Brewerton. Clinical Handbook of Eating Disorders. An Integrated Approach PDF

Clinical Handbook of Eating Disorders
Medical Psychiatry
Series Editor Emeritus
William A.Frosch, M.D.
Weill Medical College of Cornell University
New York, New York, U.S.A.
Advisory Board
Jonathan E.Alpert, M.D., Ph.D.
Massachusetts General Hospital and
Harvard University School of Medicine
Boston, Massachusetts, U.S.A.
Bennett Leventhal, M.D.

University of Chicago School of Medicine
Chicago, Illinois, U.S.A.
Siegfried Kasper, M.D.

University Hospitalfor Psychiatry
and University of Vienna
Vienna, Austria
Mark H.Rapaport, M.D.

Cedars-Sinai Medical Center
Los Angeles, California, U.S.A
1. Handbook of Depression and Anxiety: A Biological Approach, edited by Johan

A.den Boer and J.M.Ad Sitsen
2. Anticonvulsants in Mood Disorders, edited by Russell T.Joffe and Joseph
R.Calabrese
3. Serotonin in Antipsychotic Treatment: Mechanisms and Clinical Practice,
edited by John M.Kane, H.-J.Möller, and Frans Awouters
4. Handbook of Functional Gastrointestinal Disorders, edited by Kevin W.Olden
iv
24. Autism Spectrum Disorders, edited by Eric Hollander

25. Handbook of Chronic Depression: Diagnosis and Therapeutic Management,
edited by Jonathan E.Alpert and Maurizio Fava
26. Clinlcal Handbook of Eating Disorders: An Integrated Approach, edited by
Timothy D.Brewerton
27. Dual Diagnosis and Psychiatric Treatment: Substance Abuse and Comorbid
Disorders, Second Edition, edited by Henry R.Kranzler and Joyce Tinsley
28. Atypical Antipsychotics: From Bench to Bedside, edited by John G. Csernansky
and John Lauriello
29. Social Anxiety Disorder, edited by Borwin Bandelow and Dan J. Stein
ADDITIONAL VOLUMES IN PREPARATION
iii
5. Clinical Management of Anxiety, edited by Johan A.den Boer

6. Obsessive-Compulsive Disorders: Diagnosis • Etiology • Treatment, edited by
Eric Hollander and Dan J.Stein
7. Bipolar Disorder: Biological Models and Their Clinical Application, edited by
L Trevor Young and Russell T.Joffe
8. Dual Diagnosis and Treatment: Substance Abuse and Comorbid Medical and
Psychiatric Disorders, edited by Henry R.Kranzlerand Bruce J.Rounsaville
9. Geriatric Psychopharmacology, edited by J.Craig Nelson
10. Panic Disorder and Its Treatment, edited by Jerrold F.Rosenbaum and Mark
H.Pollack
11. Comorbidity in Affective Disorders, edited by Mauricio Tohen
12. Practical Management of the Side Effects of Psychotropic Drugs, edited by
Richard Balon
13. Psychiatric Treatment of the Medically III, edited by Robert G. Robinson and
William R.Yates
14. Medical Management of the Violent Patient: Clinical Assessment and
Therapy, edited by Kenneth Tardiff
15. Bipolar Disorders: Basic Mechanisms and Therapeutic Implications, edited by
Jair C.Soares and Samuel Gershon
16. Schizophrenia: A New Guide for Clinicians, edited by John G. Csernansky
17. Polypharmacy in Psychiatry, edited by S.Nassir Ghaemi
18. Pharmacotherapy for Child and Adolescent Psychiatric Disorders: Second
Edition, Revised and Expanded, David R.Rosenberg, Pablo A.Davanzo, and Samuel
Gershon
19. Brain Imaging in Affective Disorders, edited by Jair C.Soares
20. Handbook of Medical Psychiatry, edited by Jair C.Soares and Samuel Gershon
21. Handbook of Depression and Anxiety: Second Edition, Revised and
Expanded, edited by Siegfried Kasper, Johan A.den Boer, and J.M.Ad Sitsen
22. Aggression: Psychiatric Assessment and Treatment, edited by Emil F.Coccaro
23. Depression in Later Life: A Multidisciplinary Psychiatric Approach, edited by
James M.Ellison and Sumer Verma
Clinical Handbook of
Eating Disorders
An Integrated Approach
edited by
Timothy D.Brewerton
Medical University of South Carolina Charleston, South
Carolina, U.S.A.
MARCEL DEKKER, INC.

NEW YORK • BASEL
This edition published in the Taylor & Francis e-Library, 2005.
“To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of
thousands of eBooks please go to www.eBookstore.tandf.co.uk.”
Although great care has been taken to provide accurate and current information,
neither the author(s) nor the publisher, nor anyone else associated with this publication, shall
be liable for any loss, damage, or liability directly or indirectly caused or
alleged to be caused by this book. The material contained herein is not intended to
provide specific advice or recommendations for any specific situation.
Trademark notice: Product or corporate names may be trademarks or registered
trademarks and are used only for identification and explanation without intent to
infringe.
Library of Congress Cataloging-in-Publication Data
A catalog record for this book is available from the Library of Congress.
ISBN: 0-8247-4867-0
Headquarters
Marcel Dekker, Inc., 270 Madison Avenue, New York, NY 10016, U.S.A.
tel: 212–696–9000; fax: 212–685–4540
Distribution and Customer Service
Marcel Dekker, Inc., Cimarron Road, Monticello, New York 12701, U.S.A.
tel: 800–228–1160; fax: 845–796–1772
Eastern Hemisphere Distribution
Marcel Dekker AG, Hutgasse 4, Postfach 812, CH-4001 Basel, Switzerland
tel: 41–61–260–6300; fax: 41–61–260–6333
World Wide Web
http://www.dekker.com
The publisher offers discounts on this book when ordered in bulk quantities. For more
information, write to Special Sales/Professional Marketing at the headquarters
address above.
Copyright © 2004 by Marcel Dekker, Inc. All Rights Reserved.
Neither this book nor any part may be reproduced or transmitted in any form or by
any means, electronic or mechanical, including photocopying, microfilming, and
recording, or by any information storage and retrieval system, without permission
in writing from the publisher.
ISBN 0-203-02176-2 Master e-book ISBN
ISBN: 0-8247-4867-0 (Print Edition)

This book is inspired by and dedicated to all my most
important teachers—my patients.
Foreword
It is surprising that the literature on eating disorders has taken a very long time
to blossom. The original descriptions of anorexia nervosa by Gull and Lasègue date
back to 1874 and 1873 respectively, but there were long delays before its
recognition as a diagnostic entity. In its early editions, DSM merely labelled
anorexia nervosa as a gastrointestinal reaction or a feeding disturbance. It was
only in 1980, in DSM-III, that the “eating disorders” category was created and
the diagnostic criteria originally formulated by clinicians in the early 1970s
given a seal of approval. Bulimia nervosa was first described in 1979 and correctly
accorded this term in DSM-IIIR in 1987. Proposals for the recognition of a new
disorder, binge eating disorder, were put forward in the early 1990s, and
accorded grudging recognition in an appendix of DSM-IV in 1994.
The expansion of our field of study owes a great deal to the assiduous
research carried out by clinicians and scientists during the past 40 years, but
there is an additional reason not sufficiently recognised. This is the fact that
anorexia nervosa is an illness which has undergone profound changes over time.
It is not clear when this transformation began but the key socio-cultural changes
made an impact during the 1960s or even earlier. The actual increase in the
incidence of anorexia nervosa may have been fairly modest, but there has been
an outburst of eating disorders as a whole. Since the first description of the
syndrome of bulimia nervosa in 1979, its incidence has reached twice that of
anorexia nervosa. There has been an even larger number of patients with less
specified eating disorders. This proliferation explains in part why our subject
matter has become more varied and complex, and has given rise to a growing
clinical and scientific literature.
An important causative factor for the transformation of anorexia nervosa into
related disorders, is the cult of thinness which has pervaded modern westernised
and industrial societies. The components of the cult of thinness and its impact
have aroused much interest. In 1988, the social historian Joan Jacobs Brumberg
expressed deep concern about the social pressures promoting anorexia
nervosa and other eating disorders. She listed among the culprits magazines
FOREWORD ix
disseminating weight reducing diets, the fashion industry catering for the
slimmer figure, and television attributing sexual allure and professional success
to the possession of a svelte figure. These concerns seem at face value to be
entirely justified. She went further, however, in describing how Americans are
competitive even about a disease and regretted what she called “an army of
health professionals and a deluge of publications and conferences since the
1970s.” Presumably these observations were made tongue in cheek and have
fortunately not deterred Professor Brewerton and his team of contributors from
producing this new handbook much needed to do justice to the growth of our
subject.
The handbook aims at providing an integrated approach. The first feature of
integration that immediately strikes the reader is the full cover given to the well-
established disorders—not only anorexia and bulimia nervosa, but binge eating
disorder (BED) as well, and sometimes obesity is thrown in for good measure.
This has led the authors sometimes having to concede that there are only slender
research findings available, especially in BED. For example, there are no
randomised controlled trials for the evaluation of the psychological treatments
and very few on pharmacological agents in this disorder. Yet this negative
statement is useful because it demonstrates the gaps that remain to be filled.
Similarly, the chapter on Risk Factors in the various eating disorders frankly
admits that the number of studies into BED are very small, owing to the recency
of defining its research criteria.
Also important in the editor’s mission of integration is the convergence of the
main avenues of investigation—the psychosocial, the biomedical and the study of
personality. This handbook adheres therefore to the multidimensional
perspective first put forward in 1982 by Garfinkel and Garner. The significance
of this commitment should not be under-estimated. For example, Campbell in
1995 boldly asserted that the term “multifactorial,” although popular in
psychiatry, has little explanatory power and may simply serve as a cloak to
conceal our ignorance. His plea was to insist on research seeking a unitary and
necessary causal element as has been possible in many physical illnesses.
Unfortunately this worthwhile goal remains just as elusive now as it was in
1995. It remains important practically to examine risk factors in the various
fields of endeavour—sociocultural, psychological, and biomedical—because
this approach ensures that no field of study is neglected. What is particularly
striking in this handbook is the breadth and erudition of the chapters devoted to
the biomedical approach which range from genetics and molecular biology to
neuroendocrine, neuropeptide, and neurotransmitter disturbances, and finally
opens the hopeful prospects of brain imaging in patients with eating disorders.
A striking feature of the handbook is a strict objectivity in assessing the
relevant literature. In the overview of risk factors for each of the eating
x FOREWORD
disorders, the authors rightly confine their appraisal to combinations of

symptoms forming syndromes and not the symptoms themselves. Their method
of establishing causality is a demanding one. They require that the factors
suspected of causing an eating disorder are indeed precursors to the illness
itself. Consequently they largely reject the findings of mere correlations in
cross-sectional studies, and express a strong preference for prospective
longitudinal studies. These are rather rare, and thus they are obliged to use
“retrospective correlates,” meaning factors shown to predate the onset of the
disorder according to the subjects’ self-report. This stark self-discipline leads to
a drastic pruning of our previously cherished lists of causal factors, such as
family interaction and perfectionism. We are left with the following shortened
list: gender, ethnicity, higher BMI, childhood eating problems, sexual abuse,
psychiatric morbidity, low self-esteem, weight concerns and dieting.
It cannot be said that the handbook is unduly rigid in insisting on objectivity.
Fluidity has been maintained and one example will suffice. In recent years the
concept of co-morbidity has entered the domain of psychiatric disorders, and
particularly eating disorders. The term “co-morbidity” features in several
chapters ranging from the co-existence of different psychiatric symptoms, to the
secondary complications from nutritional disturbances in anorexia and bulimia
nervosa. In his chapter, Dr. Brewerton expresses the view that co-morbidity is
the rule rather than the exception, and he ingeniously finds aetiological links in
the way trauma and victimisation are apt to result in a post-traumatic stress
disorder, co-morbid with an eating disorder in some patients. In case this
fluidity is thought to be excessive, there is a suitable corrective in the chapter
dealing specifically with co-morbidity, where clear definitions and illustrations
of the multiple use of this term are provided. There is also a timely warning to
avoid co-morbidity that arises as an artefact from overlapping diagnostic
criteria.
In the field of treatment, flexibility of approach is usually an advantage, and
this is certainly illustrated in the chapter on family evaluation and therapy. The
role of the family in the genesis of eating disorders remains uncertain and
controversial. In this chapter the arguments are presented fairly and from a
historical prospective. Justice is done to the different schools of family therapy,
including the Maudsley model. Due credit is given to the systematic evaluation
of family therapy through randomised control trials undertaken by the Maudsley
school. The contributors believe the value of these trials, at least in anorexia
nervosa. On the other hand they also recognise that there may be a place for family
therapy in bulimia nervosa and binge eating disorder where such studies are
sparse or non-existent. Here is another example of integration, this time
between research-based and clinical approaches.
FOREWORD xi
Finally, a neat example of integration is provided in the very first chapter,

“Diagnostic Issues in Eating Disorders.” Leaning on the work of Sing Lee in
Hong Kong, the contributors propose a new approach to the diagnostic criteria
for anorexia nervosa, dividing them between cultureindependent criteria (e.g.,
deliberate food avoidance), and criteria which are culture-specific. The latter
include a fear of fatness in western patients, and a different reason for food
avoidance in Chinese patients, namely a desire for an enhanced spiritual or
religious experience.
This Clinical Handbook can be warmly recommended to clinicians and clinical
scientists, whether their main interest is in understanding the nature of eating
disorders or treating the patients suffering from them. Dr. Brewerton is to be
congratulated on designing a handbook with a coherent structure and
assembling a team of authors who display both erudition and well-balanced
judgement.
Gerald Russell MD, FRCP, FRCPEd., FRCPsych. (Hon.)
Emeritus Professor of Psychiatry
Consultant Psychiatrist
The Priory Hospital Hayes Grove
Bromley, Kent, England
REFERENCES
Brumberg JJ. Fasting Girls: The Emergence of Anorexia Nervosa as a Modern Disease.
Cambridge, Mass: Harvard University Press, 1988.
Campbell PG. What would a causal explanation of the eating disorders look like? In:
Szmukler G, Dare C, Treasure J, eds. Handbook of Eating Disorder: Theory,
Treatment and Research. Chichester: Wiley, 1995:49–64.
Garfinkel PE, Garner DM. Anorexia Nervosa: A Multidimensional Perspective. New
York: Brunner Mazel, 1982.
Lee S, Ho TP, Hsu LKG. Fat-phobic and non-fat phobic anorexia nervosa: a comparative
study of 70 Chinese patients in Hong Kong. Psychological Medicine 1993; 23:
999–1017.
Preface
Eating is a basic drive that is taken for granted by most people. However,
disorders of human eating behavior, including eating and feeding disorders, have
been of increasing interest and importance to clinicians over the last several
decades. This has been in part due to their increasing prevalence and the realization
that they are associated with marked degrees of an array of medical and
psychiatric comorbidities, as well as significant mortality. Parallel to other areas
in medicine, psychiatry, and psychology, progress has been swift (although
never fast enough), and this book attempts to present a concise, integrated and
up-to-date overview of the scientific advancement to date in this knowledge
base. This perspective is designed to be of particular interest and relevance to
the clinical practitioner, but it also has potential appeal for students, scholars
and researchers alike. Areas covered in this book include diagnosis and
assessment, developmental perspectives, epidemiology, course of illness,
etiology, specific and nonspecific risk factors, medical and psychiatric
comorbidity, and various treatment approaches and perspectives, including
speculations about future directions in the field. These topics are organized into
4 basic sections: (I) Diagnosis, Epidemiology, and Course (Chapters 1–5); (II)
Risk Factors, Etiology, and Comorbidity (Chapters 6–10); (III) Psychobiology
(Chapters11–14); and (IV) Treatment (Chapters 15–23).
In Chapter 1, Dr. Blake Woodside and Richelle Twose provide a modern
overview of the diagnosis and assessment of the eating disorders, not only from
the perspective of an evolving DSM, which has had its limitations, but also from
the perspective of selective populations, such as men, the very young, and the
very old, who have eating disorders. Furthermore, the authors propose and
explain the rationale for a new classification scheme for anorexia nervosa (AN)
that is “culture-independent.” In this chapter the authors underscore the fact
that accurate diagnosis is the foundation of effective treatment interventions. In
an Appendix to this chapter a very helpful outline of a sample clinical diagnostic
interview that is suitable for distinguishing between AN, bulimia nervosa (BN)
and binge eating disorder (BED) is provided.
PREFACE xiii
In Chapter 2, Drs. Jacqueline Carter, Traci McFarlane, and Marion Olmsted

extend the discussion about diagnosis into the area of assessment and review the
latest information on the most common and reliable psychometric instruments
used in the assessment and screening of eating disorder patients. Structured
interviews, self-report instruments, and motivation to change assessment tools
used in the evaluation of patients with eating disorders are covered. In addition,
structured interviews and self-report instruments used in the measurement of
comorbid psychopathology are reviewed, as well as some family function
assessment tools.
In Chapter 3, Dr. Dasha Nichols examines feeding disorders in infancy and
early childhood, such as rumination, pica, post-traumatic feeding disorder,
selective or picky eating, and general food refusal, which are increasingly
recognized clinically in pediatric populations. The complex interplay between
genetic factors and psycho-social factors, such as problematic parent-child
interactions, are discussed. Furthermore, the potential links between feeding
disorders of infancy and early childhood and eating disorders that occur later in
life are discussed.
In Chapter 4, Drs. Maria Råstam, Christopher Gillberg, Daphne van Hoeken
and Hans Hoek review studies to date on the epidemiology of disordered eating
and the formal eating disorders from a developmental perspective. Although
concentrated in the second and third decades of life, disordered eating and
eating disorders occur across the lifespan. In addition to age factors, gender and
cultural differences are also highlighted, which shed further light on the nature-
nurture interaction.
In Chapter 5, Drs. Pamela Keel and David Herzog provide an appraisal on
the long-term outcome, course and mortality of these potentially lifethreatening
illnesses. Although much remains to be learned in this area, the known
prognostic indicators for recovery and/or relapse are reviewed, as are the
results on the impact of treatment on outcome.
In Chapter 6, Dr. Corinna Jacobi, Lisette Morris, and Dr. Martina de Zwaan
provide a highly integrated and critically comprehensive overview of known risk
factors, both specific and nonspecific, for the development of AN, BN and BED.
Of particular interest to the field is the issue of causation, which serves to guide
treatment. Historically, there has tended to be a polarization of thinking about
etiology into classical nature vs. nurture arenas, but each pole falls short when it
attempts to stand alone. Available data clearly point to an interplay of factors
that cut across these classical divisions and point to an interactive model across
the life span. There does not appear to be only one cause of eating disorders,
any more so than there is only one cause of mood or anxiety disorders. The
“action is in the interaction,” as the anonymous saying goes.
xiv PREFACE
In Chapter 7, using a wealth of twin and family studies, Dr. Cindy Bulik
explores an exploding area of research and its profound clinical ramifications in
a chapter on the role of genetics in the eating disorders. These data underscore
the importance of biological factors in the development of all types of eating
disorders. However, because genetic factors do not explain 100% of the
variance, they also point out the limitations of genetics as an explanatory
paradigm and lend credence to the impact of important psychosocial forces in
shaping these disorders, perhaps by triggering and interacting with underlying
genetic and biological mechanisms.
In Chapter 8, Dr. Lisa Lilenfeld provides a comprehensive overview of axis I
psychiatric comorbidity that can be associated with AN, BN and BED.
Relationships with mood, anxiety, substance use, impulse control, psychotic,
dissociative, somatoform, attention-deficit, and disruptive behavior disorders
are all described, although the emphasis is on the most common co-occurring
conditions—mood, anxiety, and substance use disorders.
In Chapter 9, Drs. Howard Steiger and Kenneth Bruce look at the
increasingly recognized role of premorbid personality traits in the development
and perpetuation of the eating disorders. In addition, comorbid axis II
personality disorders that are typically associated with the eating disorders are
examined in depth. The implications of these data for the etiology and
treatment of eating disorders is a major focus of discussion in this chapter.
In Chapter 10, Drs. Pauline Powers and Yvonne Bannon examine in detail
the spectrum of medical (axis III) comorbidities associated with the various
eating disorders. Eating disorders are known to adversely affect literally every
organ system in the body, and it behooves the clinician to be informed of these
potentially irreversible and life-threatening problems. The importance of
working with a team of specialists, including a primary care physician who
understands and can treat these associated medical conditions is emphasized.
In Chapter 11, Dr. Howard Steiger and myself discuss neurotransmitter
dysregulation in the eating disorders. Primary focus is on the monoamines
(serotonin, norepinephrine, and dopamine), with special reference to
serotonin, which has been repeatedly found to be intimately involved in the
pathophysiology as well as the neuropsychopharmacology of the eating
disorders. A table that links various phenomenological aspects associated with
the eating disorders to each of these three neurotransmitter systems provides a
helpful perspective. Just as there is no one cause of the eating disorders, there is
also not just one neurotransmitter or neurochemical system involved in these
complex conditions.
Likewise, in Chapter 12, Drs. Ursula Bailer and Walter Kaye review the
fundamental areas of neuroendocrine and neuropeptide dysregulation in the
PREFACE xv
eating disorders, which underlie many of their clinical manifestations, including

core symptoms as well as their medical and psychiatric comorbidities.
In Chapter 13, Drs. Janet Treasure and Rudolph Uher review data from the
emerging area of brain imaging of the eating disorders, which provide further
insights into the causes and underlying brain mechanisms involved in these
tenacious conditions. Although functional brain imaging of individual eating
disorder cases has not reached the status of clinical application as of yet, the
potential for this approach in the future and the immediate clinical implications
of these group comparison studies will be expounded.
In Chapter 14, Dr. Dorothy Grice gives an overview of the role of molecular
biology in elucidating the role of genes in the etiology of the eating disorders.
Dr. Grice takes us beyond the traditional genetic methodology of twin, family,
and adoption studies into the modern realm of genetic association analyses used
to identify genes, and she provides an up-to-date appraisal of this rapidly
changing field as applied to the eating disorders.
In Chapter 15, Dr. Wayne Bowers, Dr. Arnold Andersen and Kaye Evans
explain the indications for and role of inpatient and partial hospitalization in the
treatment of patients with eating disorders. In certain clinical situations, such as
low-weight AN, or BN with severe medical and/or psychiatric comorbidity, a
structured setting is often required in order to provide the level of intensity of
care required to significantly impact these conditions. This chapter provides the
theoretical basis and specific guidelines for this often life-saving level of
treatment.
In Chapter 16, nutritionists Jillian Croll and Dr. Dianne NeumarkSztainer
expound on the basics of the practice of nutrition counseling for AN, BN, and
BED. This chapter underscores the need for a multidisciplinary approach toward
all of the eating disorders in which a knowledgeable nutritionist is involved. The
authors guide the reader through the components of a comprehensive nutritional
assessment and provide the fundamentals of nutritional rehabilitation and
weight restoration/maintenance as well as their scientific basis. The authors also
provide a daily food and feelings journal, as well as sample meal plans for
various daily caloric intake levels.
In Chapter 17, Drs. Stephen Wonderlich, Jim Mitchell, Lorraine
SwanKremier, Scott Crow, and Carol Peterson provide an overview of
cognitivebehavioral therapy (CBT) in the treatment of the eating disorders,
particularly in the outpatient setting. CBT is clearly the most extensively
studied and best empirically supported psychotherapeutic approach to the eating
disorders, especially for bulimic disorders, and its fundamental principles and
components are expounded. The authors also provide a food and liquid
monitoring form that can be used by patients to identify contextual cues
(thoughts, feelings, circumstances) associated with eating disordered behaviors.
xvi PREFACE
In Chapter 18, Dr. Deb Marcontell Michel and Susan Willard, M.S.W.
provide an overview of family evaluation and therapy for AN, BN and BED.
Basic concepts and principles of family therapy are described, as is the history of
its application in the management of eating disorders. The family characteristics
of individuals with eating disorders are appraised, and the components of a
family evaluation are illustrated. In addition, common approaches and issues
seen in the treatment of eating disorders are discussed.
In Chapter 19, Drs. Joy Jacobs, J.D., Robinson Welch, and Denise Wilfley
instruct us on the principles of Interpersonal Psychotherapy (IPT) for AN, BN
and BED. Originally developed for the treatment of depression, IPT has been
successfully adapted to the treatment of BN and BED and has been empirically
validated for these conditions. The authors review the theoretical foundations of
IPT and discuss the role of interpersonal functioning in the eating disorders. The
basic concepts of IPT in the treatment of eating disorders are explained, as well
as its treatment structure, therapeutic stance and typical phases of evolution.
Finally, the authors review data to date on outcome studies of the use of IPT in
the treatment of eating disorders.
In Chapter 20, Drs. Marsha Marcus and Michele Levine bring us up to speed
on the powerful principles and techniques of Dialectical Behavior Therapy
(DBT) as applied to the treatment of eating disorders. The authors review the
philosophy and assumptions underlying DBT and discuss the modes of
treatment used in this approach. The authors also provide a DBT diary card
adapted for use in patients with eating disorders.
In Chapter 21, Drs. Joanna Steinglass and Timothy Walsh update the reader
on the role of psychopharmacology in the treatment of AN, BN and BED. Unlike
some psychiatric disorders, psychopharmacologic treatment alone is neither
sufficient nor optimal for the treatment of any of the eating disorders, so it
should be ideally conceived as an adjunct to appropriate psychotherapeutic
approaches. This is especially true for AN for reasons that are elucidated. In this
chapter current and realistic guidelines and limitations for using
psychopharmacologic agents appropriate for these and related conditions are
provided.
In Chapter 22, I review the recent literature on the role of victimization in
eating disorders and related psychiatric comorbidity. A host of controlled studies
now clearly implicate victimization experiences, especially childhood sexual
abuse and subsequent post-traumatic stress disorder (PTSD) or symptoms, as
important risk factors in the development of BN, and perhaps AN, binge-purge
type, BED, and severe obesity (but not AN, restricting type). In addition,
victimization experiences are associated with a host of commonly seen
comorbid psychiatric disorders, including mood, anxiety, substance use,
dissociative, somatoform, impulse control, disruptive behavior, and personality
PREFACE xvii
disorders. The implications of these data for the understanding of the etiology
of eating disorders and associated psychiatric comorbidity, as well as for their
treatment in clinical practice, are highlighted. Specific principles or guidelines
used to approach the comorbid eating disordered patient are provided.
In Chapter 23, Dr. Joel Yager relies on his extensive experience, knowledge
and creative vision to speculate about future directions in the management of
eating disorders. Dr. Yager explores several areas, including future directions in
diagnosis, epidemiology, molecular genetic research, other biological
investigations, biological interventions, psychosocial interventions, and systems
of care. In addition, the impact of computers and information technology on
eating disorders management is explored.
I would like to earnestly thank all of these outstanding contributors. It is my
sincerest wish that the knowledge imparted to the reader will ultimately lead to
more lives saved, enhanced recovery from these formidable conditions, and less
suffering by patients and their loved ones. In addition, may it inspire continued
research and further advances in this field, without which this book would not
have been feasible.
Timothy D.Brewerton, M.D., D.F.A.P.A., F.A.E.D.
Contents
Foreword Gerald Russell viii

Preface xii
Contributors xxi
PART I: DIAGNOSIS, EPIDEMIOLOGY, AND COURSE

1 Diagnostic Issues in Eating Disorders: Historical 1
Perspectives and Thoughts for the Future
D.Blake Woodside and Richelle Twose
2 Psychometric Assessment of Eating Disorders 19
Jacqueline C.Carter, Traci L.McFarlane, and Marion
P.Olmsted
3 Feeding Disorders in Infancy and Early Childhood 47
Dasha Nicholls
4 Epidemiology of Eating Disorders and Disordered 71
Eating: A Developmental Overview
Maria Råstam, Christopher Gillberg, Daphne van Hoeken,
and Hans Wijbrand Hoek
5 Long-Term Outcome, Course of Illness, and 97
Mortality in Anorexia Nervosa, Bulimia Nervosa,
and Binge Eating Disorder
Pamela K.Keel and David B.Herzog
PART II: RISK FACTORS, ETIOLOGY, AND COMORBIDITY

6 An Overview of Risk Factors for Anorexia Nervosa, 117
Bulimia Nervosa, and Binge Eating Disorder
Corinna Jacobi, Lisette Morris, and Martina de Zwaan
xix
7 Role of Genetics in Anorexia Nervosa, Bulimia 165

Nervosa, and Binge Eating Disorder
Cynthia M.Bulik
8 Psychiatric Comorbidity Associated with Anorexia 183
Nervosa, Bulimia Nervosa, and Binge Eating Disorder
Lisa Rachelle Riso Lilenfeld
9 Personality Traits and Disorders Associated with 209
Anorexia Nervosa, Bulimia Nervosa, and Binge
Eating Disorder
Howard Steiger and Kenneth R.Bruce
10 Medical Comorbidity of Anorexia Nervosa, Bulimia 233
Pauline S.Powers and Yvonne Bannon
PART III: PSYCHOBIOLOGY

11 Neurotransmitter Dysregulation in Anorexia 257
Nervosa, Bulimia Nervosa, and Binge Eating Disorder
Timothy D.Brewerton and Howard Steiger
12 Neuroendocrine and Neuropeptide Dysregulation in 283
Eating Disorder
Ursula F.Bailer and Walter H.Kaye
13 Neuroimaging of the Eating Disorders 297
Janet Treasure and Rudolf Uher
14 Molecular Biology of Anorexia Nervosa, Bulimia 323
Nervosa, Binge Eating Disorder, and Obesity
Dorothy Grice
PART IV: TREATMENT

15 Management for Eating Disorders: Inpatient and 349
Partial Hospital Programs
Wayne A.Bowers, Arnold E.Andersen, and Kay Evans
16 Nutrition Counseling for Anorexia Nervosa, Bulimia 379
Jillian K.Croll and Dianne Neumark-Sztainer
xx
17 An Overview of Cognitive-Behavioral Approaches to 405

Eating Disorders
Stephen A.Wonderlich, James E.Mitchell, Lorraine Swan-
Kremier, Carol B.Peterson, and Scott J.Crow
18 An Overview of Family Evaluation and Therapy for 427
Eating Disorder
Deborah Marcontell Michel and Susan G.Willard
19 Interpersonal Psychotherapy for Anorexia Nervosa, 451
M.Joy Jacobs, R.Robinson Welch, and Denise E.Wilfley
20 Use of Dialectical Behavior Therapy in the Eating 475
Disorders
Marsha D.Marcus and Michele D.Levine
21 Psychopharmacology of Anorexia Nervosa, Bulimia 491
Joanna E.Steinglass and B.Timothy Walsh
22 Eating Disorders, Victimization, and Comorbidity: 511
Principles of Treatment
Timothy D.Brewerton
23 Future Directions in the Management of Eating 549
Disorders
Joel Yager
Index 571
Contributors
Arnold E.Andersen University of Iowa Hospital & Clinics, Iowa City,

Iowa, U.S.A.
Ursula F.Bailer Anorexia and Bulimia Nervosa Research Module, Western
Psychiatric Institute and Clinic, University of Pittsburgh School of Medicine,
Pittsburgh, Pennsylvania, U.S.A.
Yvonne Bannon Department of Psychiatry & Behavioral Medicine, College
of Medicine, University of South Florida, Tampa, Florida, U.S.A.
Wayne A.Bowers University of Iowa Hospital & Clinics, Iowa City, Iowa,
U.S.A.
Timothy D.Brewerton Institute of Psychiatry, Medical University of
South Carolina, Charleston, South Carolina, U.S.A.
Kenneth R.Bruce Eating Disorders Program, Douglas Hospital, and
McGill University, Montreal, Quebec, Canada
Cynthia M.Bulik Department of Psychiatry, University of North Carolina,
Chapel Hill, North Carolina, U.S.A.
Jacqueline C.Carter The Eating Disorders Program, University of
Toronto, Toronto, Ontario, Canada
Jillian K.Croll Eating Disorders Institute, St. Louis Park, Minnesota,
U.S.A.
Scott J.Crow Eating Disorders Research Program, Department of
Psychiatry, University of Minnesota, Minneapolis, Minnesota, U.S.A.
Martina de Zwaan Department of Psychosomatic Medicine and
Psychotherapy, Friedrich Alexander University Erlangen-Nuremberg,
Erlangen, Germany
Kay Evans University of Iowa Hospital & Clinics, Iowa City, Iowa, U.S.A
Christopher Gillberg Department of Child and Adolescent Psychiatry,
Göteborg University, Göteborg, Sweden, and St. George’s Hospital Medical
School, London, England
xxii CONTRIBUTORS
Dorothy Grice* Center for Neurobiology and Behavior, University of

Pennsylvania, Philadelphia, Pennsylvania, U.S.A.
David B.Herzog Department of Psychiatry, Massachusetts General
Hospital, Boston, Massachusetts, U.S.A.
Hans Wijbrand Hoek Department of Research, Parnassia, The Hague,
The Netherlands, and Mailman School of Public Health, Columbia
University, New York, New York, U.S.A.
Corinna Jacobi Department of Clinical Psychology and Psychotherapy II,
FB I—Psychology, University of Trier, Trier, Germany
M.Joy Jacobs Center for Eating and Weight Disorders, San Diego State
University/University of California San Diego Joint Doctoral Program in
Clinical Psychology, San Diego, California, U.S.A.
Walter H.Kaye Western Psychiatric Institute and Clinic, University of
Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, U.S.A.
Pamela K.Keel Department of Psychology, Harvard University,
Cambridge, Massachusetts, U.S.A.
Michele D.Levine Department of Psychiatry, Western Psychiatric Institute
and Clinic, and Behavioral Medicine and Eating Disorders Program,
University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania,
U.S.A.
Lisa Rachelle Riso Lilenfeld Department of Psychology, Georgia State
University, Atlanta, Georgia, U.S.A.
Marsha D.Marcus Department of Psychiatry, Western Psychiatric
Institute and Clinic, and Behavioral Medicine and Eating Disorders Program,
University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania,
U.S.A.
Traci L.McFarlane The Eating Disorders Program, University of Toronto,
Toronto, Ontario, Canada
Deborah Marcontell Michel Tulane University School of Medicine, New
Orleans, Louisiana, U.S.A.
James E.Mitchell Department of Neuroscience, University of North
Dakota School of Medicine and Health Sciences, and the Neuropsychiatric
Research Institute, Fargo, North Dakota, U.S.A.
*Current affiliation: Child and Adolescent Psychiatry, Department of Psychiatry,

University of Medicine and Dentistry of New Jersey, New Jersey Medical School,
Newark, New Jersey, U.S.A.
CONTRIBUTORS xxiii
Lisette Morris Department of Clinical Psychology and Psychotherapy,

Georg-Elias-Miiller Institute of Psychology, University of Göttingen, Göt-
tingen, Germany
Dianne Neumark-Sztainer Division of Epidemiology, School of Public
Health, University of Minnesota, Minneapolis, Minnesota, U.S.A.
Dasha Nicholls Head of Feeding and Eating Disorders Service,
Department of Child and Adolescent Mental Health, Great Ormond Street
Hospital, London, England
Marion P.Olmsted The Eating Disorders Program, University of Toronto,
Carol B.Peterson Eating Disorders Research Program, Department of
Psychiatry, University of Minnesota, Minneapolis, Minnesota, U.S.A.
Pauline S.Powers Department of Psychiatry & Behavioral Medicine,
College of Medicine, University of South Florida, Tampa, Florida, U.S.A.
Maria Råstam Department of Child and Adolescent Psychiatry, Göteborg
University, Göteborg, Sweden
Howard Steiger Eating Disorders Program, Douglas Hospital, Montreal,
Quebec, Canada
Joanna E.Steinglass New York State Psychiatric Institute, Columbia
University, New York, New York, U.S.A.
Lorraine Swan-Kremier Department of Neuroscience, University of
North Dakota School of Medicine and Health Sciences, and the
Neuropsychiatric Research Institute, Fargo, North Dakota, U.S.A.
Janet Treasure Department Academic Psychiatry, Thomas Guy House,
London, England
Richelle Twose University of Toronto, Toronto, Ontario, Canada
Rudolf Uher Eating Disorder Unit, Institute of Psychiatry, London,
England
Daphne van Hoeken Department of Research, Parnassia, The Hague, The
Netherlands
B.Timothy Walsh Department of Clinical Psychopharmacology, New York
State Psychiatric Institute, Columbia University, New York, New York,
U.S.A.
R.Robinson Welch Department of Psychiatry, Washington University
School of Medicine, St. Louis, Missouri, U.S.A.
Denise E.Wilfley Department of Psychology, Washington University
School of Medicine, St. Louis, Missouri, U.S.A.
xxiv CONTRIBUTORS
Susan G.Willard Department of Psychiatry & Neurology, Pediatrics,

Tulane University School of Medicine, New Orleans, Louisiana, U.S.A.
Stephen A.Wonderlich Department of Neuroscience, University of
North Dakota School of Medicine and Health Sciences, and the
Neuropsychiatric Research Institute, Fargo, North Dakota, U.S.A.
D.Blake Woodside Toronto General Hospital and University of Toronto,
Joel Yager Department of Psychiatry, University of New Mexico School of
Medicine, Albuquerque, New Mexico, U.S.A.
1
Diagnostic Issues in Eating Disorders
Historical Perspectives and Thoughts for the
Future
D.Blake Woodside
Toronto General Hospital and University of Toronto
Richelle Twose
University of Toronto
Diagnostic issues have always been important in the field of eating disorders,
whether debating the role of gender in these conditions or examining possible
etiologic factors. With recent advances in areas as diverse as culture and
genetics, reexamining thinking about diagnosis in anorexia nervosa, bulimia
nervosa, and binge eating disorder becomes increasingly important.
This chapter will review some of the historical issues in the diagnosis of
eating disorders and examine diagnostic schemas for special populations. A final
section will address some thoughts about possible revisions to the diagnostic
criteria in the light of recent developments in the field. A sample clinical
diagnostic interview schedule is attached as an appendix.
DIAGNOSTIC CRITERIA
Anorexia Nervosa
The essential diagnostic criteria for anorexia nervosa has undergone more
gradual and subtle shifts over time relative to bulimia nervosa. The Diag nostic
and Statistical Manual of Psychiatric Disorders (DSM) criteria (Table 1) are largely
reflective of Russell’s 1970 description of three essential factors: (a) purposeful
loss of body weight, (b) amenorrhea in females, and (c) psychopathology
expressed as an intense fear of becoming fat (1,2). Many of the successive
alterations are merely semantic in nature.
2 WOODSIDE AND TWOSE
TABLE 1 DSM-III Diagnostic Criteria for Anorexia Nervosa
Source: American Psychiatric Association, 1980.
The weight loss threshold for diagnosis of anorexia nervosa has become less
strict and severe with each revision of the DSM. In the DSM-III, a loss of 25%
of original body weight (or loss of original body weight plus projected weight
gain if an adolescent) is required (3). In subsequent versions, this threshold falls
to 15% and becomes only a “rough” guideline. In fact, the relevant text in the
DSM-IIIR (Table 2) states that weighing 85% of that expected is merely an
“arbitrary but useful guide” (4).
In contrast, the evolution of the requirement for amenorrhea has become
increasingly stringent over time. Contrary to Russell’s description (1970), the
DSM-III does not list amenorrhea as part of the diagnostic criteria but merely
mentions it in the relevant text on the section (3). In later versions, this aspect
of the disorder becomes mandatory for diagnosis, and the nature of amenorrhea
itself is further specified (4,5). However, the presence of this criterion continues
to spark controversy among experts in the field, since in approximately 15% of
cases menstruation ceases prior to weight loss, and in some individuals
amenorrhea persists for a period of time following restorative weight gain (1).
One group (6) compared women who meet all DSMIIIR criteria for anorexia
nervosa to women who met all criteria except amenorrhea and found no
significant differences on a wide variety of clinical and psychometric variables.
The final issue that is relevant is the increasingly common use of oral
contraceptives in this population, either for contraception or as a treatment for
osteoporosis. Most experts suggest that in the presence of oral contraceptive use
this criterion should be waived.
While the usefulness of this feature appears to be less and less important, it may
be some time before it is abandoned.
Successive versions of the DSM have increasingly emphasized the
psychopathological aspects of the illness. For example, criterion C in the
DSM-IV, (Table 3) was expanded to include a description of three different
DIAGNOSTIC ISSUES 3
TABLE 2 DSM-IHR Diagnostic Criteria for 307.10 Anorexia Nervosa
examples of how one’s body weight or shape is cognitively experienced in a

pathological way. Here the mental “disorder” is said to likely involve body
image distortion, its disproportionate influence on self-esteem, and the failure
to recognize the dangers of a low body weight (5). Denial of the illness is also
included in DSM-IV for the first time. DSM-III states only in the surrounding
text on the disorder that individuals with anorexia nervosa “steadfastly deny the
illness and are…resistant to therapy” (3). The DSM-IV, however, notes “denial
of the seriousness of the current low body weight” as an essential diagnostic
criterion (5).
Another recent change has been the shift away from percent average weight
to body mass index (BMI), which is weight in kilograms divided by height in
meters squared. DSM-IV suggests a BMI of 17.5 as an appropriate standard for
the diagnosis of anorexia nervosa. This change has a number of benefits both in
terms of diagnostic homogeneity across different countries and with the
adolescent age group. However, it does not take into account ethnocultural
differences and these require further study.
The increasing importance of recognizing both the physical and psychological
effects of starvation has been recognized in DSM-IV, where sections have been
added identifying common associated physical and psy chological symptoms.
This allows for increased sophistication in the diagnosis of comorbid axis I
disorders, some of which may be confused with starvation effects. This allows
for distinguishing between anorexia nervosa and depression, obsessive-
compulsive disorder, and specific phobias.
There have been several attempts to deal with the diagnosis of comorbid
bulimia nervosa and/or purging behaviors. In DSM-IIIR, if both conditions
were present, both diagnoses were made. However, DSM-IV introduced a
TABLE 3 DSM-IV Diagnostic Criteria for 307.1 Anorexia Nervosa
Source: America Psychiatric Association, 1994.
change in the way this was handled, allowing a diagnosis of anorexia nervosa to
“trump” a diagnosis of bulimia nervosa. In DSM-IV, anorexia is subclassified into
the restricting and binge-purge subtypes. This distinction remains somewhat
controversial, particularly in the light of some of the recent genetic findings in
anorexia and bulimia, which will be reviewed below. It remains to be seen as to
whether this is the optimal way in which to proceed with this area of
comorbidity.
Bulimia Nervosa
Bulimia was not recognized as a mental disorder in the DSM until the
publication of the third edition in 1980 (Table 4) (3). This followed its
identifi cation by Russell during the previous year (7). The original diagnostic
criteria in DSM-III were overinclusive and led to some confusion among
clinicians and researchers. The publication of the DSM-IIIR and, later, the
DSM-IV addressed many of the original version’s shortcomings (4,5).
Bulimia, or “ox-like eating,” the original term for the illness itself, paints an
incomplete picture of the syndrome it is meant to characterize. It does not
depict the patient’s characteristic psychopathology regarding her intense fear
of fatness and obsession with body shape and weight, nor the presence of
compensatory behaviors that are so important in the illness. The renaming of
DIAGNOSTIC ISSUES 5
TABLE 4 DSM-III Diagnostic Criteria for Bulimia
the condition as bulimia nervosa in DSM-IV helped to clarify that the illness
included important psychopathological features as well as abnormal eating.
The DSM-III criteria for bulimia is composed of both “monothetic and
polythetic criteria,” i.e., both essential and optional symptoms, respectively,
which complicates diagnosis of this disorder (8). The subsequent elimination of
the optional indicators in the DSM-IIIR (Table 5) functioned to create a more
homogeneous description (and thus a homogeneous group to research/study)
and greatly simplified the task of diagnosing this disorder.
The original description of binge eating in DSM-III, “the consumption of a
large amount of food in a discrete period of time” (3), was insufficient to
characterize this phenomenon. DSM-IIIR added the requirement for lack
of control, differentiating bingeing qualitatively from normal eating (4).
Provisions governing the time restrictions defining a binge have also evolved
over time. Originally, a binge was temporally restricted to “usually under two
hours” (3). The DSM-IIIR, on the other hand, abandoned the requirement of a
time limit, favoring a less restrictive “discrete period of time” to define the
binge event (4). The most recent criteria (5) simply provide a 2-hour period as
a useful guideline. More importantly, the criterion defines the context in which
a binge would occur. It is characterized as an amount of food that is “definitely
larger than most people would eat during a similar period of time under similar
circumstances” (5). The use of this specification distinguishes binge eating from
instances in which non-disordered individuals may overeat (i.e., at a holiday
party).
TABLE 5 DSM-IIIR Diagnostic Criteria for 307.51 Bulimia Nervosa
In addition, the frequency with which binge episodes and inappropriate

compensatory strategies must occur was not established until publication of the
DSM-IIIR (and DSM-IV; Table 6), where a minimum of two episodes per week
over 3 months is required for diagnosis (4,5). However, it is not clear that the
current frequency requirement of twice per week for 3 months is meaningful (9).
One group compared a sample of bulimia nervosa patients who differed only in
whether they binged once or twice per week on average and found no
differences across a wide spectrum of variables. It is not clear at present what
frequency of bingeing is associated with the typical psychopathology of the
condition.
Psychopathologically, the original diagnostic criteria for bulimia did not
require the presence of a preoccupation with weight and shape. The
accompanying text merely mentions that “individuals with bulimia usually
exhibit great concern about their weight” (3). This statement decreased the
emphasis on the individual’s pathological disturbance in attitudes, beliefs, and
cognitions, and their resulting distress—attributes that distinguish mental
pathology. This weakness is corrected in the DSM-IIIR, where a “persistent
overconcern with body shape and weight” is listed as an essential criterion
required for diagnosis, thereby addressing the sufferer’s abnormal attitudes
(criterion E) (4). This criterion was then strengthened in DSM-IV to include
wording related to evaluation of self (criterion D) (5).
The original formulation of compensatory behaviors in DSM-III focused on
physiology, i.e., the relief of fullness as opposed to the psychological
significance of the behavior. Induced vomiting is also mentioned last in a series
of physiological factors, such as abdominal pain and sleep, that terminate a binge
(criterion B) (3). Behaviors designed to counteract the effects of binge eating did
not become a mandatory requirement for diagnosis until the DSM-IIIR (4), where
compensatory behaviors were linked to a fear of weight gain (4) or to
modulating affect, suggesting the multiple purposes of the behaviors.
DIAGNOSTIC ISSUES 7
TABLE 6 DSM-IV Diagnostic Criteria for 307.51 Bulimia Nervosa
In the DSM-IV, the diagnosis of bulimia nervosa becomes more specific, as

the disorder is further divided into two subtypes: (a) individuals who regularly
engage in self-induced vomiting, or the misuse of laxatives, diuret ics, or
enemas following a binge (purging subtype: BN-P), versus (b) those who
compensate for binge eating primarily through the use excessive exercise or
fasting (nonpurging subtype: BN-N/P). While the purging form of the illness is
more common in treatment settings, the nonpurging form is reported to be
more common in the community, accounting for 70% of cases (10). A careful
examination for compensatory behavior in the form of dieting or exercise is
necessary to distinguish this form of the illness from binge eating disorder, as
will be reviewed below.
Binge Eating Disorder

Research criteria for a new syndrome, “binge eating disorder” (Table 7), are
included in Appendix B “Criteria Sets and Axes Provided for Further Study” of
the DSM-IV (8). However, individuals currently fulfilling criteria for this
disorder would technically still fall under the “Eating Disorder Not Otherwise
Specified” (EDNOS; Table 8) category (4). Many of the suggested diagnostic
criteria resemble symptoms of bulimia nervosa, except for the absence of the
regular use of “inappropriate compensatory behaviors” following the binge
TABLE 7 DSM-IV Research Criteria for Binge Eating Disorder
eating (4). The proposed criteria for binge eating disorder are reminiscent of the
original characterization of bulimia in the DSM-III (3,11). This is likely a
reflection of the need for further study and characterization of this illness, in
order to create a more uniform list of criteria and thus describe/encompass a
less heterogeneous set of individuals.
The characteristics of binge eating in binge eating disorder are virtually
identical to those described in bulimia nervosa, with the exception of the
frequency and duration with which they occur. The time threshold for a binge is
lengthened, defined in terms of entire days rather than discrete 2-hour
episodes. In addition, the research criteria for binge eating disorder must be
met for 6 consecutive months, rather than 3, as required for bulimia nervosa.
Individuals who qualify for diagnosis of binge eating disorder may still
occasionally engage in inappropriate compensatory behaviors, but these
behaviors are not regularly used to counteract weight gain from a binge (5).
More research is necessary to determine the frequency/upper threshold with
which these individuals may purge and still maintain the essential features of this
disorder as distinct from bulimia nervosa. Perhaps the only major difference
between bulimia nervosa and binge eating disorder is the presence or absence of
weight consciousness (8).
DIAGNOSTIC ISSUES 9
TABLE 8 DSM-IV 307.50 Eating Disorder Not Otherwise Specified
Unlike those with bulimia nervosa or anorexia nervosa-B/P, individuals

with binge eating disorder binge-eat when they are not physically hungry. That
is, the binge eating does not occur as simply a physiological response to
starvation (11).
Furthermore, the marked distress experienced by individuals with binge
eating disorder occurs not only following the binge, as in bulimia nervosa and
anorexia nervosa-B/P, but during the binge eating as well. Perhaps this is due to
the potential length episode, as one binge may last an entire day.
The weight history, current weight, diet history, and present dieting
behavior of these individuals varies. Again, additional research is required to
determine whether these factors will be important for diagnosis. Empirical
work to date suggests that these individuals have tendency to be more obese and
have a history of more marked weight fluctuations than individuals who are
obese and do not binge-eat. These factors are usually attributed to their
numerous unsuccessful dieting attempts and intermittent periods of binge eating
behavior. Furthermore, compared to obese individuals who do not binge-eat,
individuals with binge eating disorder describe pathological cognitions,
attitudes, and emotions, experienced as or characterized by “higher rates of self-
loathing, disgust about body size, depression, anxiety, somatic concern, and
interpersonal sensitivity” (5).
CLINICAL IDENTIFICATION OF EATING

DISORDERS
Appendix 1 provides a sample clinical diagnostic interview for eating disorders,
suitable for distinguishing between anorexia nervosa, bulimia nervosa, and binge
eating disorder.
COMMON DIFFERENTIAL DIAGNOSES IN EATING

DISORDERS IN ADULTS
A number of other psychiatric disturbances will periodically be confused with
anorexia nervosa or bulimia nervosa. Major depression will of course lead to
significant weight loss in a minority of patients. These individuals will deny any
interest in weight loss itself and will most usually still be eating a wide range of
foods, if in smaller quantities. When such individuals are treated they gain
weight rapidly and show none of the associated psychological comorbidity of
anorexia nervosa.
A rare patient will develop food-specific delusions that cause him or her to
avoid eating and to lose large amounts of weight. These cases can usually be
distinguished from anorexia nervosa by the presence of widespread psy chotic
features that include many other issues other than food, weight, and shape.
An occasional patient will present with symptoms of obsessive-compulsive
disorder and rituals around food. These patients may behaviorally resemble
those with anorexia nervosa and may also have extensive cognitive schemata
around their eating habits. Most typically, however, they are uninterested in
weight loss per se and do not have a wish to avoid food. They will experience
their food avoidance as largely ego-dystonic while recognizing that their
avoidance behaviors afford them a brief interval of relief from their chronic
anxiety.
Choking and swallowing phobias are relatively commonly found
masquerading as anorexia nervosa, as they are often associated with extreme
degrees of food avoidance and very significant weight loss. In such cases, a
careful history will elicit a pattern of increasing food avoidance that is not linked
to an interest in reducing calories but rather to the avoidance of a class of foods,
such as hard foods. Over time, the avoidance pattern will become more and
more generalized. These patients also deny any fear of fatness and will typically
be able to identify with some precision the exact time of onset of their food
avoidance. Such patients are readily treated with systematic desensitization,
although extreme cases may require an initial period of hospitalization to get the
process started. A variant of this problem is the individual with a choking phobia
who has lost so much weight that a duodenal feeding tube has been inserted.
DIAGNOSTIC ISSUES 11
Often the patient will not have eaten or drunk anything by mouth for a number
of years and will require extensive desensitization in order to be able to eat
normally again.
SPECIAL POPULATIONS
Males
Men have been identified as suffering from eating disorders from the beginning
of the English language literature, where Morton (12) presented both a male
and a female case of anorexia nervosa. More recent comprehensive
examinations (13–15) have demonstrated that the illnesses, when occurring in
men, are essentially indistinguishable from cases occurring in women.
Diagnostic criteria for bulimia nervosa and binge eating disorder are identical
in men and women. The requirement for amennorrhea must obviously be
waived in men; while some authors have suggested replacing this with various
substitutes (14), such as diminished sexual drive, these suggestions have not
generally been adopted. At the present time it is acceptable to diagnose
anorexia nervosa in a man if he meets all the other diagnostic criteria for the
condition.
The Very Young

There has been some controversy attached to the possibility that children
younger than 12 years might be prone to the development of eating disorders.
This controversy stems in part from a debate about the capability of younger
children to generate the cognitive set required to behave in an anorectic fashion.
Nonetheless, there is increasing evidence that these illnesses may occur in some
form in increasingly younger girls and boys.
This has led to the development of some suggested alternations to the
diagnostic criteria that take into account the lack of physical and psychological
development in a prepubertal child. One group (16) have provided a thorough
review of this topic. Diagnostically, they comment that some revision of the
usual adult criteria must be accepted. Table 9 shows Russell’s (2) criteria for
anorexia nervosa in the young, focusing on the presence of morbid fears of food
and fatness and the failure to achieve normal developmental milestones.
TABLE 9 Russell’s Criteria for Anorexia Nervosa in the Young
The Very Old

An occasional report (17) comments on the occurrence of new eating pathology
in the elderly. At the present time, no alterations in the diagnostic criteria have
been proposed for this population.
EFFECTS OF CULTURE ON THE DIAGNOSIS

OF ANOREXIA NERVOSA
An interesting recent line of thought is of interest to diagnosticians. For many
years, clinicians would periodically report atypical cases of anorexia nervosa
where weight loss was pursued, but not for the sake of thinness; these often
were described as “ascetic” anorexia nervosa, referring to the most typical
presentation involving a desire to live a simpler life.
In the mid-1990s, a psychiatrist in Hong Kong, Sing Lee, began to
publish about his experience in treating young Chinese girls with anorexia
nervosa (18–21). He noted that his patient population was roughly equally
divided into two groups. The first resembled traditional Western description of
anorexia nervosa, demonstrating a fear of fatness or drive for thinness
accompanied by food avoidance, dieting, and significant weight loss. Most of
these individuals also experienced body image distortion.
The second group, which he labeled “Chinese anorexia,” clearly avoided food
and lost weight—but had no interest in thinness per se, no fear of food or of fat,
and no body image disturbance. These individuals attributed their food
avoidance to spiritual or religious desires, such as the wish to achieve a more
pure level of existence through self-denial. Lee and his colleagues proposed that
the drive for thinness/fear of fatness and associated drive for thinness paradigm
were actually culture-bound aspects of anorexia nervosa. In his schema, the core
culture-independent aspects of the illness are deliberate food avoidance and the
achievement of a very low weight, lower than normal individuals can achieve.
From this viewpoint, the precise rationale or cognitive content would be

culture specific, and not otherwise central to the diagnosis of the condition.
No other formal research has been done on this intriguing concept. The
advantage of the concept is that it would help account for trends in incidence of
anorexia across different time periods and cultures, and would allow for an
understanding of the effect of the dissemination of Western, largely American
culture worldwide. Clinically, in our own treatment setting we see examples of
this phenomenon, both in recent immigrants from Hong Kong and in
individuals from the subcontinent of India. These latter individuals often present
with somatic complaints as the rationale for their food avoidance and weight
loss. So abdominal pain, constipation, or swallowing difficulties become the
reason for cutting out more and more foods.
The most interesting clinical observation that we could add here relates to
the effect of placing such individuals in a group therapy treatment program
primarily attended by women with classical North American anorexia nervosa.
The inevitable outcome seems to be that over the course of 2 or 3 months the
women of Chinese or Indian origin acculturate to the dominant culture of the
program, gradually abandoning their original rationale for food avoidance and
developing a fear of fatness, a drive for thinness, and body image distortion! It is
of course not certain that this represents an improvement, although it does allow
for a more homogeneous group.
The other advantage of this approach to the diagnosis of anorexia nervosa is
that it dovetails neatly with some of the information becoming available about
the genetic influences on the disorder, as will be reviewed below.
Table 10 presents a preliminary attempt to construct a set of diagnostic
criteria according to these culturally sensitive principles. In this schema, the
core requirements are deliberate food avoidance and the achievement of a very
low weight by the standards of the culture involved. The content of the
cognitive set become culture-specific associated symptoms and are no longer
core elements of the diagnosis. Such a schema would represent a marked shift in
thinking, especially for Western clinicians and researchers, but deserves some
consideration.
EFFECT OF GENETIC CAUSALITY ON DIAGNOSIS

IN EATING DISORDERS
Some information has recently been reported concerning the heritable genetic
components of both anorexia nervosa and bulimia nervosa that may be relevant
in the diagnoses of these conditions. The Price Foundation collaborative
group has recently published some preliminary results of their studies in this
area (22,23). Their findings on anorexia nervosa show significant association
TABLE 10 Proposed Culture-lndependent Diagnostic Criteria for Anorexia Nervosa
between the restricting form of the illness and high perfectionism to a site on
chromsome 1. The addition of increasing levels of bingeing or purging symptoms
reduces the magnitude of the finding, suggesting that on a genetic basis, the
restricting form of the illness has some characteristics distinct from the binge-
purge form. The report of a significant finding for a separate area of the genome
(chromosome 10) adds to the evidence that there may be distinct genetic
liabilities for the type of dietary restriction associated with anorexia nervosa and
the purging of bulimia nervosa.
Further work in this area may lead to a totally new diagnostic schema that is
related to the genetic etiology of the conditions.
CONCLUSION
The field of diagnosis in eating disorders faces many challenges in the next few
years. These relate primarily to the presentation of novel data concerning
the cultural and genetic determinants of the conditions. These data have the
potential to dramatically alter the way in which we view the nature qf the
psychiatric diagnostic labels for these conditions.
It seems likely that the next revision of DSM will include formal diagnostic
criteria for binge eating disorder. However, the relationship between this
condition and the other eating disorders remains unclear. As well, the ED-NOS
classification could benefit from some clarification, possibly along the lines of
“AN-like” and “BN-like.”
Such changes, while difficult to adapt to, will in the long run allow for a
greater understanding of the nature of the illnesses.
APPENDIX 1
Sample Clinical Interview Format
1. Introduction and overview

• patient’s view of the problem
• overall course of the development of the illness
2. Weight history
• height
• current weight
• highest-ever weight
• lowest weight since onset of illness
• typical weight ranges
• ideal weight
3. Typical daily intake
• breakfast
• lunch
• snacks
• dinner
Quantify amounts, types of food eaten and liquid consumed—is
fluid avoided or restricted?
• foods avoided
• safe foods
• diet products
4. Bingeing
• carefully define and distinguish from overeating
• onset, frequency, initiating and ameliorating factors
• specific details of a typical binge
• usual response
• nocturnal binges, odd binges, stealing
5. Purging
For all of: vomiting, laxatives, diuretics, diet pills, thyroid, exercise:
• onset, frequency, initiating and ameliorating factors
• ipecac
• surgery
• consumption of inedible items
6. Complications of starvation/bingeing and purging
• endocrine (menses)
• metabolic (K+, Hgb, ECG)
• Dermatological (skin, hair, nails)
• Gastrointestinal (bloating, early satiety, parotid swelling, dental caries,
hematemesis, hematochezia, bowel function)
• cardiac (palpitations, edema)
• psychological
sleep pattern
concentration
food preoccupation
mood
7. Medical history
• significant illnesses
• current medications
8. Past treatment history
9. Past psychiatric history
• Axis I screen
10. Family history
• includes family/genetic history
11. Personal and developmental history
• includes screen for sexual abuse
12. Relationship history
REFERENCES
1. Fairburn CG, Garner DM. Diagnostic criteria for anorexia nervosa and bulimia
nervosa: the importance of attitudes to shape and weight. In: Garner DM, Garfinkel
PE, eds. Diagnostic Issues in Anorexia Nervosa and Bulimia Nervosa. New York:
Brunner/Mazel, 1988.
2. Russell GFM. Premenarchal anorexia nervosa and its sequelae. J Psychiat Res
1985; 19:363–369.
3. American Psychiatric Association. Diagnostic and Statistical Manual of Mental
Disorders. 3rd ed. Washington, DC.
Disorders. 3rd ed. Revised. Washington, DC.
Disorders. 4th ed. Washington, DC.
6. Garfinkel PE, Lin E, Goering P, Spegg C, Goldbloom D, Kennedy S, Kaplan
A, Woodside DB. Purging and non-purging forms of bulimia nervosa in a
community sample. Int J Eating Disord 1996; 20:23–231.
7. Russell GFM. Bulimia nervosa: an ominous variant of anorexia nervosa. Psychol
Med 1979; 9:429–448.
8. Pope HG Jr, Hudson JI, Spitzer RL, Williams JBW. Revisions in DSM-III criteria
for bulimia nervosa. In: Garner DM, Garfinkel PE, eds. Diagnostic Issues in
Anorexia Nervosa and Bulimia Nervosa. New York: Brunner/Mazel, 1988:80–111.
9. Garfinkel PE, Lin E, Goering P, Spegg C, Goldbloom D, Kennedy S, Kaplan A,
Woodside DB. Bulimia nervosa in a Canadian community sample. Am J Psychiatry
1995; 152:1052–1058.
10. Garfinkel PE, Lin E, Goering P, Spegg C, Goldbloom D, Kennedy S, Kaplan A,
Woodside DB. Is amenorrhea necessary for the diagnosis of anorexia nervosa?
Br J Psychiatry 1996; 168:500–506.
11. Keys A, Brozek J, Henschel A, Mickelsen O, Taylor HL. The Biology of Human
Starvation, Vol. 1. Minneapolis: University of Minnesota Press, 1950.
12. Morton R. Phthisologica: Or a Treatise of Consumption. London: S Smith and
B Walford, 1694.
13. Woodside DB, Garfinkel PE, Lin E, Goering P, Kaplan AS, Goldbloom DS,
Kennedy SH. Men with full and partial syndrome eating disorders: community
comparisons with non-eating disordered men and eating disordered women. Am J
Psychiatry 2001; 158:570–574.
14. Crisp AH, Burns T. Primary anorexia nervosa in the male and female: a comparison
of clinical features and prognosis. In: Andersen AA, ed. Males with Eating
Disorders. New York: Brunner/Mazel, 1990:77–99.
15. Andersen A. Males with Eating Disorders. New York: Brunner/Mazel, 1990.
16. Treasure J, Thompson P. Anorexia nervosa in childhood. Br J Hosp Med 1988;
40:362–369.
17. Pobee, Kodwo A, LaPalio, Lawrence R. Anorexia nervosa in the elderly: a
multidisciplinary diagnosis. Clin Gerontologist 1996; 16:3–9.
18. Lee S, Ho TP, Hsu LKG. Fat phobic and non-fat phobic anorexia nervosa: a
comparative study of 70 Chinese patients in Hong Kong. Psychol Med 1993; 23:
999–1017.
19. Lee S. How abnormal is the desire for slimness? A survey of eating attitudes and
behaviour among Chinese undergraduates in Hong Kong. Psychol Med 1993;
23(2):437–451.
20. Lee S, Ho TP, Hsu LKG. Fat phobic and non-fat phobic anorexia nervosa—a
comparative study of 70 Chinese patients in Hong Kong. Psychol Med 1993; 23:
999–1004.
21. Hsu, George L, Lee, Sing. Is weight phobia always necessary for a diagnosis of
anorexia nervosa? Am J Psychiatry 1993; 150:1466–1471.
22. Grice DE, Halmi KA, Fichter MM, Strober M, Woodside DB, Treasure JT, Kaplan
AS, Magistretti PJ, Goldman D, Kaye WH, Bulik CM, Berrettini WH. Evidence
for a susceptibility gene for restricting anorexia nervosa on chromosome 1. Am J
Hum Genet 2002; 70:787–792.
23. Bulik CM, Devlin B, Bacanu SA, Thornton L, Klump KL, Fichter MM, Halmi KA,
Kaplan AS, Strober M, Woodside DB, Bergen AW, Ganjei K, Crow S, Mitchell J,
Rotondo A, Mauri M, Cassano G, Keel P, Berrettini WH, Kaye WH. Significant
linkage on chromosome 10p in families with bulimia nervosa. Am J Hum Genet
2003; 72:200–207.
2
Psychometric Assessment of Eating Disorders
Jacqueline C.Carter, Traci L.McFarlane, and Marion
P.Oimsted
University of Toronto
Comprehensive psychometric assessment is essential for diagnosis, case

formulation, treatment planning, and assessment of progress. The use of valid
and reliable measures is also vital to clinical research. Certain distinctive
features of eating disorders can make the clinical assessment process
challenging. Individuals with eating disorders have a reputation for minimizing
the seriousness of their symptoms. This is likely related to feelings of
ambivalence about changing certain valued features of the disorder (e.g., food
restriction, low weight). It may also be attributable to the shame and secrecy
surrounding some eating-disordered behaviors (e.g., binge eating and
vomiting). These factors can lead to denial or distortion in self-report (1). For
these reasons, the establishment of a trusting relationship is a prerequisite to
eliciting accurate information from eating disorder patients (2). The assessor
should communicate an empathic awareness of the mixed feelings many
individuals experience about acknowledging certain symptoms, considering
change, or starting treatment. In addition, incorporating the principles of
motivational viewing into the assessment process can be useful (3). Taking a
collaborative approach and focusing on the aspects of the illness that are
experienced as distressing to the patient may be particularly helpful.
The purpose of this chapter is to describe psychometric measures of the
specific and general psychopathology of eating disorders. It is beyond the scope
of the chapter to review all existing measures. For the assessment of specific
psychopathology, we have decided to focus on widely used tests that were
designed to be used as screening instruments in nonclinical populations as well
as measures designed to measure clinical eating disorders. The focus is on more
comprehensive measures, but selected measures of specific constructs (e.g.,
motivation to change) are discussed as well. Both structured interviews and self-
report instruments are covered. For a detailed review of other eating disorder
assessment methods, see Allison (4). With respect to general psychopathology,
there are a multitude of well-established measures of common associated
symptoms, including depression, anxiety, self-esteem, and interpersonal
20 CARTER ET AL.
difficulties. The interested reader is referred to Groth-Marnat (5) for more

information about general psychological assessment. The present chapter
focuses on the tests that we would recommend for the purposes of treatment
planning and measuring treatment outcome in persons with eating disorders.
The areas covered are in line with the American Psychiatric Association Practice
Guidelines for eating disorders (6). Both adult and child/adolescent measures
are discussed. A clinical interview is also an important component of the
assessment process and is necessary to obtain a psychiatric history and make a
diagnosis. This aspect of assessment is covered in Chapter 1 of this volume on
diagnosis.
SPECIFIC PSYCHOPATHOLOGY
Specific psychopathology refers to features that are distinctive to eating
disorders. The clinical features of the two recognized eating disorders, anorexia
nervosa (AN) and bulimia nervosa (BN), overlap significantly. The central
features of BN are binge eating and inappropriate compensatory behaviors to
prevent weight gain. In the fourth edition of the Diagnostic and Statistical Manual
of Mental Disorders (DSM-IV; 7), binge eating has two defining characteristics:
(a) the consumption of a large amount of food given the circumstances; and
(b) a sense of loss of control. Compensatory behaviors are divided into two
categories: purging and nonpurging. Purging techniques include self-induced
vomiting and laxative or diuretic misuse. Nonpurging compensatory behaviors
include strict dieting, fasting, and excessive exercise. The central features of AN
are extreme food restriction and a refusal to maintain body weight at or above a
minimally normal weight for age and height (i.e., body mass index ≤ 17.5).
Approximately half of those presenting with AN also report binge eating and/or
purging. Undue influence of body shape and weight on self-evaluation is
common to both disorders. Binge eating disorder, a provisional diagnostic
category in the DSM-IV, is characterized by recurrent binge eating in the
absence of compensatory behaviors.
Structured Interviews
Structured interviews have the advantage of affording a more detailed
examination of symptoms as compared with self-report inventories as they allow
the opportunity for probing and clarification. On the other hand, interviews
rely on the subject’s self-report and the results may therefore still be influenced
by recall bias, denial, or minimization of symptoms (8). A number of structured
interviews to measure eating disorder symptoms have been developed.
Examples include the Clinical Eating Disorder Rating Instrument (9) and the
PSYCHOMETRIC ASSESSMENT 21
Structured Interview for Anorexia and Bulimia (10). However, these

instruments have not been widely used, and very little information is available
on their psychometric properties. Two of the most widely used structured
interviews for the assessment of eating disorders are the Eating Disorder
Examination (11) and the Yale-Brown-Cornell Eating Disorder Scale (12,13).
The eating disorder section of the Structured Clinical Interview for DSM-IV
(SCID; 14,15) has also been widely employed to make eating disorder diagnoses
in clinical research. It will be discussed later in this chapter under “General
Psychopathology.”
Eating Disorder Examination (EDE; 11)

The EDE is considered by many investigators to be the best available method for
assessing the specific psychopathology of eating disorders (e.g., 16). The EDE
consists of four subscales: Restraint, Shape Concern, Weight Concern, and
Eating Concern. The subscale scores range from 0 (no pathology) to 6 (extreme
pathology). Individual item scores provide either frequency or severity rating
for key behavioral or attitudinal aspects of eating disorders. In addition, there
are individual items to assess the presence and frequency of key behaviors,
including binge eating and weight control behaviors, over the previous 28 days.
To allow DSM-IV diagnoses, certain items assess the previous 2 months as well.
Operationalized diagnostic criteria are provided in the interview manual (11).
The EDE is investigator based (meaning that key terms are defined by the
interviewer, not the respondent) and requires properly trained assessors. It may
take up to an hour to administer. At the outset, the interviewer refers to a
calendar and helps the subject identify key events to enhance recall of the time
period being examined. All key concepts are defined in the manual, and explicit
instructions are given as to how to rate each item. The judgment as to whether
an episode of eating is a “large amount” is made by the interviewer.
The EDE has been demonstrated to be a reliable and valid instrument. The
internal consistency of the four subscales is satisfactory (17). High interrater
reliability (18,19) and good test-retest reliability have been found (20) in
different settings. In addition, the EDE has been demonstrated to have good
criterion-related validity (17,19) and convergent validity (18). Normative data
for AN, BN, and normal controls are available (11). A children’s version of the
EDE interview has been developed (21).
Yale-Brown-Cornell Eating Disorder Scale (YBC-EDS; 12)

The YBC-EDS is a clinician-rated semistructured interview designed to measure
eating disorder symptom severity. Target symptoms are preoccupations and
22 CARTER ET AL.
rituals. Preoccupations are defined as “ideas, thoughts, images or impulses that

repeatedly enter your mind; they may seem to occur against your will.” Rituals
are defined as “behaviors or acts that you feel driven to perform.” Target
symptoms are identified and assessed in terms of time occupied by symptoms,
interference with functioning, distress caused by symptoms, and degree of
control over symptoms. Items are rated on a 0 (none/not present) to 4 (extreme)
scale. The four preoccupation items and the four ritual items are added to
obtain preoccupation and ritual subtotals. The sum of these yields a total score.
The YBC-EDS is not intended for use as a diagnostic instrument.
The YBC-EDS has been demonstrated to have high interrater reliability and
good internal consistency (12). Its convergent validity was established by
comparing scores on the YBC-EDS with scores on other validated measures of
eating disorder symptoms (12).
Self-Report Instruments
Self-report instruments offer a number of advantages over structured interview
measures. They are completely standardized, require little or no formal training
to administer, are economical, and may be completed in a group format. On the
other hand, Wilson (16) identified a number of shortcomings shared by most
self-report measures of eating disorder features, including the lack of a specific
time frame and the failure to ask directly about the frequency of key eating
disorder behaviors. In addition, most fail to define key terms such as “binge
eating.”
Numerous measures of certain aspects of eating disorder psychopathology are
available, e.g., the Restraint Scale (22), the Three Factor Eating Questionnaire
which primarily measures restraint and disinhibition (23), and the Binge
Eating Scale (24). For the purposes of this chapter, we will concentrate on
comprehensive measures that cover a broad range of the core cognitive and
behavioral features of eating disorders. Selected measures of specific constructs
(e.g., motivation to change, body image) will also be discussed. The following
self-report measures may be used as screening instruments in nonclinical
samples or as measures of eating disorder symptom severity for clinical or
research purposes. None are intended to be the sole basis for making a
diagnosis, although they may be a useful adjunct to a diagnostic interview.
Eating Attitudes Test (EAT; 25,26)

The EAT is a self-report questionnaire that was developed to measure the
symptoms of AN in adults and older adolescents (16 years and older) (27). The
original EAT consisted of 40 items; an abbreviated 26-item version was later
derived through factor analysis (EAT-26; 26). Each item is rated on a 6point
Likert scale ranging from “never” to “always.” The most symptomatic answer
receives a score of 3, the next adjacent response a score of 2, and so on. The
three least symptomatic responses receive a score of 0. No specific time frame is
given. In addition to a total score, the EAT-26 yields three subscales: Dieting,
Bulimia/Food Preoccupation, and Oral Control. The EAT-26 total score is highly
correlated with the EAT-40 total score and the 14 eliminated items were found
to be redundant (26). A children’s version of the EAT-26 intended for use with
those younger than 16 years has been developed (ChEAT; 28).
The EAT has been widely employed as a screening instrument in
epidemiological research to identify suspected cases of eating disorders. For
screening purposes, a total score above 20 on the EAT-26 is the recommended
cutoff. On average, most studies have found the EAT-26 to have a relatively low
positive predictive value (e.g., 29,30). That is, a significant proportion of those
scoring above 20 do not have clinical eating disorders upon diagnostic
interview. Regarding the rate of false negatives, one study reported that 1 % of
low scorers (i.e., <20) were subsequently identified as cases.
Carter and Moss (31) reported the test-retest reliability of the EAT-40 over a
2- to 3-week period to be 0.84. Both the EAT and EAT-26 have been found to
correlate moderately to highly with other measures of eatingdisordered
attitudes and behavior (e.g., 32–35). Gross et al. (32) found that the EAT-40
differentiated BN subjects from normal controls. However, it has not been
found to differentiate between AN and BN (36). The test-retest reliability
coefficient for the 26-item children’s version of the EAT following a 3-week
interval has been found to be 0.81 and it has been shown to have good internal
consistency (28). Scores on the 26-item ChEAT have been shown to correlate
significantly with measures of weight management behavior and body
dissatisfaction among middle-school girls (37).
Eating Disorder Inventory (EDI; 38)

The EDI is a widely used self-report measure of symptoms commonly associated
with AN and BN. It comprises three subscales that assess attitudes and behaviors
concerning eating, weight, and shape (Drive for Thinness, Bulimia, Body
Dissatisfaction), and five subscales that measure psychological constructs
clinically relevant to the eating disorders (Ineffectiveness, Perfectionism,
Interpersonal Distrust, Interoceptive Awareness, Maturity Fears). A second
edition of the EDI (EDI-2) has been published that retains the original 64 items
and adds 27 items comprising three additional subscales: Asceticism, Impulse
Regulation, and Social Insecurity (27). Items are rated using a 6point scale
ranging from “never” to “always.” The EDI does not specify a time frame for
24 CARTER ET AL.
responding to the items. Raw subscale scores are plotted on profile forms to
allow comparison with norms for eating disorder patients (N = 889) and a
female college student control group (N = 205). The EDI is ssuitable for those
12 years and older. Criterion-related and construct validity for the eight
original EDI subscales have been well documented (e.g., 32, 38–40). Wear
and Pratz (41) obtained 3-week test-retest reliability coefficients above
0.80 for all of the original subscales except Maturity Fears. A cutoff score of 14
on the Drive for Thinness subscale has been recommended for screening
purposes (27). However, this measure is most useful as a measure of eating
disorder symptom severity for clinical or research purposes.
Kids Eating Disorders Survey (KEDS; 42,43)

The Kids Eating Disorders Survey (KEDS) is a 12-item screening questionnaire
designed for eating-disordered behaviors and attitudes among latencyage and
early adolescent children. Detailed scoring instructions have been published
elsewhere (44). The KEDS assesses the presence or absence of weight loss
dieting; fasting; self-induced vomiting; excessive exercise; use of diet pills,
diuretics, and laxatives; as well as binge eating, with a detailed definition of
binge eating provided. Body dissatisfaction is measured through the use of figure
drawings. Respondents are asked to select a current and an ideal figure
drawing. The body dissatisfaction score is the difference between the two
drawings. The KEDS has been shown to have good internal consistency
(Chronbach’s α = 0.73) and total score test-retest reliability after an average
interval of four months of 0.83 (43).
Revised Bulimia Test (BULIT-R; 45)

The BULIT-R is a 36-item questionnaire that was developed to measure the
symptoms of BN. It contains 28 items that are scored and 8 unscored items that
ask about weight control behavior. Items are rated using a five-choice
multiple-choice format. The BULIT-R has been found to have high internal
consistency (Cronbach’s coefficient α = 0.97) and a 2-month test-retest
reliability of 0.95 (45). It also has demonstrated concurrent and criterionrelated
validity (46). Thelen et al. (45) reported norms of bulimia nervosa (N = 24) and
female control (N = 116) participants. The BULIT-R may be employed as a
screening instrument or as a measure of symptom severity. A cutoff score of 85
has been recommended for screening purposes (45).
Eating Disorder Examination Questionnaire (EDE-Q; 47)

The self-report version of the EDE, the EDE-Q, was designed to address some
of the shortcomings of self-report instruments (47). Like the EDE, the 36-item
EDE-Q has a 28-day time frame and asks directly about the frequency of key eating
disorder behaviors. It is intended for use with older adolescents and adults. Its
items are closely based on the corresponding questions from the EDE interview
and it uses the same 7-point rating scheme. The EDE-Q also generates four
subscale scores (Restraint, Eating Concern, Shape Concern, and Weight
Concern) as well as a global score that is the average of the four subscales.
Respondents rate each item on a 7-point rating scale (i.e., 0–6) indicating the
number of days out of 28 on which particular behaviors, attitudes, or feelings
occurred. Scores of 4, 5, or 6 are interpreted as more likely to be of clinical
severity.
There have been several studies on the psychometric properties of the
EDE-Q. Luce and Crowther (48) reported 2-week test-retest reliability
coefficients ranging from 0.81 to 0.94 across the four subscales and 0.57 to 0.70
for items measuring the frequency of key behaviors. Given the 28-day time frame
of the EDE-Q, the somewhat lower stability of the behavioral items may reflect
fluctuations in symptom frequency.
Five studies have examined the convergent validity of the EDE-Q by
comparing it with the EDE interview (47,49–52). In general, good agreement
was found across all subscales. There was also good agreement in most
studies for the assessment of self-induced vomiting and laxative misuse, but
significant differences were found for the frequency of binge eating. Finally,
Wilson et al. (53) reported discriminant validity data on the EDE-Q. In this
study, several EDE-Q items were shown to discriminate between obese binge
eaters and nonbinge eaters. The self-report version appears to be an adequate
substitute for the EDE interview in the assessment of most eating disorder
features. A children’s version has been developed, and adolescent norms for this
version have recently been published (54). The EDE-Q is particularly useful as a
self-report measure of the frequency of key behaviors for clinical or research
purposes.
Mlzes Anorectic Cognitions Scale (MACS-R; 55)

The MACS-R is designed to measure attitudes and beliefs commonly associated
with eating disorders. Items are rated on a 5-point Likert scale ranging from
“strongly disagree” to “strongly agree.” Three subscales were derived through
factor analysis: (a) self-control and self-esteem; (b) weight and approval; and
(c) rigid weight regulation and fear of weight gain (56). Test-retest reliability
26 CARTER ET AL.
was found to be 0.78 over a 2-month interval for the original version and the
coefficient a was 0.90 for the revised version. The test-retest reliability of the
revised scale has not been studied. A number of studies support the concurrent
and discriminant validity of the MACS, and it has been shown to be sensitive to
change with treatment (57). This measure may be useful for elucidating eating
disorder cognitions and measuring cognitive change with treatment.
The Shape- and Weight-Based Self-Esteem Inventory

(SAWBS; 58)
Determining self-worth based on an assessment of the body (i.e., shape- and
weight-related self-esteem) is the core cognitive psychopathology in eating
disorders (59), and it has been shown to predict relapse in BN (60). The
SAWBS Inventory is a self-report measure that involves completing a pie
diagram to measure shape- and weight-related self-esteem in adults. An
adolescent version has also been developed by the authors to be used with
adolescents aged 13–18 years (61). Participants are asked to select from a list of
personal attributes the ones that are important in terms of how they felt about
themselves (e.g., weight and shape, personality, intimate relationships) over the
past 4 weeks. A second step involves rank ordering each attribute. Then
participants are asked to divide a circle into pieces such that the size of each
piece reflects how much their self-esteem is based on each of the ranked
attributes. The shape- and weight-based self-esteem score is the angle of the
shape and weight piece of the pie. The researchers found a mean angle of 58°
for the weight and shape wedge in their sample of non-eating disordered
women. They also found that greater evaluation of the self on the basis of
weight and shape was related to higher levels of depression and lower levels of
overall self-esteem. In another study, Geller et al. (62) found that
eatingdisordered patients reported greater weight-based self-esteem (mean
angle 145°) than both a psychiatric control group (mean angle 63°) and a
student control group (mean angle 59°). Overall, the SAWBS Inventory has
been shown to have good psychometric properties in both adults and
adolescents, including test-retest reliability after a 1-week interval of 0.81 and
0.77, respectively (58,61,62).
Motivation to Change
As previously mentioned, individuals with eating disorders are typically
ambivalent about recovery. This is related to the ego-syntonic nature of certain
feature of eating disorders (e.g., dietary restriction and weight con trol). The
development and validation of measures to assess readiness to change in eating

disorders is still at an early stage.
University of Rhode Island Change Assessment Scale

(URICA; 63,64)
This widely used 32-item self-report measure of readiness to change (also
commonly referred to as The Stages of Change Questionnaire) was originally
developed to assess motivation to change among substance abusers. However,
since its item do not refer to any specific behaviors but rather to “problems that
need changing,” it has been used in different populations, including eating
disorders (e.g., 65). It has four subscales corresponding to four stages of
change: Precontemplation (e.g., “As far as I’m concerned, I don’t have any
problems that need changing”), Contemplation (e.g., “It might be worthwhile
to work on my problem”), Action (e.g., “I am really working hard to change”),
and Maintenance (e.g., “I am working to prevent myself from having a relapse of
my problem”). Items are rated on a 5-point Likert scale ranging from “strongly
agree” to “strongly disagree” and a mean score is yielded for each subscale. One
limitation of using this questionnaire to measure readiness to change in eating
disorders is that the scores are not easily interpreted because the items do not
refer to disorder-specific behaviors or attitudes.
Anorexia Nervosa Stage of Change Questionnaire

(ANSOCQ; 66)
The ANSOCQ is a newly developed measure of motivation to change that is
based on the stages-of-change model but is tailored to the specific symptoms of
AN. It is designed to measure readiness to reach a normal weight, to cease
extreme weight-control behaviors, to eat normally, and to consume avoided
foods. Each item contains five statements corresponding to five stages of
change: Precontemplation (e.g., “As far as I’m concerned, I do not need to gain
weight”), Contemplation (e.g., “In some ways, I think I might be better off if I
gained weight”), Preparation (e.g., “I have decided that I will attempt to gain
weight”), Action (e.g., “At the moment I am putting in a lot of effort of gain
weight”), and Maintenance (e.g., “I am working to maintain the weight gains I
have made”). For each item, respondents are asked to select the statement that
best describes their current attitude or behavior. The ANSOCQ has been
demonstrated to have good internal consistency and 1week test-retest reliability
as well as acceptable concurrent and predictive validity (66). This instrument
may be useful in assessing the impact of level of motivation to change on
treatment outcome in eating disorders as well as in assessing the effectiveness of
28 CARTER ET AL.
motivational enhancement interventions. Further study of this measure is

needed.
Concerns About Change Questionnaire (CAC; 67)

The CAC is a 112-item self-report measure of fears about change that is still
under development. It was designed to apply to all psychiatric conditions
including eating disorders, anxiety disorders, and substance abuse. Respondents
are asked to rate on a 5-point scale the extent to which each item applies to
them. The questionnaire comprises 17 rational subscales (e.g., Unable to
Change, Fear of Risks, Fear of Maturity, Fear of Sexuality, Fear of the Process of
Change, Fear of Loss of Accomplishment, Fear of Interpersonal Loss). The
internal consistency of the subscales is high (0.80–0.91). Although it requires
further validation, this measure may be useful in determining the functional
significance of the eating disorder and the obstacles to recovery.
GENERAL PSYCHOPATHOLOGY
General psychopathology refers to features that are shared with other
psychiatric conditions. In eating disorders, commonly associated
psychopathology includes depression, anxiety, low self-esteem, interpersonal
problems, and personality disturbance. In addition, features of obsessive-
compulsive disorder, substance abuse, posttraumatic stress disorder, and
dissociative disorder may be present. Such concerns may represent premorbid
conditions, or may indeed be caused or exacerbated by the eating disorder
itself. Many of these symptoms have been shown to improve with successful
management of the eating disorder. Patients who are severely depressed or
severely anxious may need to be stabilized before undertaking a specialized
symptom-focused eating disorder treatment program. Many comorbid concerns
are entangled with the eating disorder and will need to be addressed during the
eating disorder treatment (e.g., obsessive-compulsive disorder), while
longstanding interpersonal and personality disturbances may require additional
treatment after the eating disorder has been resolved. Numerous well-
established measures of general psychopathology exist, and it is beyond the scope
of this chapter to review them all. We will describe the instruments that we
recommend for clinical and research purposes.
Structured Interviews
Structured Clinical Interview for DSM-IV (SCID; 14,15)

The SCID is a widely used semistructured interview for making DSM-IV axis I
and axis II (personality disorder) diagnoses. It incorporates the use of obligatory
questions, operational diagnostic criteria, a categorical system for rating
symptoms, and an algorithm for arriving at a final diagnosis. In addition, it allows
the interviewer to probe and restate questions, challenge the respondent, and
ask for further clarification. The SCID can take between 1 and 3 hours to
administer. Although it was originally developed for use in research by trained
clinicians, research staff with varying levels of prior clinical experience can be
trained to reliably administer the SCID. The users guides for the SCID-I and
SCID-II provide basic training information. In addition, there is an 11-hour
videotape training program that includes examples of interviews with patients.
The SCID for Dissociative Disorders (SCID-D; 68) was developed to enable
clinically trained interviewers to assess the nature and severity of the five
dissociative symptoms (amnesia, depersonalization, derealization, identity
confusion, and identity alteration) and to diagnose the presence of DSM-IV
dissociative disorders. The reliability and validity of the SCID I and II as well as
the SCID-D have been documented in several studies (68–71). Additional
information about SCID training can be found at www.scid4.org.
Diagnostic Interview for Children and Adolescents

(DICA; 72,73)
The DICA is a semistructured interview designed to determine the presence or
absence of symptoms and other criteria required for DSM-IV lifetime diagnoses
in children and adolescents. There are three separate interviews designed for
children (6–12 years), adolescents (13–18 years), and parents. The interview is
designed to be administered by highly trained interviewers and takes
between 1–2 hours to complete. Although results vary depending on the
disorder, high test-retest reliability and moderate convergent validity have been
demonstrated (74,75).
Yale-Brown Obsessive Compulsive Scale (Y-BOCS; 76,77)

The Y-BOCS is a 10-item semistructured interview measure of
obsessivecompulsive symptoms. It was not designed for use as a diagnostic
instrument. Each item is rated from 0 (no symptoms) to 4 (extreme
symptoms). The total Y-BOCS score is the sum of items 1–10 (range, 0–40).
30 CARTER ET AL.
There are separate subtotals for the severity of obsessions (items 1–5) and
compulsions (items 6–10). Symptoms are assessed in terms of how much of the
respondent’s time they occupy, the extent to which they interfere with normal
functioning, the degree of subjective distress they cause, and the extent to
which they are actively resisted by the patient and can be controlled by the
patient. The YBOCS has been documented to have strong psychometric
properties. It shows good internal consistency (76,78), excellent interrater
reliability (76,78), and acceptable test-retest reliability (78). Evidence for its
convergent validity has been established in studies that compared the Y-BOCS
total score with other standard self-report and behavioral measures of obsessive-
compulsive disorder (77–79). A self-report version of the Y-BOCS has been
developed and has been shown to be highly correlated with the YBOCS
interview (80,81). There is also a children’s version of the Y-BOCS that has
been shown to be reliable and valid (82).
Clinician-Administered PTSD Scale (CAPS; 83)

The CAPS is a 34-item semistructured interview designed to measure the 17
symptoms of PTSD according to the DSM-IIIR (84). Each symptom is rated on
a 5-point scale in terms of frequency and intensity. The CAPS yields two total
scores, one for frequency and one for intensity, as well as two subscores for
each of the reexperiencing, avoidance, and arousal subscales. It may be used to
make current and lifetime PTSD diagnoses and also for measuring the severity
of PTSD symptomatology. The CAPS shows high interrater reliability, test-
retest reliability, and internal consistency (83,85,86). It has also been shown to
have good concurrent validity, as indicated by significant correlations with
self-report measures of PTSD (83).
SELF-REPORT QUESTIONNAIRES
Depression
Beck Depression Inventory, Second Edition (BDI-II; 87)

The BDI-II is a 21-item self-report instrument for measuring the severity of
depression in adults and adolescents 13 years and older. The BDI-II was
developed as an indicator of the presence and degree of depressive symptoms
consistent with the DSM-IV criteria. Respondents are asked to endorse
symptoms that apply to them during the past 2 weeks. Scores of 0–13 reflect
minimal depression, 14–19 mild depression, 20–28 moderate depression, and
scores of 29 or more severe depression. This measure is easily administered and

scored. However, because depression is associated with suicidal risk in eating-
disordered patients it should be interpreted by a professional with appropriate
clinical training and experience. Moreover, the clinician should pay particular
attention to the responses to item 2 (pessimism) and item 9 (suicidal thoughts
or wishes) as indicators of possible suicide risk. The BDI-II has been shown to
have sound psychometric properties, including coefficient a values of 0.92 for
patients and 0.93 for college students, and test-retest reliability after a 1-week
interval of 0.93 (87).
Depression Anxiety and Stress Scales (DASS; 88,89)

This self-report measure is available either in a 42-item or a 21-item version.
The DASS has the advantage of assessing and discriminating between depression
(e.g., “I felt down-hearted and blue”) and anxiety symptoms (e.g., “I found
myself getting agitated”), as well as a more general stress dimension.
Participants are asked to respond to each statement using a 4 point scale ranging
from 0 (did not apply to me at all) to 3 (applied to me very much or most of the
time) using the time frame of “over the past week.” The factor structure and
reliability of the DASS have been established in clinical populations, and it has
been shown to have very good concurrent and discriminant validity (90).
Anxiety
Beck Anxiety Inventory (BAI; 91)

The BAI is a 21-item self-report measure of anxiety symptoms. The BAI asks
respondents to rate symptoms that have bothered them over the past week
using a 4-point scale ranging from 0 (not at all) to 4 (severely). This measure
places an emphasis on symptoms of physical hyperarousal (e.g., numbness or
tingling, heart pounding or racing, difficulty breathing). The BAI has sound
psychometric properties (91) and has also demonstrated treatment
sensitivity (92). This is a useful measure to screen for the presence and severity
of anxiety symptoms.
The BAI and the DASS (described under depression) can be used to screen
for the presence and degree of physical hyperarousal indicating anxiety and
general stress. However, to identify specific anxiety disorders more specialized
measures are required. It is beyond the scope of this chapter to review all possible
means of identifying and measuring specific anxiety disorders. Some suggestions
include the Fear Questionnaire (93) to measure the severity of common
32 CARTER ET AL.
phobias, the Panic Disorder Severity Scale (94) to assess panic disorder severity
in patients already diagnosed with panic disorder, the Penn State Worry
Questionnaire (95) to measure the tendency to worry excessively indicative of
generalized anxiety disorder, the Social Phobia Inventory (96) to measure the
fear, avoidance, and physiological arousal associated with social phobia, and the
Yale-Brown Obsessive-Compulsive Scale (81). For more information about
these and other scales, the reader is referred to Antony et al. (97) for a
comprehensive review of anxietyrelated assessment measures.
Self-Esteem
Rosenberg Self-Esteem Scale (RSES; 98,99)

This is a 10-item self-report inventory that assesses self-esteem in adolescents
and adults. This measure is significantly correlated with measures of depression
and neuroticism (98,99), and other measures of self-esteem (100). The RSES
has strong construct, convergent, and discriminant validity (99,101) and has
been shown to predict outcome in the management of bulimia nervosa (102).
Respondents are asked to strongly agree, agree, disagree, or strongly disagree to
a series of statements about the self (e.g., “I feel I do not have much to be proud
of”) considering how they “usually feel.” Scores range from 10 (extremely low
self-esteem) to 40 (extremely high self-esteem). In a recent study of 100
respondents, it was shown that eating-disordered participants scored
significantly lower (mean score 22) than both chronic dieters (mean score 31)
and a student control group (mean score 33) in terms of trait self-esteem as
measured by the RSES (103).
Substance Abuse
Michigan Alcoholism Screening Test (MAST; 104)

The MAST is a 25-item self-report questionnaire that was developed to detect
alcohol-related problems in adults. The MAST can be used clinically with
individual patients, both for diagnosis and as a means of providing a short but
comprehensive profile of the medical, psychological, and social functions that
are interfered with by the use of alcohol (105). The scale has been shown to
have high internal consistency (106), satisfactory test-retest reliability, and good
convergent validity (107).
Adolescent Drinking Index (ADI; 108)

The ADI is a 24-item self-report measure that was developed to screen for
alcohol-related problems in adolescents (12–17 years old). The ADI is designed
to identify adolescents who should be referred for further alcohol evaluation or
treatment. The instrument contains two subscales: self-medicating problem
drinking and aggressive, rebellious drinking behavior. The ADI has high internal
consistency, good test-retest reliability, and has demonstrated convergent and
discriminant validity (108,109).
Drug Abuse Screening Test (DAST; 110)

The DAST is a 20-item self-report measure designed to detect drug abuse in
adults during the past year. This measure has high internal consistency, and
good concurrent and discriminant validity (110,111). There is also an
adolescent version of this measure that has demonstrated sound psychometric
properties in adolescents aged 13–19 years (112).
Other Psychological Symptoms
Dissociative Experiences Scale (DES; 113,114)

The DES is a self-report questionnaire that was developed to measure normal
and pathological dissociative phenomena in adults (18 years and older). The
original DES consists of 28 descriptions of dissociative experiences.
Respon dents are asked to indicate on a visual analog scale what percentage of
the time each experience occurs to them in their daily life. Subscales include
amnesia, depersonalization-derealization, and absorption experiences. The total
score is an average of each response.
The scale has been shown to have good psychometric properties including a
coefficient a of 0.96 and test-retest reliability of 0.93 for the total score (115).
In addition, the DES has demonstrated convergent and predictive validity (116)
and can be used as a screening measure to identify dissociative disorders in a
clinical population (117). There is also an adolescent version of the scale
(Adolescent Dissociative Experiences Scale; 118) that has been shown to have
sound psychometric qualities in adolescents aged 12–18 (118–120).
Brief Symptom Inventory (BSI; 121)

The BSI is a 53-item self-report symptom inventory designed to measure a
range of psychological symptoms. It is essentially a brief form of the Symptom
34 CARTER ET AL.
Checklist-90-R (122). The BSI is useful to screen for a wide variety of

psychological symptoms in clinical and research settings where time is limited.
Respondents are asked to indicate how much a problem has distressed or
bothered them during the past 7 days. Each item of the BSI is rated on a 5-point
scale ranging from 0 (not at all distressed) to 4 (extremely distressed). There
are nine subscales (i.e., Somatization, Obsessive-compulsive, Interpersonal
Sensitivity, Depression, Anxiety, Hostility, Phobic Anxiety, Paranoid Ideation,
and Psychoticism) and three global indices (i.e., global severity index, positive
symptom total, positive symptom distress index) (121). The a coefficients for
all nine subscales are very good, ranging from a low of 0.71 on the Psychoticism
dimension to a high of 0.85 on Depression. Test-retest reliability after a
2-week interval was high for the global indices scales, ranging from 0.80 to
0.90 (121,123,124). Norms are provided for adults (i.e., nonpatients and
psychiatric patients) and for adolescents (aged 13–19) (121).
PTSD Symptom Scale—Self-Report (PSS; 125,126)

The PSS-SR is a 17-item self-report scale designed to measure the presence and
frequency of PTSD symptoms over the previous 2 weeks in individuals with a
known trauma history. Its items correspond to the 17 DSM-IIIR diagnostic
criteria for PTSD. Each item is rated from 0 (not at all) to 3 (very much). The
total severity score is the sum of the severity ratings for the 17 items. The
PSS-SR has been shown to have satisfactory internal consistency, high test-retest
reliability, and good convergent validity with the SCID PTSD module (125).
For diagnostic purposes, the PSS-SR should be used in combination with a
diagnostic interview.
Interpersonal Functioning and Personality
Inventory of Interpersonal Problems (IIP; 127)

The IIP is available as a 32- or 64-item self-report instrument that identifies
salient interpersonal difficulties that may have important clinical implications.
This measure evaluates overall interpersonal difficulty, and also assesses eight
specific domains of interpersonal functioning (i.e., domineering/controlling,
vindictive/self-centered, cold/distant, socially inhibited, nonassertive, overly
accommodating, self-sacrificing, intrusive/needy). The eight subscales are
organized around two major factors: control/submission (dominance) and
positive/negative valence (affiliation). Items are divided into two sections: one
section begins with “The following are things you find hard to do with other
people” (e.g., join in on groups, confront people with problems that come up);
the other section begins with “The following are things that you do too much”
(e.g., I am overly generous to other people, I argue with other people too
much). Response options range from “not at all” to “extremely.” Both the IIP-64
and the IIP-32 have moderate to high internal consistency for the total scales
corresponding to 0.96 and 0.93, respectively, and both have test-retest
reliability after a 1-week interval of 0.78 in a nonclinical sample. Test-retest
reliability is higher in clinical samples. Normative information is based on adults
aged 18–89 (127,128).
Dimensional Assessment of Personality Pathology—Basic

Questlonnaire (DAPP-BQ; 129)
The DAPP-BQ is a 290-item self-report questionnaire developed to assess
dimensions of personality pathology. It is composed of 18 subscales derived
through factor analysis corresponding to easily interpreted and theoretically
meaningful personality patterns. The DAPP-BQ scales are internally consistent;
estimates of coefficient a for the scales range from 0.87 to 0.94. It also has
demonstrated good convergence with other self-report measures of personality
traits (e.g., 130,131). Norms for a general adult population control group are
available (129). This instrument was used in a recent study of personality
pathology in patients with eating disorders (132). For a detailed discussion of
the assessment of personality disorder, the reader is referred to Clark and
Harrison(133).
Family Functioning
Disturbances in family functioning are common in the eating disorders, and
family therapy has been found to be helpful, particularly for younger
patients (134). Therefore, it can be useful to include a measure of family
functioning in the assessment battery. For a detailed discussion of self-report
instruments for family assessment, see Skinner (135). We recommend using
one of the following measures.
Family Environment Scale (FES; 136)

The FES is a 90-item true-false self-report measure designed to assess an
individual’s perceptions of the current social environment or climate of his or
her whole family unit. It may be administered to individuals 12 years and older.
Ten subscales reflecting three dimensions of family functioning were derived
through factor analysis: (a) Interpersonal Relationships (i.e., Cohesion,
36 CARTER ET AL.
Conflict, and Expressiveness subscales); (b) Personal Growth (i.e.,

Independence, Achievement Orientation, Intellectual-Cultural Orientation,
Active Recreational Orientation, and Moral-Religious Emphasis subscales); and
(c) System Maintenance (i.e., Family Organization and Control subscales). The
10 subscales have been shown to be internally consistent with Cronbach’s a
coefficients ranging from 0.61 to 0.78. Adequate test-retest reliability for a
2-month interval has also been demonstrated (136), and its criterion-related
validity has been established in numerous studies (137). The manual presents
normative data for normal and distressed families.
Family Assessment Measure (FAM-III; 138,139)

The FAM-III is a self-report measure of family functioning that was designed to
measure both family strengths and weaknesses from the perspective of different
family members. It may be administered to family members who are at least
10 years old. The basic processes assessed by the FAM-III include Task
Accomplishment, Role Performance, Communication, Affective Expression,
Involvement, Control, and Values and Norms. The FAM-III consists of three
subscales: a 50-item General Scale examining the overall family system; a
42-item Dyadic Relationships Scale measuring the quality of relationships
between specific pairs in the family; and a 42-item Self-Rating Scale assessing an
individual’s perception of his or her functioning in the family. It takes about
30–40 minutes to administer. The FAM-III subscales have been shown to have
good internal consistency (a coefficients of 0.86–0.95) and adequate 12-day
test-retest reliability (median correlations of 0.57–0.66) (138). FAM-III
scores have been shown to differentiate between clinical and nonclinical families
(140,141). Studies of its construct validity have found that the FAM-III is highly
correlated with other measures of family functioning including the FES
(142,143). The manual presents norms and interpretative guidelines.
SUMMARY AND RECOMMENDATIONS

This chapter has provided an overview of psychometric instruments relevant to
the assessment of eating disorders that can be used clinically and/or in research.
A comprehensive assessment should include measures of specific eating disorder
features as well as measures of general psychopathology. We have described
measures of eating disorder symptoms that may be used as screening
instruments as well as those designed to measure clinical eating disorders.
Recommended measures of general psychopathology were also discussed.
In specialist treatment settings, we would recommend using the EDE to
measure core diagnostic features plus an additional dimensional self-report
measure of the severity of eating disorder symptomatology (e.g., EDI) to measure

specific psychopathology. For the measurement of general psychopathology, at a
minimum we would suggest including a self-report measure of depression,
anxiety, self-esteem, interpersonal difficulties, and family functioning. Including
a measure of motivation to change may also be helpful in tailoring the treatment
plan to the patient’s level of readiness to change. For example, patients who are
not ready to make behavioral changes may be referred to a purely
psychoeducational intervention. Providing patients with feedback about their
test results is another important element of the clinical assessment process but is
beyond the scope of this chapter. It goes without saying that such feedback
should be provided by qualified professionals. In primary care settings, a
screening instrument (e.g., EAT or KEDS) may be useful in determining
whether a referral for specialist treatment should be considered.
REFERENCES
1. Vitousek K, Daly J, Heiser C. Reconstructing the internal world of the
eatingdisordered individual: overcoming denial and distortion in self-report. Int J Eat
Disord 1991; 10:647–666.
2. Crowther JH, Sherwood NE. Assessment. In: Garner DM, Garfinkel PE, eds.
Handbook of Treatment for Eating Disorders. New York: Guilford Press, 1997:
34–39.
3. Miller WR, Rollnick S. Motivational Interviewing. 2d ed. New York: Guilford
Press, 2002.
4. Allison DB. Handbook of Assessment Methods for Eating Behaviors and Weight-
Related Problems—Measures, Theory and Research. Thousand Oaks,CA: Sage
Publications, 1995.
5. Groth-Marnat G. Handbook of Psychological Assessment. 3d ed. New York: John
Wiley and Sons, 1996.
6. American Psychiatric Association. Practice guidelines for the treatment of patients
with eating disorders. Am J Psychiatry 2000; 157:1–39.
Disorders. 4th ed. Washington, DC, 1994.
8. Pike KM, Loeb K, Walsh BT. Binge eating and purging. In: Allison DB,
ed. Handbook of Assessment Methods for Eating Behaviors and Weight-Related
Problems: Measures, Theory and Research. Thousand Oaks, CA: Sage
Publications, 1995:303–346.
9. Palmer R, Christie M, Cordle C, Davies D, et al. The Clinical Eating Disorder
Rating Instrument (CEDRI): a preliminary description. Int J Eat Disord 1987; 6:
9–16.
10. Fichter MM, Elton M, Engel K, Meyer A-E, et al. Structured Interview for Anorexia
and Bulimia Nervosa (SIAB): development of a new instrument for the assessment
of eating disorders. Int J Eat Disord 1991; 10:571–592.
38 CARTER ET AL.
11. Fairburn CG, Cooper Z. The Eating Disorder Examination. In: Fairburn CG,
Wilson GT, eds. Binge Eating: Nature, Assessment and Treatment. 12th ed. New
York: Guilford Press, 1993:317–360.
12. Mazure CM, Halmi KA, Sunday SR, Romano SJ, Einhorn AM. The YaleBrown-
Cornell Eating Disorder Scale: development, use, reliability and validation.
J Psychosom Res 1994; 28:425–445.
13. Sunday SR, Halmi KA, Einhorn A. The Yale-Brown-Cornell Eating Disorder Scale:
a new scale to assess eating disorder symptomatology. Int J Eat Disord 1995; 18:
237–245.
14. First MB, Gibbon M, Spitzer RL, Williams JB. User’s Guide for the Structured
Clinical Interview for DSM-IV Axis I Disorders. New York: Biometrics Research,
1996.
15. First MB, Gibbon M, Spitzer RL, Williams JB, Benjamin LS. User’s Guide for the
Structured Clinical Interview for DSM-IV Axis II Personality Disorders.
Washington, DC: American Psychiatric Press, 1997.
16. Wilson GT. Assessment of binge eating. In: Fairburn CG, Wilson GT, eds. Binge
Eating: Nature, Assessment and Treatment. New York: Guilford Press, 1993:
317–360.
17. Cooper Z, Cooper PJ, Fairburn CG. The validity of the Eating Disorder
Examination and its subscales. Br J Psychiatry 1989; 154:807–812.
18. Rosen JC, Vara L, Wendt S, Leitenberg H. Validity studies of the Eating Disorder
Examination. Int J Eat Disord 1990; 9:519–528.
19. Wilson GT, Smith D. Assessment of bulimia nervosa: an evaluation of the Eating
Disorder Examination. Int J Eat Disord 1989; 8:173–179.
20. Rizvi SL, Peterson CB, Crow SJ, Agras WS. Test-retest reliability of the Eating
Disorder Examination. Int J Eat Disord 2000; 28:311–316.
21. Bryant-Waugh RJ, Cooper PJ, Taylor CL, Lask BD. The use of the Eating Disorder
Examination with children: a pilot study. Int J Eat Disord 1996; 19: 391–397.
22. Herman CP, Polivy J. Restrained eating. In: Stunkard AJ, ed. Obesity.
Philadelphia: W.B. Saunders, 1980:208–225.
23. Stunkard AJ, Messick S. The three factor eating questionnaire to measure dietary
restraint, disinhibition and hunger. J Psychosom Res 1985; 29:71–81.
24. Gormally J, Black S, Daston S, Rardin D. The assessement of binge eating severity
among obese persons. Addict Behav 1982; 7:47–55.
25. Garner DM, Garfinkel PE. The Eating Attitudes Test: an index of the symptoms of
anorexia nervosa. Psychol Med 1979; 9:273–279.
26. Garner DM, Olmsted MP, Bohr Y, Garfinkel PE. The Eating Attitudes Test:
psychometric features and clinical correlates. Psychol Med 1982; 12:871–878.
27. Garner DM. Eating Disorder Inventory-2 professional manual. Odessa, FL:
Psychological Assessment Resources, 1991.
28. Maloney M, McGuire J, Daniels S. Reliability testing of a children’s version of the
Eating Attitudes Test. J Am Acad Child Adol Psychiatry 1988; 5:541–543.
29. Johnson-Sabine E, Wood K, Patton G, Mann A, Wakeling A. Abnormal eating

attitudes in London schoolgirls: a prospective epidemiological study. Psychol Med
1988; 18:615–622.
30. Meadows GN, Palmer RL, Newball EUM, Kenrick JMT. Eating attitudes and
disorder in young women: a general practice based survey. Psychol Med 1986; 16:
351–357.
31. Carter PI, Moss RA. Screening for anorexia and bulimia nervosa in a college
population: problems and limitations. Addict Behav 1984; 9:417–419.
32. Gross J, Rosen JC, Leitenberg H, Willmuth ME. Validity of the Eating Attitudes
Test and the Eating Disorder Inventory in bulimia nervosa. J Consult Clin Psychol
1986; 54:875–876.
33. Henderson M, Freeman CPL. A self-rating scale for bulimia: the “BITE.” Br J
Psychiatry 1987; 150:18–24.
34. Mizes JS. Personality characteristics of bulimics and non-eating disordered female
controls: a cognitive behavioral perspective. Int J Eat Disord 1988; 7:541–550.
35. Smith MC, Thelen MH. Development and validation of a test for bulimia.
J Consult Clin Psychol 1984; 52:863–872.
36. Williamson DA, Cubic BA, Gleaves DH. Equivalence of body image disturbances
in anorexia and bulimia nervosa. J Abnorm Psychol 1993; 102:1–4.
37. Smolak L, Levine MP. Psychometric properties of the Children’s Eating Attitudes
Test. Int J Eat Disord 1994; 16:275–282.
38. Garner DM, Olmsted MP, Polivy J. Development and validation of a
multidimensional Eating Disorder Inventory for anorexia nervosa and bulimia. Int J
Eat Disord 1983; 2:15–34.
39. Garner DM, Olmsted MP. The Eating Disorder Inventory Manual. Odessa, FL:
Psychological Assessment Resources, 1984.
40. Schoemaker C, Verbank M, Breteler R, van der Staak C. The discriminant validity
of the Eating Disorder Inventory-2. Br J Clin Psychol 1997; 36:627–629.
41. Wear RW, Pratz O. Test-retest reliability of the Eating Disorder Inventory. Int J Eat
Disord 1987; 6:767–769.
42. Childress AC, Jarrell MP, Brewerton TD. The Kids’ Eating Disorders Survey
(KEDS): Internal consistency, component analysis, and reliability. Eating
Disorders: J Treat Prev 1993; 1:123–133.
43. Childress AC, Brewerton TD, Hodges EL, Jarrell MP. The Kids’ Eating Disorders
Survey (KEDS): a study of middle school students. J Am Acad Child Adol
Psychiatry 1993; 32:843–850.
44. Brewerton TD. The use and scoring of the Kids’ Eating Disorders Survey (KEDS).
Eating Disorders: J Treat Prev 2001; 9:71–74.
45. Thelen MH, Farmer J, Wonderlich S, Smith M. A revision of the bulimia test: the
BULIT-R. Psychol Assess 1991; 3:119–124.
46. Welch G, Thompson L, Hall A. The BULIT-R: its reliability and clinical validity as
a screening tool for DSM-III-R bulimia nervosa in a female tertiary education
population. Int J Eat Disord 1993; 14:95–105.
40 CARTER ET AL.
47. Fairburn CG, Beglin SJ. Assessment of eating disorders: interview or self-report
questionnaire. Int J Eat Disord 1994; 16:363–370.
48. Luce KH, Crowther JH. The reliability of the Eating Disorder Examination—Self-
Report Questionnaire Version (EDE-Q). Int J Eat Disord 1999; 25:349–351.
49. Black CMD, Wilson GT. Assessment of eating disorders: Interview versus
questionnaire. Int J Eat Disord 1996; 20:43–50.
50. Carter JC, Aime A, Mills J. Assessment of bulimia nervosa: a comparison of
interview and self-report questionnaire methods. Int J Eat Disord 2001; 30:
187–192.
51. Grilo CM, Masheb RM, Wilson GT. A comparison of different methods for
assessing the features of eating disorders in patients with binge eating disorder.
52. Wilfley DE, Schwartz MB, Spurrel EB, Fairburn CG. Assessing the specific
psychopathology of binge eating disorder: interview or self-report? Behav Res Ther
1997; 35:1151–1159.
53. Wilson GT, Nonas CA, Rosenblum GD. Assessment of binge eating in obese
patients. Int J Eat Disord 1993; 8:173–179.
54. Carter JC, Stewart DA, Fairburn CG. Eating Disorder Examination Questionnaire:
norms for young adolescent girls. Behav Res Ther 2001; 39:625–632.
55. Mizes JS, Christiano B, Madison J, Post G, Seime R, Varnado P. Development of
the Mizes Anorectic Cognitions Questionnaire-Revised: psychometric properties
and factor structure in a large sample of eating disorders patients. Int J Eat Disord
2000; 28:415–421.
56. Mizes JS. Construct validity and factor stability of the Anorectic Cognitions
Questionnaire. Addict Behav 1991; 16:89–93.
57. Mizes JS. Validity of the Mizes Anorectic Cognitions Scale: a comparison between
anorectics, bulimics and psychiatric controls. Addict Behav 1992; 17:283–289.
58. Geller J, Johnson C, Madsen K. The role of shape and weight in self-concept: the
shape and weight based self-esteem inventory. Cog Ther Res 1997; 21:5–24.
59. Cooper Z, Fairburn CG. Confusion over the core psychopathology of bulimia
nervosa. Int J Eat Disord 1993; 13:385–389.
60. Fairburn CG, Peveler RC, Jones R, Hope RA, Doll HA. Predictors of 12month
outcome in bulimia nervosa and the influence of attitudes to shape and weight.
61. Geller J, Srikameswaran S, Cockell S, Zaitsoff Z. Assessment of shape- and weight-
based self-esteem in adolescents. Int J Eat Disord 2000; 28:339–345.
62. Geller J, Johnson C, Madsen K, Goldner EM, Remick RA, Birmingham CL. Shape
and weight-based self-esteem and the eating disorders. Int J Eat Disord 1998; 24:
285–298.
63. McConnaughy EA, DiClemente CC, Prochaska JO, Velicer WF. Stages of change
in psychotherapy: a follow-up report. Psychotherapy 1983; 26:494–503.
64. McConnaughy EA, Prochaska JO, Velicer WF. Stages of change in psychotherapy:
measurement and sample profiles. Psychotherapy 1983; 20:368–375.
65. Ward A, Troop N, Todd G, Treasure J. To change or not to change—“how” is the

question. Br J Med Psychol 1996; 69:139–146.
66. Rieger E, Touyz S, Schotte D, Beumont P, Russell J, Clarke S, Kohn M, Griffiths
R. Development of an instrument to assess readiness to recover in anorexia
67. Vitousek K, DeViva J, Slay J, Manke F. Concerns about change in the eating and
anxiety disorders. Paper presented at the Annual Meeting of the American
Psychological Association, New York, 1995.
68. Steinberg M. Structured Clinical Interview for DSM-IV Dissociative Disorders
(SCID-D). revised. Washington, DC: American Psychiatric Press, 1994.
69. First MB, Spitzer RL, Gibbon M, Williams JB, et al. The Structured Clinical
Interview for the DSM-III-R Personality Disorders (SCID-II): Part II. Multisite test-
retest reliability study. J Person Disord 1995; 9:92–104.
70. Spitzer RL, Williams JB, Gibbon M, First MB. The Structured Clinical Interview
for DSM-III-R: History, rationale and description. Arch Gen Psychiatry 1992; 49:
624–629.
71. Williams JB, Gibbon M, First M, Spitzer RL. The Structured Clinical Interview for
DSM-III-R (SCIDII): Multisite test-retest reliability. Arch Gen Psychiatry 1992; 49:
630–636.
72. Herjanic B, Reich W. Development of a structured psychiatric interview for
children: agreement between child and parent on individual symptoms. J Abnorm
Child Psychol 1982; 10:307–324.
73. Reich W. Dignostic interview for children and adolescents (DICA). J Am Acad
Child Adol Psychiatry 2000; 39:59–66.
74. De la Osa N, Ezpeleta L, Oomenech JM, Navarro JB, Losilla JM. Convergent and
discriminant validity of the structured diagnostic interview for children and
adolescents (DICA-R). Psychol Spain 1997; 1:37–44.
75. Welner Z, Reich W, Herjanic B, Jung KG, Amado H. Reliability, validity and
parent-child agreement studies of the diagnostic interview for children and
adolescents (DICA). J Am Acad Child Adol Psychiatry 1987; 5:649–653.
76. Goodman WK, Price LH, Rasmussen SA, Mazure C, Fleischmann RL, Hill CL,
Heninger GR, Charney DS. The Yale-Brown Obsessive Compulsive Scale, I.
Development, use, and reliability. Arch Gen Psychiatry 1989; 46:1006–1011.
77. Goodman WK, Price LH, Rasmussen SA, Mazure C, Delgado P, Heninger GR,
Charney DS. The Yale-Brown Obsessive Compulsive Scale, II. Validity. Arch Gen
Psychiatry 1989; 46:1012–1016.
78. Woody SR, Steketee G, Chambless DL. Reliability and validity of the YaleBrown
Obsessive-Compulsive Scale. Behav Res Ther 1995; 33:597–605.
79. Kim SW, Dysken MW, Kuskowski M. The Yale-Brown Obsessive-Compulsive
Scale: a reliability and validity study. Psychiatry Res 1990; 34:99–106.
80. Rosenfeld R, Dar R, Anderson D, Kobak KA, Greist JH. A computeradministered
version of the Yale-Brown Obsessive Compulsive Scale. Psychol Assess 1992; 4:
329–332.
42 CARTER ET AL.
81. Steketee G, Frost R, Bogart K. The Yale-Brown Obsessive Compulsive Scale:

interview versus self-report. Behav Res Ther 1996; 34:675–684.
82. Scahill L, Riddle MA, McSwiggin-Hardin M, Ort SI, King RA, Goodman WK,
Cicchetti D, Leckman JF. J Am Acad Child Adol Psychiatry 1997; 36:844–852.
83. Blake DD, Weathers FW, Nagy LM, Kaloupek DG, Klauminzer G, Charney DS,
Keane TM. A clinician rating scale for assessing current and lifetime PTSD:
theCAPS-1. Behav Therapist 1990; 13:187–188.
Disorders. 3d ed. revised. Washington, DC, 1987.
85. Weathers FW, Keane TW, Davidson JRT. Clinician-Administered PTSD Scale: a
review of the first ten years of research. Depression Anxiety 2001; 13: 132–156.
86. Blanchard EB, Hickling EJ, Taylor AE, Forneris CA, Loos W, Jaccard J. Effects of
varying scoring rules of the Clinician-Administered PTSD Scale (CAPS) for the
diagnosis of posttraumatic stress disorder in motor vehicle accident victims. Behav
Res Ther 1995; 33:471–475.
87. Beck AT, Steer RA, Brown GK. The Beck Depression Inventory Manual, 2d ed.
San Antonio, TX: Psychological Corporation Harcourt Brace & Company, 1996.
88. Lovibond SH, Lovibond PF. Manual for the Depression and Anxiety Stress Scales.
2d ed. Sydney, Australia: Psychological Foundation of Australia, 1995.
89. Lovibond SH, Lovibond PF. The structure of negative emotional states:
Comparison of the Depression Anxiety Stress Scales (DASS) with the Beck
Depression and Anxiety Inventories. Behav Res Ther 1995; 33:335–342.
90. Antony MM, Bieling PJ, Cox BJ, Enns MW, Swinson RP. Psychometric properties
of the 42-item and 21-item versions of the Depression Anxiety Stress Scales in
clinical groups and a community sample. Psychol Assess 1998; 10:176–181.
91. Beck AT, Epstein N, Brown G, Steer RA. An inventory for measuring clinical
anxiety: psychometric properties. J Consult Clin Psychol 1988; 56:893–897.
92. de Beurs E, Wilson KA, Chambless DL, Goldstein AJ, Feske U. Convergent and
divergent validity of the Beck Anxiety Inventory for patients with panic disorder
and agoraphobia. Depression Anxiety 1997; 6:140–146.
93. Marks IM, Mathews AM. Brief standard self-rating for phobic patients. Behav Res
Ther 1979; 17:263–267.
94. Shear MK, Brown TA, Sholomskas DE, Barlow DH, Gorman JM, Woods SW,
Cloitre M. Panic Disorder Severity Scale (PDSS). Pittsburgh: Department of
Psychiatry, University of Pittsburgh School of Medicine, 1992.
95. Meyer TJ, Miller ML, Metzger RL, Borkovec TD. Development and validation of
the Penn State Worry Questionnaire. Behav Res Ther 1990; 28:487–495.
96. Connor KM, Davidson JRT, Churchill LE, Sherwood A, Foa E, Wesler RH.
Psychometric properties of the Social Phobia Inventory (SPIN). Br J Psychiatry
2000; 176:379–386.
97. Antony MM, Orsillo SM, Roemer L. Practitioner’s Guide to Empirically Based
Measures of Anxiety. New York: Plenum Publishers, 2001.
98. Rosenberg M. Society and the Adolescent Self-image. Princeton, NJ: Princeton
University Press, 1965.
99. Rosenberg M. Conceiving the Self. New York: Basic Books, 1979.
100. Demo DH. The measurement of self-esteem: refining our methods. J Personality
Social Psychol 1985; 48:1490–1502.
101. Wylie RC. Measures of Self-concept. Lincoln: University of Nebraska Press, 1989.
102. Fairburn CG, Kirk J, O’Connor M, Anastasiades P, Cooper PJ. Prognostic factors
in bulimia nervosa. Br J Clin Psychol 1987; 26:223–224.
103. McFarlane T, McCabe RE, Jarry J, Olmsted MP, Polivy J. Weight-related and
shape-related self-evaluation in eating-disordered and non-eating-disordered
women. Int J Eat Disord 2001; 29:328–335.
104. Selzer ML. The Michigan Alcoholism Screening Test: the quest for a new
diagnostic instrument. Am J Psychiatry 1971; 127:1653–1658.
105. Brady JP, Foulks EF, Childress AR, Pertschuk M. The Michigan Alcoholism
Screening Tests as a survey instrument. J Oper Psychiatry 1982; 13:27–31.
106. Thurber S, Snow M, Lewis D, Hodgson JM. Item characteristics of the Michigan
Alcoholism Screening Test. J Clin Psychol 2001; 57:139–144.
107. Hedlund JL, Vieweg BW. The Michigan Alcoholism Screening Test (MAST): a
comprehensive review. J Oper Psychiatry 1984; 15:55–65.
108. Harrell AV, Wirtz PW. The Adolescent Drinking Index Professional Manual.
Odessa, FL: Psychological Assessment Resources, 1985.
109. Harrell AV, Wirtz PW. Screening for adolescent problem drinking: validation of a
multidimensional instrument for case identification. Psychol Assess 1989; 1:61–63.
110. Skinner HA. The Drug Abuse Screening Test. Addict Behav 1982; 7:363–371.
111. Gavin DR, Ross HE, Skinner HA. Diagnostic validity of the Drug Abuse Screening
Test in the assessment of DSM-III drug disorders. Br J Addict 1989; 84:301–307.
112. Martino S, Grilo CM, Fehon DC. Development of the Drug Abuse Screening Test
for Adolescents (DAST-A). Addict Behav 2000; 25:57–70.
113. Bernstein EM, Putnam FW. Development, reliability, and validity of a dissociation
scale. J Ner Men Dis 1986; 174:727–735.
114. Carlson EB, Putnam FW. An update of the Dissociative Experiences Scale.
Dissociation: Progr Dissoc Disord 1993; 6:16–27.
115. Dubester KA, Braun BG. Psychometric properties of the Dissociative Experiences
Scale. J Nerv Ment Dis 1995; 183:231–235.
116. van Ijzendoorn MH, Schuengel C. The measurement of dissociation in normal and
clinical populations: meta-analytic validation of the Dissociative Experiences Scale
(DES). Clin Psychol Rev 1996; 16:365–382.
117. Draijer N, Boon S. The validation of the Dissociative Experiences Scale against the
criterion of the SDIC-D, using receiver operating characteristics (ROC) analysis.
Dissociation: Progr Dissoc Disord 1993; 6:28–37.
118. Armstrong JG, Putnam FW, Carlson EB, Libero DZ, Smith SR. Development and
validation of a measure of adolescent dissociation: the Adolescent Dissociative
Experiences Scale. J Nerv Men Dis 1997; 185:491–497.
119. Farrington A, Waller G, Smerden J, Faupel AW. The Adolescent Dissociative
Experiences Scale: psychometric properties and difference in scores across age
groups. J Nerv Ment Dis 2001; 189:722–727.
44 CARTER ET AL.
120. Smith SR, Carlson EB. Reliability and validity of the Adolescent Dissociative
Experiences Scale. Dissociation: Progr Dissoc Disord 1996; 9:125–129.
121. Derogatis LR. Brief Symptom Inventory: Administration, Scoring and Procedures
Manual. 3d ed. Minneapolis: National Computer Systems, 1993.
122. Derogatis LR. Symptom Checklist 90-R: Administration, Scoring and Procedures
Manual. Minneapolis: National Computer Systems, 1994.
123. Croog SH, Levine S, Testa MA, Brown B, Bulpitt CJ, Jenkins CD, Klerman GL,
Williams CH. The effects of antihypertensive therapy on quality of life. N Engl J
Med 1986; 314:1657–1664.
124. Derogatis LR. BSI Bibliography. Minneapolis: National Computer Systems, 1993.
125. Foa EB, Riggs DS, Dancu CV, Rothbaum BO. Reliability and validity of a brief
instrument for assessing post-traumatic stress disorder. J Traum Stress 1993; 6:
459–473.
126. Coffey SF, Dansky BS, Falsetti SA, Saladin ME, Brady KT. Screening for PTSD in a
substance abuse sample: psychometric properties of a modified version of the PTSD
Symptom Scale Self-Report. J Traum Stress 1998; 11:393–399.
127. Horowitz LM, Alden LE, Wiggins JS, Pincus AL. Inventory of Interpersonal
Problems Manual. United States: The Psychological Corporation, 2000.
128. Horowitz LM, Rosenberg SE, Baer BA, Ureno G, Villasenor VS. Inventory of
Interpersonal Problems: psychometric properties and clinical applications.
129. Livesley WJ, Jackson DW, Schroeder ML. Dimensions of personality pathology.
Can J Psychiatry 1991; 36:557–562.
130. Clark LA, Livesley WJ, Schroeder ML, Irish S. The structure of maladaptive
personality traits: convergent validity between two systems. Psychol Assess 1996;
8:294–303.
131. Schroeder ML, Wormworth JA, Livesley WJ. Dimensions of personality disorder
and their relationship to the big five dimensions of personality. Psychol Assess 1992;
4:47–53.
132. Goldner EM, Srikameswaran S, Schhroeder ML, Livesley WJ, Birmingham CL.
Dimensional assessment of personality pathology in patients with eating disorders.
Psychiatry Res 1999; 85:151–159.
133. Clark LA, Harrison JA. Assessment Instruments. In: Livesley, WJ ed. Handbook of
Personality Disorders. New York: Guilford Press, 2001.
134. Eisler I, Dare C, Russell GFM, Szmukler G, le Grange D, Dodge E. Family
and individual therapy in anorexia nervosa: A 5-year follow-up. Arch Gen
Psychiatry 1997; 54:1025–1030.
135. Skinner HA. Self-report instruments for family assessment. In: Jacob T ed. Family
Interaction and Psychopathology. New York: Plenum Publishers Co, 1987.
136. Moos R, Moos B. Family Environment Scale Manual. Palo Alto, CA: Consulting
Psychologists Press, 1981.
137. Grotevant HD, Carlson CI. Family Assessment: A Guide to Methods and
Measures. New York: Guilford Press, 1989.
138. Skinner HA, Steinhauer PD, Santa-Barbara J. Family Assessment Measure-III.

Toronto: Multi Health Systems, 1995.
139. Skinner HA, Steinhauer PD, Sitarenios G. Family Assessment Measure (FAM) and
process model of family functioning. J Fam Ther 2000; 22:190–210.
140. Forman B. Assessing perceived patterns of behavior exchange in relationships.
J Clin Psychol 1988; 44:972–981.
141. Skinner HA, Steinhauer PD, Santa-Barbara J. The Family Assessment Measure.
Can J Comm Men Health 1983; 2:91–105.
142. Bloom BL. A factor analysis of self-report measures of family functioning. Fam
Proc 1985; 24:225–239.
143. Jacob T. The role of the time frame in the assessment of family functioning.
J Marital Fam Ther 1995; 21:281–286.
46 CARTER ET AL.
3
Feeding Disorders in Infancy and Early
Childhood
Dasha Nicholls
Great Ormond Street Hospital
London, England
This chapter examines the range of feeding problems from birth to early
childhood from a theoretical and clinical perspective. Feeding problems are
usually understood as being part of a relationship and never simply located
within the child. Feeding problems are usually defined as starting before the age
of 6 years and are rarely, if ever, defined in terms of subjective anxieties of the
child but usually in terms of observed behavior or adults’ descriptions. A
feeding difficulty becomes a problem when it is associated with additional
developmental, medical, or emotional problems. The boundary between a
feeding problem, often parent defined, and a disorder, often professionally
defined, is indistinct. Most cases presenting to a hospital setting have a
multifactorial cause and maintenance involving medical, developmental, social,
and psychological factors; and in these cases it is unlikely that the children will
simply grow out of the problem without help. The difficulty of negotiating
issues of independence and control with their caregivers can result in conflict,
and these problems are often labeled as behavioral. This chapter will give an
overview of issues relating to diagnosis and classification, control and
responsibility, and assessment, with reference to texts from which further
information can be sought.
RECOGNITION AND CLASSIFICATION

DSM-IV distinguishes three feeding disorders: two specific subtypes—pica and
rumination—and a broader diagnostic category of “feeding disorder of infancy
and early childhood.” Each will be considered in turn.
Pica
Pica describes the persistent eating of nonnutritive substances over an extended
period of time (more than a month) (see Table 1 for diagnostic criteria). It is
usually only diagnosed separately if of such severity that specific intervention is
48 NICHOLLS
TABLE 1 Diagnostic Criteria for Pica
required, such as when medical complications occur as a result of toxicity. The

type of substance eaten tends to vary with age and developmental capacity of the
child. For example, younger children may eat paint, string, or hair; older
children leaves and pebbles; and adolescents clay or soil. Pica is associated with
developmental disorders, including learning disability and pervasive
developmental disorders, where its severity is related to the degree of disability.
Poverty, neglect, and lack of supervision increase the risk for pica in vulnerable
individuals. Epidemiological data on pica are limited, and the ingestion of
nonnutritive substances may be quite common in preschool children and should
not be considered abnormal unless persistent and severe. In a sample of
472 children attending pediatric clinics, Stein et al. found that pica was
associated with parasomnias such as sleepwalking, nightmares, night terrors, and
head banging or rocking in sleep (1). Case reports have suggested a role for
selective serotonin reuptake inhibitors in the management of pica (2).
Rumination
Rumination is a syndrome characterized by the effortless regurgitation of
recently ingested food (see Table 2 for diagnostic criteria). It has been linked to
severe medical and psychosocial conditions, including malnutrition, aspiration
pneumonia, and complete social withdrawal, and is seen in three distinct
populations: infants; individuals with psychiatric and neurological disorders,
particularly developmental disabilities; and adults who do not have overt
psychiatric or neurological disorders (3). The hallmark of rumination, which
separates it from other disorders of the upper gastrointestinal tract (such as
gastroesophageal reflux disease), is that in rumination the stomach contents
appear in the mouth without retching or nausea. The subject appears to make a
conscious decision on how to handle the regurgitated food, and the experience
seems to be pleasurable. It usually occurs very soon after a meal and tends to
last for 1–2 hours.
FEEDING DISORDERS IN INFANCY AND EARLY CHILDHOOD 49
TABLE 2 Diagnostic Criteria for Rumination Disorder
Rumination is relatively rare but potentially fatal when it occurs in infants. It

is thought to be a form of self-stimulation in this age group. Observation of
rumination is all that is needed to make the diagnosis, although often rumination
ceases as soon as the infant notices the observer. Parents may not report the
symptom spontaneously but may recognize it when described. The infant who
ruminates may not retain enough nutrients and may develop potentially lethal
malnutrition, a complication that seldom, if ever, occurs in older ruminators.
Sensory and/or emotional deprivation are both associated with rumination in
children, which may explain the increased incidence of rumination in
institutionalized children, infants in intensive care units, and normal infants with
attachment disorders. In infant rumination, the aim of treatment is to provide a
nurturing environment and comforting care to the infant, and to help the
mother improve her ability to recognize and respond sensitively to her infant’s
physical and emotional needs. In mentally handicapped children, providing a
nurturing caregiver may not be sufficient and behavioral therapy may be
necessary.
In older children rumination can be associated with weight loss and
vomiting (4), and treatment includes nutritional support in combination with
medication, cognitive and relaxation techniques, and pain management. Cases
may present to gastroenterologists rather than to mental health practitioners. A
collaborative approach can be very effective, e.g., combining the use of
medication to reduce acid damage to the esophagus, with psychological therapy
aimed at identifying situations and emotions that trigger the symptoms. A
multidisciplinary team approach is associated with satisfactory recovery in most
patients.
Feeding Disorder of Infancy and Early Childhood

These two specific feeding disorders aside, there remain a number of feeding
and eating behaviors in childhood and early adolescence that are currently
50 NICHOLLS
TABLE 3 Diagnostic Criteria for Feeding Disorder of Infancy or Early Childhood
classified under the generic diagnosis of “feeding disorders of infancy and early
childhood” (see Table 3 for diagnostic criteria). The classification of feeding
disorders has been hampered by a lack of knowledge about feeding behaviors in
healthy, typically developing children, thus making the boundaries between
normal, problem, and disordered feeding hard to identify. In addition, reported
feeding problems are common. More than 50% of parents report one problem
feeding behavior and more than 20% report multiple problems (5) in children
aged 9 months to 7 years. Trying to get children to eat food during structured
mealtimes appears to provide the most tension for parents. Problem feeding is
more likely to be reported among the children of parents using strategies such
as coaxing, threatening, making multiple meals, and force feeding (5). In this
study, psychosocial variables such as marital status, socioeconomic status, or the
child’s birth order were not significantly associated with reported frequency and
number of feeding problems.
Bax [quoted in (6)] has suggested that the fact that feeding and feeding
problems are covered so extensively in the child rearing literature in part
explains the relative neglect of scientific research in the area. For example,
information about typical length of mealtimes in infants and toddlers has only
recently been established, from which a definition of slow eating was
derived (7). In this study more than 30 minutes was considered slow. The
relative paucity of knowledge about the range of normal feeding behaviors has
led to an overreliance on parental report of feeding as problematic in order to
define disorder. An exception is the work of Dahl and colleagues in Sweden,
who studied a sample of infants aged 3–12 months (8) and published follow-up
findings up to primary school age. Problem feeding was defined on the basis of
parent and nurse reports, had to be present for more than 1 month, and had to
have responded to medical and psychological advice and treatment. She
identified 50 infants fulfilling these criteria (a prevalence of 1.4%). The majority
(82%) were underweight for their age. Three main problem categories were
distinguished: refusal to eat (56%), colic (18%), and vomiting (16%), with 10%
miscellaneous others.
Feeding Disorders Presenting to Clinical Services

A number of different approaches have been taken in attempts to classify feeding
disorders based on children presenting to clinical services. There is no
international consensus on how to name these feeding problems or how they
should be classified. Furthermore, although there is an extensive literature
relating to feeding problems of infancy and early childhood (usually preschool),
there is a very little published work about feeding and eating problems
presenting in middle childhood.
The way in which feeding disorders are categorized tends to depend on the
professional context in which they have presented, and the terms used are those
that have been found to be meaningful in a clinical context. For example, taking
a medical perspective, Burklow (9) proposed a coding system for complex
feeding disorders based on combinations of five categories: structural
abnormalities, neurological conditions, behavioral issues, cardiorespiratory
problems, and metabolic dysfunction. In this scheme behavioral issues were
defined very broadly to include feeding difficulties arising from psychosocial
difficulties, negative feeding behavior shaped and maintained by reinforcement,
and/or emotional difficulties (e.g., phobias, depression). Using this system,
85% of cases were classified in the behavioral category, suggesting that a
biobehavioral conceptualization may be more appropriate. Skuse and Wolke (6)
identify the main feeding problems as colic; gastroesophageal reflux; oral-motor
dysfunction in high-risk neonates; tactile hypersensitivity; oral-motor
dysfunction and failure to thrive; and psychodynamic feeding problems. For a
comprehensive list of possible causes of feeding problems, see (10).
Speech and language therapists tend to consider the presence or absence of
oromotor dysfunction a key factor (11), grouping feeding problems according
to dysfunctional or disorganized feeding (12). Dysfunctional feeding patterns
tend not to resolve over time and are associated with other neurological
diagnoses and motor difficulties. Disorganized feeding either resolves over time
or is related to subsequent sensory problems. It can be helpful to think
separately about motor and sensory problems, although inevitably the two
interact.
There are a number of psychological classifications, and these depend on
particular approaches and whether the child, the parent, or the child-parent
interaction is the basis for defining the problem. A symptom-based approach to
classification can be helpful in behavioral management, e.g., problems with
texture, quantity, and range (13). Harris et al. (14) take an even more
52 NICHOLLS
pragmatic approach, simply identifying two types of food refusal that may occur
separately or together, i.e., refusal to take in sufficient calories and refusal to
ingest a sufficient range of foods.
From a psychodynamic perspective, Chatoor and colleagues have described a
number of feeding problems based on attachment-separation theory, the best
characterized of which are infantile anorexia (15) and posttraumatic feeding
disorder of infancy (PTFD) (16). This approach emphasizes the difficulties
infants have distinguishing somatic sensations such as hunger and satiety from
emotional feelings such as sadness, anger, and frustration. The term infantile
anorexia derives from Bruch’s early descriptions of patients with anorexia
nervosa, and relates to the interpersonal nature of the infant’s struggle to have his
or her needs met. Infant-mother relationships in this group of infants have been
characterized by a lack of reciprocity, conflict, and a struggle for control in
which food refusal is central (see below). In PTFD, children are distinguished by
showing increased resistance during feeding interactions and a marked resistance
to swallowing food. This may be an infantile precursor to what had been termed
“functional dysphagia” (17), in which children identify a specific fear of
swallowing, often associated with a history of choking.
Finally, attempts have been made to classify parental responses to feeding in
recognition of the role that managing the “balance of power” (18) has in feeding
problems. Birch and colleagues identified three types of parenting styles
applicable to feeding: highly controlling, laissez-faire, and responsive. Highly
controlling and laissez-faire parenting may interfere with self-regulation of
children’s feeding behaviors. Perhaps most importantly in this context is the
association between highly controlling parenting and maternal eating
disorders (19), and the link to feeding problems in their offspring (20,21).
Maternal feeding practices and perceptions of their child’s eating have also been
linked subsequent overweight (22).
One study of classification is particularly noteworthy in terms of its research
methodology. Crist et al. (5), used a feeding screening questionnaire, the
Behavioral Pediatrics Feeding Assessment Scale (BPFAS), to empirically derive
subtypes of feeding problem from a sample of 96 control and 249 clinically
referred subjects. Using a principal-components analysis, they identified the
following factors, which accounted for 55% of the total variance for the
combined clinical and normative groups.
Factor 1: Picky Eaters

This factor essentially represents the willingness of children to try new foods
and the variety of food groups that the child accepts. This continuum of feeding
problems runs from the “picky eating” pattern, common during the toddler
years, to the more severe end of the spectrum where children have narrowed
their food selection to the extent that they are consuming insufficient amounts of
key vitamins and minerals. Reau et al. (7) found an association between picky
eating and length of meal times and speculated that lengthened meal times
might reflect underlying oral-motor dysfunction in some cases. Recently Jacobi
et al. (23) aimed to clarify and validate the nature of picky eating by examining
the relationship between parental report of picky eating and objective
behavioral measures, and also looked at both parent and child precursors of
picky eating. This study is important because it attempts to clarify the relationship
between parent-reported problem feeding and observed disorder. Of 29 cases
classified as “picky” on the basis of parent report, objective measures confirmed
a lower number and variety of foods consumed, predominantly the avoidance of
vegetables. These authors found no association between picky eating and slow
eating. Picky eating, sometimes to quite an extreme degree, is commonly found
in children with autistic spectrum disorders. This type of feeding pattern has
also been termed selective eating (24) or perseverative feeding disorder (14), both of
which convey the extreme selectivity in preferred foods and resistance to trying
new foods [food neophobia (25)]. Overall caloric intake is often adequate for
these children, and growth and development are usually normal (26).
Rydell et al. (27) found that at primary school age choosiness was not related to
gender, social class, or ethnic background. The choosy children had modestly
elevated levels of externalizing, hyperactive, and internalizing behavior. More
pronounced choosiness was found in those with a history of refusal to eat in
infancy or at preschool age. These children had more problem behaviors in
other areas than less choosy children.
Factor 2: Toddler Refusal—General

The behaviors identified with this factor included whining or crying, tantrums,
and spitting out food. This refusal pattern was associated with younger children
and appeared to be more general in nature as opposed to food specific, and may
be linked to other types of oppositional behavior. Parent training approaches
may be the most fruitful.
Factor 3: Toddler Refusal—Textured Foods

The behaviors that loaded on this factor included problems chewing food; eating
only ground, soft food; letting food sit in the mouth without swallowing it; and
choking or gagging. This type of refusal behavior appeared to relate closely to
making the transition from soft to chewy foods, and appeared to reflect the
individual child’s difficulty in handling textured foods or the selective refusal of
54 NICHOLLS
textured foods rather than a general disruption of mealtime behavior. Clinical

experience suggests some overlap between this factor and picky eating, and
where the two co-occur the picky eating may be more persistent. Problems
with textured food are also found in children with neurodevelopmental
disorders, such as autism. Facial and oral hypersensitivity can be found along
with neurological difficulties such as epilepsy, and is also associated with
situations where oral desensitization has failed to occur. The most commonly
encountered instance of this is in children who have been tube fed. Normally
children experience a wide range of oral stimulation, at various phases putting
almost anything and everything in their mouths. If a child has had no experience
of the taste, temperature, touch, or smell of food for an extended period, the
reintroduction of oral feeding may not be pleasurable. To prevent feeding
problems from developing while on enteral feeding, children should be
encouraged to maintain feeding skills, engage in textured messy/food play, and
maintain patterns of feeding and social aspects of feeding, e.g., sitting at the
table for regular meals with others.
Factor 4: Older Children Refusal—General

This factor included behaviors associated with older children, such as delaying
eating by talking, trying to negotiate what food he or she will eat, getting up
from the table during meals, and refusing to eat much during a meal but requesting
food immediately after the meal. Much of the nutritional intake of these
children was gained through snacking between meals. Although these behaviors
lengthen the mealtime, they would seem to reflect general disruptive behavior,
rather than possible oral-motor difficulties.
Factor 5: Stallers
This factor was not as well defined as others. The authors assigned the term
“stallers” because the feature of “letting food sit in the mouth without
swallowing it” was common to all cases. It was associated with a preference for
fluids over food (e.g., would rather drink than eat; drinks milk). This feeding
pattern may be similar to restrictive eating, a term that Bryant-Waugh and
Lask (28) have used to describe children with a constitutionally small appetite,
who show limited interest in food, don’t eat much, and who grow and develop
normally in the lower centiles for weight and height. This type of feeding
pattern could therefore be considered a normal variant. Clinical presentation is
often due to anxiety about growth, and there may be a long history of attempts
to feed the child more that he or she can manage. Indeed, feeding practices such
as coercion to eat, excessive anxiety about weight gain, and conflict over food
can precipitate food refusal, vomiting, and failure to thrive.
None of the five groups outlined above would seem to identify a single factor
reflecting the more serious condition of failure to thrive, which describes a
pattern of faltering development as a result of poor weight gain. It may be that
failure to thrive can result from any of the feeding problems described above, if
sufficiently severe, and/or other factors contribute to exacerbation of feeding
problems, such as the use of coercion or other negative parent-child
interactions. Although malnutrition is the end point, disentangling the complex
contributions of child, parent, and their interactions is not always
straightforward. For example, the dichotomy of organic versus nonorganic
failure to thrive is used less often since the contribution of subtle differences in
sucking, chewing, and swallowing were noted in a group of children identified
as having nonorganic failure to thrive (29). Nonetheless, failure to thrive can
be, but is not always, associated with other evidence of neglect and deprivation.
This way of grouping feeding problems is not dissimilar to other attempts at
classification but has the advantage of being empirically derived. However,
there are a number of limitations. First, only 55% of the variance was accounted
for by these five factors. Second, larger sample sizes would be needed to
determine whether any of these patterns of feeding behavior were more
common for specific medical conditions than for others. Perhaps surprisingly,
only one prospective study has examined the role of early feeding problems in
the development of subsequent onset of eating disorders (30), although feeding
problems have been identified as a risk factor from retrospective studies (31,32)
and there is increasing interest in developmental precursors of subsequent
problem eating. In Marchi and Cohen’s study, 659 children and their mothers
were interviewed three times between 1 and 21 years of age. Picky eating in early
childhood was found to predict symptoms of anorexia nervosa in later
adolescence (30). The authors defined picky eating by the presence of three of
the following (maternally reported) behaviors of the child: “does not eat enough,”
“is often or very often choosy about food,” “usually eats slowly,” “is usually not
interested in food.” This concept of picky eating is therefore broader than that
defined above, in which only choosiness is a feature. Disinterest in food is a
characteristic most often associated with restrictive eating, classifiable in the
“stallers” group above.
What evidence there is suggests that the nature of feeding problems found in
children with and without medical conditions do not differ signifi cantly in type
but only in frequency or intensity (5,33). For a review of diagnosis and
treatment of feeding difficulties in children with developmental difficulties see
Schwarz et al. (34), and for feeding difficulties associated with illness see Harris
et al. (35).
56 NICHOLLS
Enteral Feeding
Increasingly children presenting to clinical services have been started on enteral
feeding for any of a number of reasons and are having difficulty making the
transition from tube feeding to oral feeding. Enteral feeding is a passive
experience for the child. Problems created by enteral feeding include reduction
of appetite and thirst leading to lack of recognition of hunger and thirst drive;
absence of normal response to the sight or smell of food, e.g., salivation, lack of
pleasure; reduction of opportunities to practice routine and skills required for
oral feeding, such as the use of cutlery; and fear and revulsion at certain
textures, tastes, and smells. Each of these difficulties may be present to a
greater or lesser degree. It is unrealistic to expect a child to start oral feeding
simply by withdrawing enteral feeds. Psychological approaches may consider a
program of desensitization to anxiety-provoking textures and foods, as well as
the establishment of normal mealtime behaviors in order to expose the child to
food and to encourage feeding. Play with food must be modeled by caregivers in
a relaxed, nonpressurizing manner. Self-feeding by mouthing, kissing, and
licking food should be encouraged. Reduction of enteral feeds should be
attempted once feeding skills have been established and a small but measured
amount of food or drink is taken by mouth on a daily basis. A window of hunger
and thirst may be created by tube feeding the child overnight or spacing bolus
feeds. Enteral feeds should be reduced slowly and gradually under the
supervision of a dietitian, in conjunction with the child’s medical care and with
regular weight checks. To prevent feeding problems developing while on
enteral feeding, children should be encouraged to maintain feeding skills,
engage in textured/messy food play, and maintain patterns of feeding and social
aspects of feeding, such as sitting at the table for regular mealtimes with others.
FROM FEEDING TO EATING DISORDERS: ISSUES

OF CONTROL AND SELF-REGULATION
The term “feeding disorder” suggests an interactional component between the
caregiver and child, whereas “eating disorder” implies autonomy of
selfregulation and care. Typically developing children vary enormously in the
degree of dependence they exhibit around food, and this becomes particularly
emphasized in the context of illness or disability, when the capacity of the child
to follow normal trajectories toward self-regulation are delayed or deviated.
Thus, a preschool child might be described as “eating” if he or she shows an
ability to select food appropriate to the energy needs, respond to hunger and
satiety, and exhibit socially adaptive eating behavior. By contrast, a 12-year-old
girl with anorexia nervosa may, during the course of her illness, become fully
dependent on her parents to choose appropriate food, determine her nutritional

needs, and even feed her. In practice, therefore, the age at which feeding ends
and eating begins is far from clear. Despite this, feeding problems are
traditionally considered separately from eating problems and eating disorders.
One consequence of this separation is the relatively limited examination to date
of the continuities and discontinuities between the two. Recently, there has
been an increase in interest in the association between the two, opening new
avenues of investigation toward a developmental understanding of problem
eating in later life.
Understanding the transition from feeding to eating requires a consideration
of how control and responsibility around food intake are determined and
negotiated in both a developmental and a systemic context. This process of
transfer of responsibility for eating from caregiver to child is a careful balance of
timing and encouragement—too much parental regulation and the child may
rebel; too much autonomy for the child and he or she may not be able to cope.
As such the transition from feeding to eating is highly susceptible to tension and
conflict, particularly over issues of autonomy and control. It is also a point of
communication between a child and a parent, and can be a means for
communicating distress or anxiety.
Thus, learning to manage control over self and others is an important
developmental task. The dynamic nature of who “holds” control and
responsibility in a wide range of different areas forms a major part of the
complex, changing relationship between parents and children. For example, our
views and expectations of the appropriateness and ability of children to accept
responsibility for themselves, and of the nature and degree of the control their
parents might exert, vary widely according to developmental and individual
considerations. At each developmental stage food can represent a vehicle for the
exploration and regulation of basic needs. For example, in infancy food is the
main medium through which a child can seek reassurance that his or her needs
will be met reliably and predictably. In the toddler years, characterized by the
learning of new skills and the making of first choices, the child is seeking
mastery, with help at hand for disasters. The responses of parents, caretakers,
and other significant others have an impact on the child’s developmental
progress and ability to successfully manage conflicts, dilemmas, and tasks at each
stage. For example, the caregiver has to tolerate the frustration and mess of the
toddler’s failed attempts at mastery. The individual child’s personal style is in
part understood against the backdrop of family members’ responses and
communication, which is in turn considered against its own wider social and
cultural background.
Control is a central theme in feeding and eating disorders, and can be
considered from many different perspectives, e.g., within individuals, within
58 NICHOLLS
relationships, culturally, systemically, developmentally, and across generations.

Birch and Fisher have emphasized the role of parental attempts to control and
restrict access to foods in the development of food preferences (36). Not only
can parental control of food intake be counterproductive in terms of enhancing
rather than reducing preferences for “bad” foods, but it can also, if habitual,
influence the child’s capacity for self-regulation of food intake. “Child feeding
practices that encourage or restrict children’s consumption of foods may
decrease the extent to which children use their internal signals of satiety and
hunger as a basis for adjusting energy intake” (36, p. 544).
Much of Birch and colleagues work has focused on the development of
childhood overweight, but we know too that parental eating disorders may be
associated with a wish for control, which in turn can impact on child
development. Stein et al. (19) looked at whether parental (in this case maternal)
control is specific to eating disorders and whether attempts at control with the
child are specific to feeding behavior. They found that mothers with eating
disorders were more controlling, both verbally and in actions, than mothers
with depression or control mothers, and this was related to dietary restraint.
However, rather than maternal attempts at control being restricted to mealtimes,
they found that control was most likely to be exercised during the child’s play.
The significance of this finding is the impact of controlling parental behavior on
child development outside the narrow domain of feeding and eating. The
specificity of parental overcontrol to eating disorder subjects is of particular
note. In Fairburn and colleagues’ cognitive behavioral theory of the maintenance
of anorexia nervosa, the need for control of eating is a central theme. They argue
“that an extreme need to control eating is the central feature of the disorder,
and that in Western societies a tendency to judge self-worth in terms of shape
and weight is superimposed on this need for self control” (Fairburn et al.,
1999).
The number of systematic controlled studies in this important area is
relatively small (19,37–39). The available evidence suggests that children of
mothers with eating disorders are at increased risk of disturbance, but that the
risk is not invariable and depends on a variety of factors. Further research is
needed to understand how the need for control is internalized, as well as the
role of early experiences of control and self-regulation. In particular, a number
of studies have suggested that some aspects of parental influence on eating
behaviors are gender specific. Jacobi et al. (23) found the impact of maternal
eating disorders and disturbances to be much stronger than that of fathers
and that it was specifically directed at their daughters. Maternal restraint was
predictive of worries about weight in 8-year-old daughters, and high maternal
BMI was predictive of weight control behaviors in daughters but not sons.
Although mothers with eating disorders may be at the extreme end of the
spectrum, the link between parental eating concern and reported childhood
feeding problem is evident from a number of studies as well as from clinical
experience. For example, in a community sample of 397 children whose
parents were questioned when their child was 13 months and 5 years of age,
problematical or maladaptive eating habits of the children were found to be
connected to those of their parents (40). The mother’s poor ability to enjoy
eating and high tendency to snack regardless of hunger, as well as the father’s
difficulty in weight regulation, significantly predicted persistent poor eating in
their children.
While there is increasing recognition of the association between parental
eating behavior and child eating behavior, far less is known about the longterm
course of feeding disorders and later problems both in eating and in other
aspects of control and self-regulation. We are increasingly aware from clinical
experience that a number of feeding problems can persist into later life.
Examples are children who are tube fed, in whom increased survival may also
be a factor in some patient groups; children with selective (picky) eating, which
can persist into adolescence (26) and even adulthood; and food phobias.
The mechanism by which feeding problems may be linked to eating disorders
or other types of eating problem in adolescence or adulthood is at this time
somewhat speculative. Issues of control and of the development of specific
cognitions relating to beliefs about self and others may play a part. Our capacity
to distinguish eating disorder cognitions in younger children may enhance our
understanding in this area. True eating disorders can be understood as disorders
characterized by grossly disordered or chaotic eating behavior associated with
morbid preoccupation with body weight and shape. In children as young as 7,
true eating disorders have been described and can be distinguished from
children of comparable age with other types of food avoidance. For these
disorders, anorexia nervosa and bulimia nervosa, the overall clinical
presentation is similar to that in adults, with some important distinctions that
reflect developmental and gender-based differences in expression rather than
differences in the disorder per se (41). Using the childhood version of the Eating
Disorders Examination (EDE) (42,43), Cooper et al. (38) have shown that
early-onset patients with anorexia nervosa had comparable scores to later-onset
patients on all subscales apart from Eating Concern, but that AN patients were
clearly distinguishable from those with other types of eating problems. It is
these other types of eating problems presenting in childhood, not classifiable as
eating disorders, that are of un certain nosological status (44). In particular, a
few of the subjects in the Cooper et al. study were difficult to classify and what
at first appeared to be an atypical feeding/eating problem later become more
characteristic of anorexia nervosa. A few may be diagnosed as “eating disorder
60 NICHOLLS
not otherwise specified,” and a proportion whose food avoidance dates back to
younger than 6 years would meet criteria for “feeding disorder of infancy and
early childhood.” There remain a number of children who present clinically
with disordered eating who do not fit into any of the current classification
systems. This heterogeneous group of patients need further systematic
investigation. Food avoidance emotional disorder (45), psychogenic vomiting,
food phobias, and functional dysphagia are just some of the terms that have been
used to describe problem eating behaviors in this age group (41). We have
suggested that these can be usefully thought about in relation to associated
psychopathology. For example, children with food avoidance emotional
disorder or psychogenic vomiting often have other medically unexplained
symptoms, and their parents may attribute weight loss to an undiagnosed
physical disorder. As such, classification as a somatization disorder may be
appropriate. Some children develop specific circumscribed fears in relation to
food, which are associated with specific cognitions. Generalized anxiety,
unrelated to food, may also be present. In these cases anxiety management
techniques in combination with family work can be effective once nutritional
intake has been reestablished, and hence classification as anxiety disorder, a
specific phobia, or obsessive-compulsive disorder may be appropriate depending
on clinical presentation. What each of these problems has in common with
feeding and eating disorders is a need to specifically address the nutritional as
well as the emotional needs of the child.
ASSESSMENT OF FEEDING PROBLEMS

All eating and feeding problems have biological, psychological, and social
contributing factors. Assessment is one of the most important aspects of any
approach to management and can in itself be a powerful intervention. It should
be based on a theoretical or conceptual model, and encompass biological,
psychological, behavioral, and social components. As already mentioned, both a
developmental and a systemic understanding should be emphasized. Thus, the
level of understanding will change if the child only, the child and mother, the child
and parents, or the whole family are seen. In addition, a cultural understanding
may be needed if the significance of the problem is to be understood. This is
true of basic parameters such as the interpretation of height and weight data,
and becomes even more complex when it comes to understanding issues of
control and responsibility. Any assessment protocol will need to be
comprehensive enough to gather relevant information to address the range of
feeding problems thus far described and is likely to require the skills of a
multidisciplinary team and a multidimensional approach. Assessment aims to
afdsf
FIGURE 1 Conceptual model for the maintenance of “vicious cycles” within feeding
problems. Assessment aims to identify the relative contributions of physical, emotional,
behavioral, and cognitive contributions to the maintenance of the feeding problem within
a developmental and systemic framework.
identify the relative contributions of physical, emotional, behavioral, and

cognitive contributions to the maintenance of the feeding problem in a
developmental and systemic framework (Fig. 1).
History and Development of the Feeding Problem

A thorough history should be gathered from the parents/caregivers. A good
starting point is usually the time when someone first became aware that there
might be a problem, although it is helpful to then go back and review feeding
from birth. This is often done by means of a semistructured questionnaire
administered to parents, of which a number have been published (5,14,46).
Finally the child’s current general functioning should also be assessed (including
school and social functioning). From the history factors that may have been
associated with onset can be identified, such as medical treatment; a traumatic
incident such as choking, loss events, or moves; a developmental problem; a
medical condition; beliefs about feeding and parental roles; or any combination
of these. Factors associated with onset can then be usefully differentiated from
factors that appear to maintain the problem, such as continued medical
problems/treatment causing discomfort or loss of appetite, ongoing emotional
problems such as fear of choking again, and factors such as heightened anxiety at
mealtimes or poor parental management/mealtime conflict.
62 NICHOLLS
Physical Assessment
In addition to aiding diagnosis and management decisions, information obtained
from physical assessments can be a powerful tool for both sufferers and parents/
caregivers. Furthermore, medical concerns are often the main cause for concern
or the reason that the problem has been taken seriously. The commonest causes
for concern are significant weight loss or weight gain, failure to gain weight or
height appropriately for age and stage of development, or a single low or high
BMI or other weight-to-height ratio index measurement (outside 3rd or 97th
centile for age). The potentially irreversible effects of malnutrition on physical,
psychological, and emotional growth and development in infants and children
are well described. Monitoring of growth and development should continue until
the child is growing and developing along appropriate centiles. Weight and
height charts are of more value for monitoring change than for assessing the
significance of a single measurement. More importantly, growth charts
emphasize the rate of expected weight gain for a child. Over the past few years
many countries have published BMI centiles for children, e.g., (47–49), which
will enable nutritional status to be more conveniently and routinely assessed.
It is important that the presence of unidentified and untreated or partially
treated physical symptoms, such as esophageal reflux or problems with
swallowing as a result of oromotor dyscontrol, be assessed.
Assessment of Developmental Issues

We have found it useful to take a personal history including important events in
the child’s life. This can be drawn on a time line that can then be placed next to
the one with weight and eating history, enabling potential links to be explored
more easily. The personal history should also include a developmental history,
including early feeding, milestones, the child’s medical history, plus any past
history of emotional and/or behavioral problems.
Assessment of Systemic Context

Assessment of systemic context involves gathering information about the
family, its wider context, and the family’s past contact with health
professionals. In addition to gathering information about family history of weight
and shape issues as well as family medical and psychiatric history, the aim is to
identify strengths, resources, and sources of support. This can be done through
drawing a family tree involving any members of the family who live at home. It
is important to gain some impression of the family’s social context, their ethnic
background, and any relevant associated beliefs and practices. With regard to
the eating difficulty, we would routinely ask all members of the family how they
understand it and how it affects them. Finally, we ask parents what they have
tried in terms of managing the problem and who or what they have found
helpful in the past. This will also allow a picture of help seeking and professional
networks to be identified.
Behavioral and Emotional Assessment

In younger children it may be useful to observe the feeding behavior, and video
material can be used therapeutically to demonstrate specific behaviors,
interactions, and later to show change. Observation will identify whether food
intake is limited in quantity or range, and if there is excessive, irregular, or
chaotic food intake or abnormal practices such as vomiting or purging. In
addition, patterns of behavior, such as inability to accommodate normal
variations in eating patterns and dietary content, which can manifest as
mealtime conflicts, can be identified, as can peer group interaction, social
withdrawal, social isolation, and reluctance to engage in peer activities.
In addition, assessment is needed regarding the emotional well-being of the
child, as well as to establish the level of anxiety surrounding the feeding/eating
relationship for the child, for the parents/caregivers, and for the professionals
involved. The child’s view should be sought if he or she is old enough to talk.
Otherwise low mood, mood swings, and anxiety, including panic attacks and
specific food phobias, may be observable. In young children, play and artwork
may help to identify feelings of aggression, sadness, and fear. There are also a
number of responses evident in feedback from children and adults with
Asperger’s syndrome suggesting that these children may be vulnerable to being
overwhelmed with sensory input and consequent anxiety, fear, and panic. This
may be associated with unusual cognitions regarding the effects of food or
beliefs designed to limit the anxiety, such as only eating from orange packets.
This set of emotional responses and coping with new foods is on a continuum
with that seen in extremely picky eaters, especially as they grow older and
develop beliefs about the potential harm that food may represent.
Observation of the feeding relationship also allows identification of the few
cases in which feeding practices have reached a point of coercion or become
abusive in some other way. Examples include the restriction of food, harsh and
punitive responses, and neglect. Recognition of these difficulties will inform
other aspects of assessment and a formulation reached within the framework
outlined above.
A final part of the assessment process is to ascertain where the child and the
parents are up to in terms of their wish to change, to assess their perception of
64 NICHOLLS
their own ability to achieve change, and to assess their understanding of the
process by which change can be achieved.
Other professionals may be needed in the assessment process, e.g., other
physicians, nurse specialists, speech and language therapists, dietitians,
occupational therapists, social workers, and play specialists. Liaison with the
child’s professional network outside the hospital will give an indication of
previous treatment strategies that have been tried and of the extent to which
professional concerns play a part in the context of the problem.
The assessment process allows more than a diagnosis to be established. It
encompasses a risk assessment in terms of physical compromise, vulnerability to
abuse or neglect, and risk of aggression/violence to others. Itenables
determination of the impact of the feeding problem on the child’s development
and general functioning as well as on the family. Expectations of treatment are
ascertained, together with some provisional information about level of
motivation and readiness to change. A formulation can be made within the
framework outlines and this formulation shared with the family as a starting
point for a collaborative approach to the problem.
MANAGEMENT OF FEEDING PROBLEMS

It is beyond the scope of this text to comprehensively consider the management
of feeding problems, although a number of useful publications exist
outlining approaches based on behavioral, psychodynamic, and family-based
approaches (50). Evidence-based treatments in this area are scarce, hampered in
part by the difficulties in classification of this heterogeneous group of children
and families, as discussed above. In addition, different treatment approaches
reflect the fact that children with feeding difficulties are seen by a variety of
professionals in a variety of settings, including pediatric or mental health
settings. Most commonly feeding problems present in primary care to general
practitioners and other health care professionals, from whom more complex
cases will be referred to nutritional and dietetic services or specialized pediatric
services, particularly gastroenterology. In addition, the common cooccurrence
of feeding disorders and other developmental difficulties means that child
development clinics regularly see children with feeding problems. Psychology
services may be involved early on if behavioral and interactional aspects of the
feeding problem are clear. Specialist feeding teams tend to be multidisciplinary
and involve many or all of the professionals outlined above, i.e., medical,
nursing, nutritional, and psychological expertise. For example, a
multidisciplinary team has been shown to be more effective in achieving weight
gain in cases of nonorganic failure to thrive than treatment in a primary care
setting (51).
A guiding principle of treatment is that it should enable oral feeding at

whatever developmental and physical level the child is capable. This means
supporting parents/caregivers to do the work of feeding, and supplying them
with the necessary information and skills this entails. It may be that oral feeding
is contraindicated and in these cases intervention should focus on supporting the
caregivers with this decision. As our conceptual framework suggests, it may be
easier to address maintaining factors than causal factors, although the
formulation should identify significant contributory factors in the past for both
child and parents, and any medical conditions should be clarified and managed,
e.g., reflux, choking, or aspirations due to dysphagia problems or to allergy/
intolerance, before proceeding to further work.
Although treatment is specific to the problem and to the child and family, a
number of elements are, in our view, essential. Treatment of young people with
feeding and eating disorders works best when it is collaborative and based on a
comprehensive, multidisciplinary assessment. Treatment should be appropriate
to level of complexity. Treatment should be responsive to the developmental
stage, physically and psychologically, of the child and take into account both the
degree of independence the child has within the family and the resources within
the family to support the child. In other words, the treatment should fit the
patient. Finally, assessment and treatment approaches must be reviewed,
developed, and evaluated.
OUTCOME
Remarkably little is known of the outcome of feeding disorders in the Western
world, although the effects of malnutrition are well documented from studies in
developing countries. In one of very few studies, the developmental sequelae of
infant failure to thrive were examined in 42 unreferred 6-year-olds with a
history of severe failure to thrive at age 1 year. At 6 years, children with a
history of failure to thrive were considerably smaller than matched comparisons
in terms of body mass index and height and weight centiles. Failureto-thrive
patients had more limited quantitative and memory skills than controls, but
there were no significant differences in general cognitive functioning once
maternal IQ was taken into consideration (53). In this small series, therefore,
there was little evidence of adverse effects of early malnutrition on cognitive
functioning by school age. The most comprehensive follow-up study to date has
been that of Dahl and colleagues, who followed 25 of their sample of infant
problem eaters. At primary school age, those who had initial refusal to eat
continued to present more eating problems both at home and at school than
controls but were otherwise no different on measures of general behavior,
somatic health, or growth. Whether these problem eating behaviors persisted into
66 NICHOLLS
adolescence and possible associations with later eating disorders or problems

with weight regulation is unknown.
For selective (picky) eating that has persisted into late childhood or
adolescence, in our sample of 20 children only those who had had specific
intervention improved their range of foods (26). These were selected cases
referred to a specialist service and the generalizability of these findings is
unclear. For anorexia nervosa, only a few studies have reported long-term
follow-up in younger patients only, although many studies include some young
patients (52). Overall, outcome in childhood-onset anorexia nervosa is roughly
equivalent to later-onset disorders.
SUMMARY
A developmental framework encompassing the range of feeding and eating
problems from childhood to adulthood is needed, as increasing evidence of
continuities between feeding and eating problems are found. A small but
expanding literature is emerging identifying relationships between early feeding
difficulties and later eating concerns, and in the mechanisms for systemic and
transgenerational influence of food-related beliefs and cognitions. Central to
this are issues of control and the development of selfregulation, concepts that go
far beyond the feeding behavior in terms of their importance and impact on
child development.
There are still many areas where knowledge is lacking, in terms of both
theoretical understanding and treatment approaches. In particular, issues of
classification remain far from resolved, nor is it clear in which conceptual
framework feeding problems are best classified and understood. This hampers
many aspects of further research. The relationship between parent-defined
feeding problems from primary care and community samples and cases of
feeding “disorder” presenting clinically needs further consideration. Treatment
and outcome studies are needed.
ACKNOWLEDGMENT
My thanks to Catherine Dendy for her comments and conceptual framework.
REFERENCES
1. Stein MA, Mendelsohn J, Obermeyer WH, Amromin J, Benca R. Sleep and
behavior problems in school-aged children. Pediatrics 2001; 107(4):E60.
2. Stein DJ, Bouwer C, Van Heerden B. Pica and the obsessive-compulsive spectrum
disorders. S Afr Med J 1996; 86(suppl 12): 1582–1586.
3. Olden KW. Rumination. Curr Treat Options Gastroenterol 2001; 4(4):351–358.

4. Khan S, Hyman PE, Cocjin J, Di Lorenzo C. Rumination syndrome in adolescents.
J Pediatr 2000; 136(4):528–531.
5. Crist W, Napier-Phillips A. Mealtime behaviors of young children: a comparison of
normative and clinical data. J Dev Behav Pediatr 2001; 22(5):279–286.
6. Skuse D, Wolke D. The nature and consequences of feeding problems in infancy.
In: Cooper P, Stein A, eds. Feeding Problems and Eating Disorders in Children and
Adolescents. Reading: Harwood Academic, 1992:1–25.
7. Reau NR, Senturia YD, Lebailly SA, Christoffel KK. Infant and toddler feeding
patterns and problems: normative data and a new direction. Pediatric Practice
Research Group. J Dev Behav Pediatr 1996; 17(3): 149–153.
8. Dahl M, Sundelin C. Early feeding problems in an affluent society. I. Categories
and clinical signs. Acta Paediatr Scand 1986; 75(3):370–379.
9. Burklow KA, Phelps AN, Schultz JR, McConnell K, Rudolph C. Classifying
complex pediatric feeding disorders. J Pediatr Gastroenterol Nutr 1998; 27(2):
143–147.
10. Rudolph CD. Feeding disorders in infants and children. J Pediatr 1994; 125
(6 Pt 2):S116-S124.
11. Wickendon M. The development and disruption of feeding skills: how speech and
language therapists can help. In: Southall A, Schwartz A, eds. Feeding Problems in
Children: A Practical Guide. Abingdon, Oxford: Radcliffe Medical Press, 2000:
3–23.
12. Palmer MM, Crawley K, Blanco IA. Neonatal Oral-Motor Assessment scale: a
reliability study. J Perinatol 1993; 13(l):28–35.
13. Douglas J. Behavioural approaches to the assessment and management of feeding
problems in young children. In: Southall A, Schwartz A, eds. Feeding Problems in
Children: A Practical Guide. Abingdon, Oxford: Radcliffe Medical Press, 2000:
42–57.
14. Harris G, Booth IW. The nature and management of eating problems in preschool
children. In: Cooper PJ, Stein A, eds. Feeding Problems and Eating Disorders in
Children and Adolescents. Monographs in Clinical Pediatrics No 5. Reading:
Harwood Academic, 1992:61–85.
15. Chatoor I, Hirsch R, Ganiban J, Persinger M, Hamburger E. Diagnosing infantile
anorexia: the observation of mother-infant interactions. J Am Acad Child Adolesc
Psychiatry 1998; 37(9):959–967.
16. Chatoor I, Ganiban J, Harrison J, Hirsch R. Observation of feeding in the diagnosis
of posttraumatic feeding disorder of infancy. J Am Acad Child Adolesc Psychiatry
2001; 40(5):595–602.
17. Koon R. Conversion dysphagia in children. Psychosomatics 1983; 24:182–184.
18. Birch LL, Fisher JA. Appetite and eating behavior in children. Pediatr Clin North
Am 1995; 42(4):931–953.
19. Stein A, Woolley H, Murray L, Cooper P, Cooper S, Noble F, et al. Influence of
psychiatric disorder on the controlling behaviour of mothers with 1-year-old
68 NICHOLLS
infants. A study of women with maternal eating disorder, postnatal depression and
a healthy comparison group. Br J Psychiatry 2001; 179:157–162.
20. Whelan E, Cooper PJ. The association between childhood feeding problems and
maternal eating disorder: a community study. Psychol Med 2000; 30(l):69–77.
21. McCann JB, Stein A, Fairburn CG, Dunger DB. Eating habits and attitudes of
mothers of children with non-organic failure to thrive. Arch Dis Child 1994; 70(3):
234–236.
22. Birch LL, Fisher JO. Mothers’ child-feeding practices influence daughters’ eating
and weight. Am J Clin Nutr 2000; 71(5):1054–1061.
23. Jacobi C, Agras WS, Hammer L. Predicting children’s reported eating disturbances
at 8 years of age. J Am Acad Child Adolesc Psychiatry 2001; 40(3):364–372.
24. Bryant-Waugh R. Overview of the eating disorders. In: Lask B, Bryant-Waugh R,
eds. Anorexia nervosa and related eating disorders in childhood and adolescence.
Hove, East Sussex: Psychology Press, 2000:27–40.
25. Pliner P, Loewen ER. Temperament and food neophobia in children and their
mothers. Appetite 1997; 28:239–254.
26. Nicholls D, Christie D, Randall L, Lask B. Selective eating: symptom, disorder or
normal variant? Clin Child Psychol Psychiatry 2001; 6(2):257–270.
27. Rydell AM, Dahl M, Sundelin C. Characteristics of school children who are choosy
eaters. J Genet Psychol 1995; 156(2):217–229.
28. Bryant-Waugh R, Lask B. Eating disorders in children. J Child Psychol Psychiatry
1995; 36(2): 191–202.
29. Reilly SM, Skuse DH, Wolke D, Stevenson J. Oral-motor dysfunction in children
who fail to thrive: organic or non-organic? Dev Med Child Neurol 1999; 41 (2):
115–122.
30. Marchi M, Cohen P. Early childhood eating behaviors and adolescent eating
disorders. J Am Acad Child Adol Psychiatry 1990; 29(1):112–117.
31. Rastam M. Anorexia nervosa in 51 Swedish adolescents: premorbid problems and
comorbidity. J Am Acad Child Adol Psychiatry 1992; 31(5):819–829.
32. Jacobs BW, Isaacs S. Pre-pubertal anorexia nervosa: a retrospective controlled
study. J Child Psychol Psychiatry 1986; 27(2):237–250.
33. Stark LJ, Jelalian E, Powers SW, Mulvihill MM, Opipari LC, Bowen A, et al.
Parent and child mealtime behavior in families of children with cystic fibrosis.
J Pediatr 2000; 136(2): 195–200.
34. Schwarz SM, Corredor J, Fisher-Medina J, Cohen J, Rabinowitz S. Diagnosis and
treatment of feeding disorders in children with developmental disabilities.
Pediatrics 2001; 108(3):671–676.
35. Harris G, Blissett J, Johnson R. Food refusal associated with illness. Child Psychol
Psychiatry Rev 2000; 5(4): 148–156.
36. Birch LL, Fisher JO. Development of eating behaviors among children and
adolescents. Pediatrics 1998; 101(3 Pt 2):539–549.
37. Patel P, Wheatcroft R, Park RJ, Stein A. The children of mothers with eating
disorders. Clin Child Fam Psychol Rev 2002; 5(1):1–19.
38. Cooper PJ, Watkins B, Bryant-Waugh R, Lask B. The nosological status of early onset
anorexia nervosa. Psychol Med 2002; 32(5):873–880.
39. Agras S, Hammer L, McNicholas F. A prospective study of the influence of eating-
disordered mothers on their children. Int J Eat Disord 1999; 25(3):253–262.
40. Saarilehto S, Keskinen S, Lapinleimu H, Helenius H, Simell O. Connections
between parental eating attitudes and children’s meagre eating: questionnaire
findings. Acta Paediatr 2001; 90(3):333–338.
41. Nicholls D, Bryant-Waugh R. Children and young adolescents. In: Treasure J,
Schmidt U, van Furth E, eds. Handbook of Eating Disorders. Chichester: John
Wiley and Sons, 2003:415–433.
42. Fairburn CG, Cooper Z. The eating disorders examination. 12th ed. In: Fairburn
CG, Wilson GT, eds. Binge Eating: Nature, Assessment and Treatment. New
43. Bryant-Waugh R, Cooper P, Taylor C, Lask B. The use of the Eating Disorder
Examination with children: a pilot study. Int J Eat Disord 1996; 19(4):391–397.
44. Nicholls D, Chater R, Lask B. Children into DSM IV don’t go: a comparison of
classification systems for eating disorders in childhood and early adolescence. Int J
Eat Disord 2000; 28(3):317–324.
45. Higgs JF, Goodyer IM, Birch J. Anorexia nervosa and food avoidance emotional
disorder. Arch Dis Child 1989; 64:346–351.
46. Birch LL, Fisher JO, Grimm-Thomas K, Markey CN, Sawyer R, Johnson SL.
Confirmatory factor analysis of the Child Feeding Questionnaire: a measure of
parental attitudes, beliefs and practices about child feeding and obesity proneness.
Appetite 2001; 36(3):201–210.
47. Cole TJ, Freeman JV, Preece MA. Body mass index reference curves for the UK,
1990. Arch Dis Child 1995; 73(l):25–29.
48. Williams S. Body Mass Index reference curves derived from a New Zealand birth
cohort. N Z Med J 2000; 113(1114):308–311.
49. Luciano A, Bressan F, Zoppi G. Body mass index reference curves for children aged
3–19 years from Verona, Italy. Eur J Clin Nut 1997; 51(1):6–10.
50. Southall A, Schwartz A. Feeding Problems in Children: A Practical Guide.
Abingdon, Oxford: Radcliffe Medical Press, 2000.
51. Bithoney WG, McJunkin J, Michalek J, Snyder J, Egan H, Epstein D. The effect of
a multidisciplinary team approach on weight gain in nonorganic failure-tothrive
children. J Dev Behav Pediatr 1991; 12(4):254–258.
52. Steinhausen HC. Outcome of anorexia nervosa in the younger patient. J Child
Psychol Psychiatry 1997; 38(3):271–276.
53. Boddy J, Skuse D, Andrews B. The developmental sequelae of nonorganic failure to
thrive. J Child Psychol Psychiatry Nov 2000; 41(8): 1003–1014.
70 NICHOLLS
4
Epidemiology of Eating Disorders and
Disordered Eating: A Deveiopmental
Overview
Maria Råstam
Göteborg University, Göteborg, Sweden
Christopher Gillberg
Göteborg University, Göteborg, Sweden, and St. George’s
Hospital Medical School,
London, England
Daphne van Hoeken
Parnassia, The Hague, The Netherlands
Hans Wijbrand Hoek
Parnassia, The Hague, The Netherlands, and Mailman School of
Public Health, Columbia University,
How common are the eating disorders? The social and economic burden for the
individual and the society makes the epidemiology of eating disorders an
important issue. Furthermore, studies of representative samples of eating
disorders are the soundest base for the examination of etiologic factors and
outcome. The most consistent finding across time and studies seems to be that
anorexia nervosa predominantly affects adolescent girls.
How do the eating disorders develop over time? Do they belong to a
spectrum with disordered eating at one end and anorexia and bulimia nervosa at
the other, with the partial syndromes of anorexia and bulimia nervosa in the
middle? How often does disordered eating develop into a (partial) eating
disorder? Most individuals with anorexia nervosa develop bulimia nervosa or a
partial syndrome of anorexia or bulimia nervosa in the course of their
recovery (1). There are no data as to how often those subjects with a diagnosis
of Eating Disorders-Not Otherwise Specified (ED-NOS) progress to the full
syndrome of anorexia or bulimia nervosa.
72 RÅSTAM ET AL.
Epidemiological studies have to counter a number of methodological

problems (2,3). Problems specific to the eating disorders are their low
prevalence and the tendency of eating disorder subjects to conceal their illness
and avoid professional help (4). These make it necessary to study a very large
number of subjects from the general population in order to reach enough
differential power for the cases. This is highly time and cost intensive. Several
strategies have been used to circumvent this problem, in particular case register
and other record-based studies, two-stage studies, and studies of special
populations.
The limitations of record-based studies are considerable (4). Registerbased
frequencies represent cases detected in inpatient and, occasionally, outpatient
care. Treated patients represent only a minority of all cases. Findings from case
registers and hospital records are of more value to treatment planning than for
generating hypotheses on the etiology of disease because there is no direct
access to the subjects and the additional information that is available is usually
limited and of a demographic nature only.
At present a two-stage screening approach is the most widely accepted
procedure for case identification. First a large population is screened for the
likelihood of an eating disorder by means of a screening questionnaire,
identifying an at-risk population (first stage). Then definite cases are established
using a personal interview on subjects from this at-risk population as well as on
a randomly selected sample of those not at risk (second stage) (5).
Methodological problems of two-stage studies are poor response rates,
sensitivity and specificity of the screening instrument, and the often restricted
size of the interviewed groups, particularly those not at risk (6).
Studies of special populations address a particular segment of the general
population, selected a priori for being at increased risk, such as female high
school or university students, athletes, or a particular age cohort. The major
methodological problem associated with this type of study is the specificity of
the findings to the selected subset of the general population.
EATING DISORDERS
Classification
In the sections on anorexia nervosa and bulimia nervosa only studies using strict
definitions of these eating disorders (meeting Russell, DSM, or ICD criteria) are
discussed. Another category, the ED-NOS, is a mixed category. It includes
heterogeneity of patients who do not meet all criteria for anorexia nervosa or
bulimia nervosa but who have symptoms severe enough to qualify them as
EPIDEMIOLOGY 73
having a clinically significant eating disorder. In DSM-IV (7) a provision was

made for a separate eating disorder category to be researched further, the binge
eating disorder (BED). Only limited epidemiological information is available on
ED-NOS and BED to date. The epidemiology of eating disorders has
been reviewed before, e.g., Hoek (3), Hsu (4), Van Hoeken et al. (8), and
Hoek et al. (9).
Disordered Eating
Dieting is very common among young women and is considered a major risk factor
for disordered eating and the development of eating disorders (10,11). Fairburn
and colleagues compared subjects with bulimia nervosa (12), BED (13), and
anorexia nervosa (14) with each other, with healthy control subjects without an
eating disorder (general risk factors), and with subjects with other psychiatric
disorders (specific risk factors), recruited from general practices in
Oxfordshire, England. After screening with self-report questionnaires, a
retrospective risk factor interview was carried out that addressed the premorbid
period. The results suggest that both bulimia nervosa and BED are most likely to
develop in dieters who are at risk of obesity and psychiatric disorder in
general (12,13). Disordered eating in teenagers may be an important future
health risk. In a community-based longitudinal investigation in New York State,
10% of girls and 1% of boys (out of 717 individuals, 51% female) were
diagnosed with DSM-IV eating disorder diagnosis in adolescence. Compared to
those without an eating disorder diagnosis, they had significantly more physical
problems and anxiety and depressive disorders during early adolescence.
Furthermore, data indicated that problems with eating or weight during
adolescence may be associated with physical and mental health problems during
early adulthood, regardless of whether the full syndrome of an eating disorder
had been present or not (15).
In Sweden in 1880, a health survey of schools for girls (from the upper
classes) showed that 35% of 3000 adolescent girls had “chlorosis,” defined as
being pale, weak, and anemic with no or irregular menstruations. The cause
was ascribed to the half starvation recommended for girls of that social standing.
The illness was reported occasionally to be epidemic in higher school grades (16).
Nowadays even young children have grasped the ideal of a thin female body. In
a British study, one of five 9-year-old girls was dieting, and, although more
common in heavier girls, dieting was not restricted to that group (17). In a
study of 318 girls and boys aged 9 12 years (18), 45% wanted to lose weight,
significantly more girls than boys. Dieting occurred in 37% (Table 1), and 6.9%
had high scores on the Children’s Eating Attitudes Test (ChEAT), indicating
disordered eating. Trying to replicate the Maloney study on children in the
74 RÅSTAM ET AL.
same age range in Israel, Sasson and coworkers (19) found 8.8% high scorers on
the ChEAT.
In a cross-sectional self-report study of eating attitudes and behaviors in girls
attending junior high and high schools in Toronto, Ottawa, and Hamilton (20),
one in five girls were dieting (Table 1). Bingeing and purging at a frequency
consistent with a diagnosis of DSM-IV bulimia nervosa occurred in 4.9%.
Disordered eating attitudes and behaviors tended to increase gradually
throughout adolescence, especially in heavier teenagers (20). Table 1
summarizes one-stage (or the first stage of two) surveys of disordered eating in
children and adolescents with a participation rate of at least 70%.
In a questionnaire study (28), 35% out of 1241 Swedish schoolgirls aged
14–18 years were dieting. A follow-up of 130 females with dieting plus mental
problems gave an accumulated prevalence of 1.2% of anorexia nervosa for the
whole female population (29). In South Carolina (22), 3129 middle school
students (out of 4282) completed a self-report questionnaire, the Kids’ Eating
Disorders Survey (KEDS). Dieting and disordered eating was found to be fairly
common in 9-to 16-year-olds (Table 1). In a second stage, designed to validate
the screening tool, altogether 165 subjects, a mixture of high scorers on the
KEDS and randomly selected students, were personally interviewed. One
subject fulfilled criteria for DSM-IIIR anorexia nervosa and 22 for ED-NOS.
With the new male body ideal in the media, young men have become
increasingly engaged in body building and the use of anabolic steroids (30, 31).
Many of these young men are reminiscent of young females with anorexia nervosa
in their compulsive pursuit of more training and a specific body image. Still,
generally, teenage girls diet because they feel fat whereas most boys who diet
are fat (32).
In a representative sample of 4746 adolescents from 31 Minnesota public
schools, about 6% were self-reported vegetarians (33). The vegetarians were
more likely than nonvegetarians to be female and to be involved in weight
control behaviors, including some with an eating disorder. Some individuals
may use vegetarianism as a disguise for dieting. In a communitybased
study on teenage-onset anorexia nervosa, 25% introduced their diet as
vegetarianism (34), and during a 10 year follow-up period altogether 41 % had
a vegetarian phase (1). One cannot exclude the possibility that for some
individuals a drastic change of diet per se might be a risk for the development of
an eating disorder.
TABLE 1 Surveys of Disordered Eating in School Years
For Tables 1–4 ANIS, Anorexia Nervosa Inventory Scale; BCDS, Bulimic Cognitive Distortions Scale; ChEAT, Childrens Eating Attitudes Test-26;
DIS, Diagnostic Interview Schedule; DSED, Diagnostic Survey for Eating Disorders; EAT-26, Eating Attitudes Test-26; EDE, Eating Disorder
Examination; EDE-Q, Eating Disorder Examination-Questionnaire; EDE-S, Eating Disorder Examination-Screening Version; EDI, Eating Disorder
Inventory; KEDS, Kids’ Eating Disorders Survey. a first stage of two-stage study.
EPIDEMIOLOGY 75
76 RÅSTAM ET AL.
Anorexia Nervosa
Prevalence
The current standard for prevalence studies of eating disorders are studies
employing a two-stage selection of cases. Table 2 summarizes the two-stage
surveys of anorexia nervosa in young females. All researchers have succeeded in
obtaining high response rates of 85% or more, except Meadows et al. (38), who
obtained a response rate of 70%. Those two-stage surveys that identified cases
found a prevalence rate of strictly defined anorexia nervosa of between 0.2%
and 0.8% of young females, with an average prevalence of 0.3%. These rates
are possibly minimal estimates. Most studies found much higher prevalence
rates for partial syndromes of anorexia nervosa. One twostage study (23)
screened 607 14- to 17-year-olds with a wide battery of questionnaires. The
surveys was completed by 583 participants. The first stage reported on eating
behaviors in teenagers (Table 1). In a second stage, 399 parents of screen-
positive and control subjects were interviewed. The adolescents were not
examined. At the first screening a clinical eating disorder was suspected in two
cases, one a girl with anorexia and the other a girl with bulimia nervosa, and this
was confirmed in parents’ interviews. No more clinically significant eating
disorders were found.
Two other studies give prevalence figures for the entire population. In a
general-practice study in the Netherlands, Hoek (46) found a raw point
prevalence rate of 18.4 per 100,000 of the total population (95% CI
12.7–26.8) on January 1, 1985. Lucas et al. (47) used a very extensive case finding
method. It included all medical records of health care providers, general
practitioners, and specialists in Rochester, Minnesota. They also screened
records mentioning related diagnostic terms for possible undetected cases. They
found an overall sex-and age-adjusted point prevalence of 149.5 per 100,000
(95% CI 119.3–179.7) on January 1, 1985. A main explanation for this
difference can be found in the inclusion of probable and possible cases by Lucas
et al. Definite cases constituted only 39% (82 of 208) of all incident cases
identified in the period 1935–1989 (48). Applying this rate to the point
prevalence of 149.5 gives an estimated point prevalence of 58.9 per 100,000
for definite cases in Rochester, Minnesota on January 1, 1985. The remaining
difference with the point prevalence reported by Hoek (46) could be explained
by the greater variety of medical sources searched by Lucas et al. (47).
TABLE 2 Two-Stage Surveys of Prevalence of Anorexia Nervosa in Young Females
a
EPIDEMIOLOGY 77
Not found by screening (EAT score below threshold).

78 RÅSTAM ET AL.
Incidence
Incidence rate differences between groups are better clues to etiology than
prevalence rate differences because they refer to recently started disease (49). The
incidence studies of anorexia nervosa have used psychiatric case registers,
medical records of hospitals in a circumscribed area, registrations by general
practitioners, or medical records of health care providers in a community.
Table 3 summarizes the results of the studies on the incidence of anorexia nervosa
that report overall rates for a general population sample. The overall rates vary
considerably, ranging from 0.10 in a hospital-recordsbased study in Sweden in
the 1930’s to 12.0 in a medical-records-based study in the USA in the 1980’s,
both per 100,000 population per year.
Incidence rates derived from general practices on average represent more
recently started eating disorders than those based on other medical records.
There were two studies of this type. In the study by Hoek and colleagues (58),
general practitioners using DSM-IIIR criteria have recorded the rate of eating
disordersin a large (1985:7V=151,781), representative sample (l.l%) of the
Dutch population. The incidence rate of anorexia nervosa was 8.1 per 100,000
person years (95% CI 6.1–10.2) during 1985–1989. During the study period
63% of the incident cases were referred to mental health care, accounting for an
incidence rate of anorexia nervosa in mental health care of 5.1 per year per
100,000 population. Turnbull et al. (59) searched the UK General Practice
Research Database (GPRD), covering 550 general practitioners and 4 million
patients, for first diagnoses of anorexia in the period 1988–1993. A randomly
selected subset of cases was checked with DSM-IV criteria, from which
estimates for adjusted incidence rates were made. For anorexia nervosa they
found an age-and sex-adjusted incidence rate of 4.2 (95% CI 3.4–5.0) per
100,000 population in 1993.
Lucas et al. (47,48) used the most extensive case finding method (see the
section on prevalence). Over the period 1935–1989, they report an overall age-
and sex-adjusted incidence rate of anorexia nervosa of 8.3 per 100,000 person-
years (95% CI 7.1–9.4).
Age and Sex

Earlier reports indicate a peak incidence in the late teens (55,60), whereas later
studies indicate a peak age of onset between 14 and 15 years (61). Adolescent
onset points to the role of pubertal changes such as body composition
changes (62) and the rapid rise in estrogen levels (63). The psychological and
social changes of adolescence have been shown to be important (14,39). The
trend toward an earlier menarche suggests that the peak onset age might be even
EPIDEMIOLOGY 79
TABLE 3 Incidence of Anorexia Nervosa per Year per 100,000 Population
earlier (64). Incidence rates of anorexia nervosa are highest for females 15–19
years old. These constitute approximately 40% of all identified cases and 60%
of female cases. For example, Lucas et al. (48) report an incidence rate of 73.9
per 100,000 person-years for 15- to 19-year-old women during the period
1935–1989, with a continual rise since the 1930s to a top rate of 135.7 for the
period 1980–1989. On an overall female rate of 15.0 per 100,000 population
per year, Lucas et al. (48) report a rate of 9.5 for 30- to 39year-old women, 5.9
for 40- to 49-year-old women, 1.8 for 50- to 59-year-old women, and 0.0 for
women 60 years and older.
The majority of male incidence rates reported were below 0.5 per
100,000 population per year, e.g., Turnbull et al. (59). In those studies where
it is reported, the female-to-male ratio usually is around 11:1, e.g., Hoek
et al. (58).
80 RÅSTAM ET AL.
Time Trends
In spite of some reports of an increased rate of anorexia nervosa from the
mid-1960s to the mid-1980s in certain regions (48), there is no consensus
that anorexia nervosa prevalence generally has increased over the corresponding
time period. Case register studies prior to the 1980s show at most a slight
increase over time of incident anorexia nervosa cases (3). The studies done in the
1980s show widely diverging incidence rates. Most likely there is a
methodological explanation for these differences. The main problem lies in the
need for long study periods. This results in a sensitivity of these studies to minor
changes in absolute incidence numbers and in methods, e.g., variations in
registration policy, demographic differences between populations, faulty
inclusion of readmissions, the particular method of detection used, or the
availability of services (65–67).
From the studies that have used long study periods, it may now be concluded
that there is an upward trend in the incidence of anorexia nervosa since the
1950s. The increase is most substantial in females 15–24 years of age. Lucas
et al. (48) found that the age-adjusted incidence rates of anorexia nervosa in
females 15–24 years old showed a highly significant linear increasing trend from
1935 to 1989, with an estimated rate of increase of 1.03 per 100,000 person-
years per calendar year. In 10- to 14-year-old girls a rise in incidence was
observed for each decade since the 1950s. The rates for men and for women 25
and older remained relatively low.
All record-based studies will grossly underestimate the true incidence
because not all patients will be referred to mental health care or become
hospitalized (68). The increase in incidence rates of registered cases implies at
least that there is an increased demand for health care facilities for anorexia
nervosa.
Bulimia Nervosa
Prevalence
In 1990 Fairburn and Beglin gave a review of the prevalence studies on bulimia
nervosa (6). This landmark review yielded the generally accepted prevalence
rate of 1 % of young females with bulimia nervosa according to DSM criteria.
Table 4 summarizes two-stage surveys of bulimia nervosa in young females that
have been published since the review by Fairburn and Beglin. Despite the
different classifications used—DSM-III (72) versus DSM-IIIR (73)—and
different types of prevalence rates provided [lifetime prevalence, e.g.,
sgfsdfdsdgfsdfs
TABLE 4 Two-Stage Surveys of Prevalence of Bulimia Nervosa in Young Females
EPIDEMIOLOGY 81
82 RÅSTAM ET AL.
Bushnell et al. (69), versus point prevalence, e.g., Rathner and Messner (43)],
the aggregated prevalence rate according to DSM criteria remains 1%. The
prevalence of subclinical eating disorders is substantially higher than that of full-
syndrome bulimia nervosa, e.g., Whitehouse et al. (42): 1.5% for fullsyndrome
and 5.4% for partial-syndrome bulimia nervosa.
The results of three studies not using a two-stage procedure for case finding
are reported here as well because they are likely to represent the entire
population of women. Garfinkel et al. (74) assessed eating disorders in a ran
dom, stratified, nonclinical community sample, using a structured interview for
the whole sample. They reported a lifetime prevalence for bulimia nervosa of
1.1 % in women and of 0.1 % in men aged 15–65, using DSM-IIIR criteria.
This is a study of a representative sample of women in the United States and
provides a good estimate of the lifetime prevalence of bulimia nervosa in the
United States. Another study from the United States designed for examining the
relationship among assault, bulimia nervosa, and BED (DSM-IV criteria) by
interviewing by telephone a national representative sample of 3006 women
yielded a prevalence for bulimia nervosa of 2.4% (75). In a study by Soundy
et al. (76), no attempt was made to determine prevalence rates for bulimia
nervosa because they considered the information about how long symptoms had
persisted and the long duration (mean 39.8 months) of symptoms before
diagnosis to be too unreliable.
Incidence
There have been few incidence studies of bulimia nervosa. The most obvious
reason is the lack of criteria for bulimia nervosa in the past. Most case registers
use the International Classification of Diseases, currently ICD-10 (77). The
ICD-9 (78) and previous versions did not provide a separate code for bulimia
nervosa. Bulimia nervosa has been distinguished as a separate disorder by
Russell in 1979 (79) and DSM-III in 1980 (72). Before 1980 the term “bulimia”
in medical records designated symptoms of heterogeneous conditions
manifested by overeating, but not the syndrome as it is known today.
Therefore, it is difficult to examine trends in the incidence of bulimia nervosa
or a possible shift from anorexia nervosa to bulimia nervosa, which might have
influenced the previously described incidence rates of anorexia nervosa.
Three studies are of main interest: those of Soundy et al. (76), Hoek
et al. (58), and Turnbull et al. (59). Soundy and colleagues used methodology
similar to that in the long-term anorexia nervosa study by Lucas et al. (47)
screening all medical records of health care providers, general practitioners, and
specialists in Rochester, Minnesota over the period 1980–1990 for a clinical
diagnosis of bulimia nervosa as well as for related symptoms (76). Hoek and
EPIDEMIOLOGY 83
colleagues studied the incidence rate of bulimia nervosa using DSM-IIIR criteria
in a large general-practice study representative of the Dutch population,
covering the period 1985–1989 (58). Turnbull and colleagues screened the
General Practice Research Database (GPRD), covering a large, representative
sample of the English and Welsh population, for first diagnoses of anorexia
nervosa and bulimia nervosa in 1993 (59). The three studies all report an annual
incidence of bulimia nervosa around 12 per 100,000 population: 13.5 for
Soundy et al., 11.5 for Hoek et al., and 12.2 for Turnbull et al.
Another general-population study of a relatively small population (lower than
50,000 inhabitants) is of some interest because it was based on a homogeneous
population from a small Danish island (80). The Danish Psychiatric Case
Register and the local records of in-and outpatients were screened, psychiatrists
and physicians in primary care were interviewed, as were all school medical
officers. In the period 1985–1989, the average annual incidence of bulimia
nervosa was 6.8 per 100,000 population.
Age and Sex

Soundy et al. (76) report an incidence of bulimia nervosa of 26.5 for females
and of 0.8 for males per 100,000 population, yielding a female-to-male ratio of
33:1. Hoek et al. (58) report similar rates of 21.9 for females and 0.8 for males
per 100,000 population, yielding a female-to-male ratio of 27:1. For the highest
risk group of 20- to 24-year-old females, rates close to 82 per 100,000 are
found: 82.7 for Soundy et al. and 82.1 for Hoek et al. Hoek et al. report a rate
of 8.3 per 100,000 for women aged 35–64. Turnbull et al. (59) report an
annual incidence of 1.7 per 100,000 people (men and women) aged 40 and
over.
Time Trends
Soundy et al. (76) found yearly incidence rates to rise sharply from 7.4 per
100,000 females in 1980 to 49.7 in 1983, and then remain relatively constant at
around 30 per 100,000 females. This would seem to be related to the
publication, and following implementation in the field, of DSM-III in
1980 (72), introducing bulimia nervosa as an official diagnostic category. Hoek
et al. (58) report a nonsignificant trend for the incidence rates of bulimia
nervosa to increase by 15% each year in the period 1985–1989. Turnbull
et al. (59) noted a highly significant, threefold increase in bulimia nervosa
incidence rates for women aged 10–39 in the period 1988–1993, increasing
from 14.6 in 1988 to 51.7 in 1993. These incidence rates of bulimia nervosa can
only serve as minimal estimates of the true incidence rate. The reasons are the
84 RÅSTAM ET AL.
lack of data, the greater taboo surrounding bulimia nervosa, and its smaller
perceptibility in comparison with anorexia nervosa.
Eating Disorders NOS

ED-NOS [DSM-IV (7)] refers to a mixed category of individuals who do not
meet all criteria for anorexia nervosa or bulimia nervosa, but who have
symptoms severe enough to qualify them as having a clinically significant eating
disorder. The heterogeneity of the category makes comparison of rates between
studies difficult.
In a two-stage cohort study of Australian students, initially aged 14–15 years
over 3 years with 6-month intervals (11), 3.3% of all girls and 0.3% of all boys
had partial syndromes of an eating disorder at the start of the study. Girls who
engaged in severe dieting (8%) had an 18-fold increase in the risk of developing
an eating disorder within 6 months compared with nondieters. Of course,
severe dieting may already be a sign of an eating disorder under way. Moderate
dieting, present in 60% of all girls, was associated with a fivefold increase in risk
for eating disorder. General psychiatric symptomatology as a separate factor
increased the risk for an eating disorder seven times. All together, 6.6% of all
girls and 1.2% of the boys were classified as having an eating disorder at least
once during the study.

One example of ED-NOS is binge eating disorder (BED), which has the status
of “Diagnostic Category in Need of Further Research” in the DSM-IV (7). BED
is characterized by eating large amounts of food with no sense of control and no
use of compensatory behaviors to avoid the weight gain that is a common
consequence. Weight concern and worry about body shape seem to be a clinical
feature of BED, regardless of BMI (81,82). Studies, including a two-stage
community study (83), support BED as a diagnosis distinct from bulimia
nervosa of the purging type. Furthermore, BED seems to have a different
outcome than nonpurging bulimia nervosa, suggesting a distinction between
nonpurging bulimia nervosa and BED. A general problem with the comparison
of studies of BED—and bulimia nervosa for that matter—lies in the definition of
a binge. Studies differ in the way a binge is defined, resulting in subject groups
that are not fully comparable.
We know of only one prevalence study using a two-stage case identification
procedure in the general population. Cotrufo et al. (84) identified two cases of
BED in a group of 919 females aged 13–19 years, giving a prevalence rate
EPIDEMIOLOGY 85
of 0.2%. The low rate may be due to the relatively young age of the investigated
population. Also, the sample size is rather small for a low-frequency disorder.
In a telephone survey, Dansky et al. (75) found DSM-IV BED to be present in
1% of 3006 adult females. Hay (85) conducted interviews to determine the
prevalence of bulimic-type eating disorders on all subjects in a large general-
population sample (3001 interviews). The mean age of the cases was 35.2 years.
Using DSM-IV criteria, a point prevalence for BED of 1 % was found. Using a
broader definition by Fairburn and Cooper (86), the prevalence was estimated at
2.5%. A weakness of the study was that diagnoses were based on a very limited
number of questions (two gating questions and three additional probes). No
information was given regarding the sensitivity and specificity of the
instrument.
Time Trends
To study the prevalence of binge eating and weight control practices in West
Germany over time, 2130 adult subjects were examined in 1997 and compared
with 1773 adults in 1990 (87). Participation was almost 70%. The same self-
report questionnaire was used in 1990 and in 1997 (complemented with a
personal interview for a subsample in 1997). The prevalence for BED dropped
from 1.5% in 1990 to 0.7% in 1997 in women and from 2.4% to 1.5% in men.
The prevalence for bulimia nervosa dropped from 2.4% to 1.1% in women and
from 2.0% to 1.1 % in men. None of the differences were statistically significant.
SOME DEMOGRAPHIC VARIABLES
Social Class
Most psychiatric disorders show a higher prevalence in the lower socioeconomic
classes. It is difficult to determine whether this is the result of the social
selection process or whether it is caused by social factors (88).
For anorexia nervosa, there has been a traditional belief of an upper social
class preponderance. In reviewing the evidence, Gard and Freeman (89)
concluded that the relationship between anorexia nervosa and high socioeconomic
status is unproven, due to data collection biases including sample size, clinical
status, and referral patterns. A recent study on a large comprehensive clinical
database, covering 692 referrals to a U.K. national specialist center in 33 years’
time, challenges this conclusion: McLelland and Crisp (90) found referrals for
anorexia nervosa from the two highest social classes to be almost twice as high
as expected. They present evidence that their findings are unrelated to
86 RÅSTAM ET AL.
differences in clinical features or in access to their service. In a two-stage survey

of all 15-year-olds in the city of Göteborg, Sweden, the subjects with anorexia
nervosa did not differ from a sex-, age-, and schoolmatched comparison group
concerning social class (91).
For bulimia nervosa, Gard and Freeman (89) conclude that—similar to most
psychiatric disorders—there seems to be preponderance in the lower socio-
economic groups.
Level of Industrialization
It is commonly thought that anorexia nervosa is a Western illness: there
appears to be a developmental gradient across countries, with a predominance
in industrialized, developed countries, linking the disorder to an affluent
society (3). This gradient has been hypothesized to be connected with the
sociocultural theory. This theory holds that eating disorders are promoted by a
“Western” culture favoring slimness as a beauty ideal for females.
By consequence eating disorders would be less prevalent in underdeveloped,
nonWestern cultures. Unfortunately, to date few developing countries have the
facilities and means to arrive at reliable epidemiological data.
According to a recent extensive review of studies from the Far East (92),
body dissatisfaction and dieting rates were similar to those in the West. Dieting
occurred in more than 60% of females in Japan; prevalence rates were around
0.03% for anorexia nervosa and ranged from 1.9% to 2.9% for bulimia
nervosa. Community studies in China found the anorexia nervosa prevalence to
be 0.01 %, and bulimia nervosa rates ranged from 0.5% to 1.3%. The very low
rate of anorexia nervosa in the Far East may seem surprising. It could be due to
a slightly different behavioral phenotype of the syndrome, including cases
masquerading as bulimia nervosa. Reports from Eastern Europe are very scant.
In Hungary and Poland two-stage studies have been conducted; see Tables 2 and
4 (44,70). A questionnaire lifestyle study of 453 medical students between 1995
and 1999 in Slovakia showed excessive underweight (BMI=17.5 or lower)
in 1.0% of men and 2.8% of women (93), suggesting a possible eating disorder.
In Iran, Nobakht and Dezhkam (25) examined disordered eating with EAT-26 in
3100 female students, with no attrition (Table 1). Of 749 high scorers, 99.8%
completed another selfreport questionnaire to assess DMS-IV eating disorder
diagnoses. Accumulated prevalence for full-syndrome anorexia nervosa
was 0.9%, for the partial syndrome 1.84%, and for full-syndrome bulimia
nervosa 3.23%, for the partial syndrome 4.79%. Comparing 59 Iranian female
students living in Tehran to 45 female students of Iranian descent living in Los
Angeles, Abdollahi and Mann (94) found similar amounts of self-reported
disordered eating in the two samples. In a population of 351 secondary school
EPIDEMIOLOGY 87
girls in Cairo, Egypt, 11.4% scored positively on the Eating Attitude Test
(EAT) Questionnaire and were subsequently interviewed (95). Three cases
fulfilled Russell’s criteria for a diagnosis of bulimia nervosa (1.2%), and an
additional 12 individuals (3.4%) were deemed to have a partial syndrome of
bulimia nervosa.
Girls and women born in countries with a low prevalence of anorexia
nervosa seem to have an increased risk of developing the disorder after moving
to a high-frequency region (96,97). Some recent publications cast doubts on the
validity of the sociocultural theory, at least for anorexia nervosa (98,99). For
example, Hoek et al. (100) found an incidence of anorexia nervosa on the
Caribbean island of Curaçao within the lower range of rates reported in
Western countries.
Level of Urbanization
Hoek et al. (58) report that the incidence of bulimia nervosa is three to five
times higher in urbanized areas and cities than in rural areas, whereas ano rexia
nervosa is found with almost equal frequency in areas with different degrees of
urbanization. The drift hypothesis, relating urbanization differences to migration
for educational reasons, is rejected because the differences remain after
adjusting for age. Other social factors involved might be an increased pressure
to be slender and decreased social control in urbanized areas. If these hold true,
this would imply that anorexia nervosa is less sensitive to social factors than
bulimia nervosa, has a more biological origin, and is more driven by other
factors such as a tendency toward asceticism and compulsive behavior.
Occupation
Some occupations appear to be linked to an increased risk of the development
of an eating disorder (101). Typical examples are professions in the world of
fashion (102) and ballet (2). We do not know whether this is a causal factor or
rather the result of disturbed attitudes around body and shape. In other words,
are preanorectics attracted by the ballet world, or are the requirements of the
profession conducive to the development of anorexia nervosa?
According to a meta-analysis of 34 studies, elite athletes had a higher risk of
eating disorders than nonathletes. On the other hand, nonelite athletes were
more satisfied with their bodies than controls, and they seemed to have a
reduced risk of eating disorders (103). In a recent study of 181 elite women
runners in the United Kingdom (104), 7 (4%) had anorexia nervosa, 2
individuals (1%) had bulimia nervosa, and 20 (11%) females had an EDNOS.
88 RÅSTAM ET AL.
CONCLUSIONS
For anorexia nervosa an average prevalence rate of 0.3% was found for young
girls. Although Soundy et al. (76) caution for the possibility of unreliable
information, figures of an average prevalence rate of bulimia nervosa of 1 % in
women and of 0.1 % in men seem accurate. A tentative conclusion is that the
prevalence of BED is more likely to be 1 % than 4% or more.
Assuming that even the studies with the most complete case finding methods
yield an underestimate of the true incidence, as state of the art we conclude that
the overall incidence of anorexia nervosa is at least 8 per 100,000 population
per year and the incidence of bulimia nervosa is at least 12 per 100,000
population per year. The incidence rate of anorexia nervosa has increased
during the past 50 years, particularly in females aged 10–24 years. The
registered incidence of bulimia nervosa has increased, at least during the first
5 years after bulimia nervosa was added to the DSM-III (72). For BED not
enough incidence information is available to summarize.
Little is known about how many people with or without (a combination of)
known risk factors develop an eating disorder over time. Dieting seems a
general nonspecific risk factor, increasing the risk of developing an eating
disorder by about fivefold.
COMMENTS
The value of epidemiology lies in its particular methodology that gives rise to
population-based disease rates and ratios. When properly established, these
rates and ratios provide a scientific basis on the community level for treatment
planning and etiological model building. Epidemiological information is needed
to examine and extend on clinical observations.
The basic epidemiological measures are incidence and prevalence rates. For
the purpose of treatment planning, there is an ongoing need for prevalence
information at the local level. For reasons of time and cost efficiency, this is best
done by monitoring existing health care consumption registers. Attention must
be paid to the interpretation of changing consumption rates in relation to
changes in health care recruitment and admission policy. When the adequacy or
accuracy of case definition and registration is questioned, efforts are needed to
improve registration.
For the purpose of etiological model building, the mere determination of
prevalence and incidence rates is not enough. Although more is becoming
known on general and specific risk factors for the onset of an eating disorder
there still is an impressive gap. Furthermore, the developmental mechanisms of
these factors are largely unknown. The general conclusion is that dieting
EPIDEMIOLOGY 89
behavior plays a role in the pathogenesis of anorexia nervosa, bulimia nervosa,

and BED. However, not all dieters proceed to develop an eating disorder. A
prospective follow-up study of initially healthy dieters sampled from the general
population may shed light on the mechanisms that turn dieting into an eating
disorder.
To circumvent the power and cost problems caused by the relatively low
rates of eating disorders in the general population, a few suggestions for the
design of economically feasible studies providing generalizable, reliable results
on risk factors and mechanisms are given:
There is a need for prospective, follow-up designs using initially healthy

subjects at high risk for developing an eating disorder, such as young
girls, dieters, and participants in weight-restricted sports, including
ballet. Depending on the question to be answered, these could be
matched on sex, age, and socio-economic status with initially healthy
intermediate-and low-risk groups.
For lower risk groups, such as males or older persons, a prospective
design is too cost inefficient and a case-control design is more
appropriate. Cases should be collected at as low a level of entry into the
health care system as possible, preferably primary care. Same-sex siblings
and other same-sex persons matched for age-and socio-economic status
could be of use as controls. To facilitate hypothesis testing and the
exchange of knowledge and ideas, the formation of a multicenter
database of these rare cases would mean a great improvement.
For both prospective follow-up studies and case-control studies of
eating disorders, a comprehensive assessment of biological,
psychological, familial, and social variables is needed. The factors and
mechanisms studied should be based on findings from previous research
such as that by Fairburn et al. (12). To decide on the effect of weight-and
shape-centered beauty ideals on the frequency of eating disorders, studies
are needed that compare the distribution of eating disorders between
groups differing in weight- and shape-related attitudes.
Finally, an issue to be solved for epidemiological studies on eating disorders is

the reliance on a categorical approach of caseness, particularly for the “newer”
diagnoses of bulimia nervosa, the ED-NOS, and BED. By focusing on incident
clinical cases and ignoring atypical or subclinical cases, etiological reasoning may
miss the crucial developmental elements in what have been called “broad-
spectrum” disorders.
90 RÅSTAM ET AL.
SUMMARY
The average prevalence rate for young girls is 0.3% for anorexia nervosa and 1%
for bulimia nervosa. The overall incidence is at least 8 per 100,000 person-years
for anorexia nervosa and 12 per 100,000 person-years for bulimia nervosa.
The incidence rate of anorexia nervosa has increased during the past
50 years, particularly in females 10–24 years old. The registered incidence of
bulimia nervosa has increased, at least during the first 5 years after bulimia
nervosa was introduced in the DSM-III.
There is a need for prospective, follow-up designs using initially healthy
subjects at high risk for developing an eating disorder. Depending on the
question to be answered, these could be matched on sex, age, and
socioeconomic status with initially healthy intermediate-and low-risk groups.
Detailed and reliable registration of case definition, demographic, and other
characteristics of the patient, symptoms, and concomitants of the dis ease or
disorder remain of the utmost importance to advance evidence-based treatment
and prevention.
REFERENCES
1. Wentz E, Gillberg C, Gillberg IC, Råstam M. Ten-year follow-up of adolescent-
onset anorexia nervosa: psychiatric disorders and overall functioning scales. J Child
Psychol Psychiatry 2001; 42:613–622.
2. Szmukler GI. The epidemiology of anorexia nervosa and bulimia. J Psychiatr Res
1985; 19:143–153.
3. Hoek HW, van Hoeken D. Review of the prevalence and incidence of eating
disorders. Int J Eat Disord 2003; 34:383–396.
4. Hsu LKG. Epidemiology of the eating disorders. Psychiatric Clin North Am 1996;
19:681–700.
5. Williams P, Tarnopolsky A, Hand D. Case definition and case identification in
psychiatric epidemiology: review and assessment. Psychol Med 1980; 10:101–114.
6. Fairburn CG, Beglin SJ. Studies of the epidemiology of bulimia nervosa. Am J
Psychiatry 1990; 147:401–408.
8. van Hoeken D, Lucas AR, Hoek HW. Epidemiology. In: Hoek HW, Treasure JL,
Katzman MA, eds. Neurobiology in the Treatment of Eating Disorders.
Chichester: John Wiley and Sons, 1998:97–126.
9. Hoek HW, van Hoeken D, Katzman MA. Epidemiology and cultural aspects of
eating disorders: a review. In: Maj M, Halmi K, Lopez-Ibor JJ, Sartorius N, eds.
Eating Disorders. Vol. 6. Chichester: John Wiley and Sons, 2003:75–104.
EPIDEMIOLOGY 91
10. Patton GC, Johnson-Sabine E, Wood K, Mann AH, Wakeling A. Abnormal eating
attitudes in London schoolgirls—a prospective epidemiological study: outcome at
twelve month follow-up. Psychol Med 1990; 20:383–394.
11. Patton GC, Selzer R, Coffey C, Carlin JB, Wolfe R. Onset of adolescent eating
disorders: population based cohort study over 3 years. BMJ 1999; 318:765–768.
12. Fairburn CG, Welch SL, Doll HA, Davies BA, O’Connor ME. Risk factors for
bulimia nervosa: a community-based case-control study. Arch Gen Psychiatry 1997;
54:509–517.
13. Fairburn CG, Doll HA, Welch SL, Hay PJ, Davies BA, O’Connor ME. Risk factors
for binge eating disorder: a community-based case-control study. Arch Gen
Psychiatry 1998; 55:425–432.
14. Fairburn CG, Cooper Z, Doll HA, Welch SL. Risk factors for anorexia nervosa:
three integrated case-control comparisons. Arch Gen Psychiatry 1999; 56:
468–476.
15. Johnson JG, Cohen P, Kasen S, Brook JS. Eating disorders during adolescence and
the risk for physical and mental disorders during early adulthood. Arch Gen
Psychiatry 2002; 59:545–552.
16. Johannisson K. The Dark Continent. Stockholm: Nordstedts Förlag, 1994: 131.
17. Hill AJ, Draper E, Stack J. A weight on children’s minds: body shape
dissatisfactions at 9-years old. Int J Obes Relat Metab Disord 1994; 18:383–389.
18. Maloney MJ, McGuire J, Daniels SR, Specker B. Dieting behavior and eating
attitudes in children. Pediatrics 1989; 84:482–489.
19. Sasson A, Lewin C, Roth D. Dieting behavior and eating attitudes in Israeli
children. Int J Eat Disord 1995; 17:67–72.
20. Jones JM, Bennett S, Olmsted MP, Lawson ML, Rodin G. Disordered eating
attitudes and behaviours in teenaged girls: a school-based study. CMAJ 2001; 165:
547–552.
21. Whitaker A, Davies M, Shaffer D, Johnson J, Abrams S, Walsh BT, Kalikow K.
The struggle to be thin: a survey of anorexic and bulimic symptoms in a non-
referred adolescent population. Psychol Med 1989; 19:143–163.
22. Childress AC, Brewerton TD, Hodges EL, Jarrell MP. The Kids’ Eating Disorders
Survey (KEDS): a study of middle school students. J Am Acad Child Adol
Psychiatry 1993; 32:843–850.
23. Steinhausen HC, Winkler C, Meier M. Eating disorders in adolescence in a Swiss
epidemiological study. Int J Eat Disord 1997; 22:147–151.
24. Nakamura K, Hoshino Y, Watanabe A, Honda K, Niwa S, Tominaga K, Shimai S,
Yamamoto M. Eating problems in female Japanese high school students: a
prevalence study. Int J Eat Disord 1999; 26:91–95.
25. Nobakht M, Dezhkam M. An epidemiological study of eating disorders in Iran. Int
J Eat Disord 2000; 28:265–271.
26. Koskelainen M, Sourander A, Helenius H. Dieting and weight concerns among
Finnish adolescents. Nord J Psychiatry 2001; 55:427–431.
27. Carter JC, Stewart DA, Fairburn CG. Eating disorder examination questionnaire:
norms for young adolescent girls. Behav Res Ther 2001; 39:625–632.
92 RÅSTAM ET AL.
28. Nylander I. The feeling of being fat and dieting in a school population. An
epidemiologic interview investigation. Acta Sociomed Scand 1971; 3:17–26.
29. Schleimer K. Dieting in teenage schoolgirls. A longitudinal prospective study. Acta
Paediatr Scand Suppl 1983; 312:1–54.
30. Irving LM, Wall M, Neumark-Sztainer D, Story M. Steroid use among adolescents:
findings from Project EAT. J Adol Health 2002; 30:243–252.
31. Labre MP. Adolescent boys and the muscular male body ideal. J Adol Health 2002;
30:233–242.
32. Lau B, Alsaker FD. Dieting behavior in Norwegian adolescents. Scand J Psychol
2001; 42:25–32.
33. Perry CL, McGuire MT, Neumark-Sztainer D, Story M. Characteristics of
vegetarian adolescents in a multiethnic urban population. J Adol Health 2001; 29:
406–416.
34. Råstam M. Anorexia nervosa in 51 Swedish adolescents: premorbid problems and
comorbidity. J Am Acad Child Adol Psychiatry 1992; 31:819–829.
35. Button EJ, Whitehouse A. Subclinical anorexia nervosa. Psychol Med 1981; 11:
509–516.
36. Szmukler GI. Weight and Food Preoccupation in a Population of English
Schoolgirls. Columbus, Ohio: Ross, 1983.
37. King MB. Eating disorders in a general practice population. Prevalence,
characteristics and follow-up at 12 to 18 months. Psychol Med Monogr Suppl
1989; 14:1–34.
38. Meadows GN, Palmer RL, Newball EUM, Kenrick JMT. Eating attitudes and
disorder in young women: a general practice based survey. Psychol Med 1986; 16:
351–357.
39. Johnson-Sabine E, Wood K, Patton G, Mann A, Wakeling A. Abnormal eating
attitudes in London schoolgirls—a prospective epidemiological study: factors
associated with abnormal response on screening questionnaires. Psychol Med 1988;
18:615–622.
40. Råstam M, Gillberg C, Garton M. Anorexia nervosa in a Swedish urban region. A
population-based study. Br J Psychiatry 1989; 155:642–646.
41. Whitaker A, Johnson J, Shaffer D, Rapoport JL, Kalikow K, Walsh BT, Davies M,
Braiman S, Dolinsky A. Uncommon troubles in young people: prevalence
estimates of selected psychiatric disorders in a nonreferred adolescent population.
Arch Gen Psychiatry 1990; 47:487–496.
42. Whitehouse AM, Cooper PJ, Vize CV, Hill C, Vogel L. Prevalence of eating
disorders in three Cambridge general practices: hidden and conspicuous morbidity.
Br J Gen Pract 1992; 42:57–60.
43. Rathner G, Messner K. Detection of eating disorders in a small rural town: an
epidemiological study. Psychol Med 1993; 23:175–184.
44. Wlodarçzyk-Bisaga K, Dolan B. A two-stage epidemiological study of abnormal
eating attitudes and their prospective risk factors in Polish schoolgirls. Psychol Med
1996; 26:1021–1032.
EPIDEMIOLOGY 93
45. Gual P, Perez-Gaspar M, Martinez-Gonzalez MA, Lahortiga F, de IralaEstevez J,

Cervera-Enguix S. Self-esteem, personality, and eating disorders: baseline
assessment of a prospective population-based cohort. Int J Eat Disord 2002; 31:
261–273.
46. Hoek HW. The incidence and prevalence of anorexia nervosa and bulimia nervosa
in primary care. Psychol Med 1991; 21:455–460.
47. Lucas AR, Beard CM, O’Fallon WM, Kurland LT. 50-year trends in the incidence
of anorexia nervosa in Rochester, Minn.: a population-based study. Am J
Psychiatry 1991; 148:917–922.
48. Lucas AR, Crowson CS, O’Fallon WM, Melton LJ III. The ups and downs of
anorexia nervosa. Int J Eat Disord 1999; 26:397–405.
49. Eaton WW, Tien AY, Poeschla BD. Epidemiology of schizophrenia. In: den Boer
JA, Westenberg HGM, van Praag HM, eds. Advances in the Neurobiology of
Schizophrenia. Chichester: John Wiley and Sons, 1995:27–57.
50. Theander S. Anorexia nervosa. A psychiatric investigation of 94 female patients.
Acta Psychiatr Scand Suppl 1970; 214:1–194.
51. Willi J, Grossmann S. Epidemiology of anorexia nervosa in a defined region of
Switzerland. Am J Psychiatry 1983; 140:564–567.
52. Willi J, Giacometti G, Limacher B. Update on the epidemiology of
anorexia nervosa in a defined region of Switzerland. Am J Psychiatry 1990; 147:
1514–1517.
53. Jones DJ, Fox MM, Babigian HM, Hutton HE. Epidemiology of anorexia nervosa
in Monroe County, New York: 1960–1976. Psychosom Med 1980; 42:551–558.
54. Kendell RE, Hall DJ, Hailey A, Babigian HM. The epidemiology of anorexia
nervosa. Psychol Med 1973; 3:200–203.
55. Szmukler G, McCance C, McCrone L, Hunter D. Anorexia nervosa: a psychiatric
case register study from Aberdeen. Psychol Med 1986; 16:49–58.
56. Hoek HW, Brook FG. Patterns of care of anorexia nervosa. J Psychiatr Res 1985;
19:155–160.
57. Møller-Madsen S, Nystrup J. Incidence of anorexia nervosa in Denmark. Acta
Psychiatr Scand 1992; 86:197–200.
58. Hoek HW, Bartelds AIM, Bosveld JJF, van der Graaf Y, Limpens VEL, Maiwald
M, Spaaij CJK. Impact of urbanization on detection rates of eating disorders. Am J
Psychiatry 1995; 152:1272–1278.
59. Turnbull S, Ward A, Treasure J, Jick H, Derby L. The demand for eating disorder
care. An epidemiological study using the General Practice Research Database. Br J
Psychiatry 1996; 169:705–712.
60. Crisp AH, Palmer RL, Kalucy RS. How common is anorexia nervosa? A prevalence
study. Br J Psychiatry 1976; 128:549–554.
61. Casper RC. Introduction to special issue. J Youth Adol 1996; 25:413–418.
62. Swenne I. Changes in body weight and body mass index (BMI) in teenage girls
prior to the onset and diagnosis of an eating disorder. Acta Paediatrica 2001; 90:
677–681.
94 RÅSTAM ET AL.
63. Young JK. Estrogen and the etiology of anorexia nervosa. Neurosci Biobehav Rev
1991; 15:327–331.
64. Kaplowitz PB, Slora EJ, Wasserman RC, Pedlow SE, Herman-Giddens ME. Earlier
onset of puberty in girls: relation to increased body mass index and race. Pediatrics
2001; 108:347–353.
65. Wakeling A. Epidemiology of anorexia nervosa. Psychiatry Res 1996; 62:3–9.
66. Williams P, King M. The “epidemic” of anorexia nervosa: another medical myth?
Lancet 1987; 1:205–207.
67. Fombonne E. Anorexia nervosa. No evidence of an increase. Br J Psychiatry 1995;
166:462–471.
68. Gillberg C, Råstam M, Gillberg IC. Anorexia nervosa: who sees the patients and
who do the patients see? Acta Paediatrica 1994; 83:967–971.
69. Bushnell JA, Wells JE, Hornblow AR, Oakley-Browne MA, Joyce P. Prevalence of
three bulimia syndromes in the general population. Psychol Med 1990; 20:
671–680.
70. Szabo P, Tury F. The prevalence of bulimia nervosa in a Hungarian college and
secondary school population. Psychother Psychosom 1991; 56:43–47.
71. Santonastaso P, Zanetti T, Sala A, Favaretto G, Vidotto G, Favaro A. Prevalence of
eating disorders in Italy: a survey on a sample of 16-year-old female students.
Psychother Psychosom 1996; 65:158–162.
Disorders. 3d ed. Washington, DC,1980.
Disorders. 3d ed. revised. Washington, DC, 1987.
74. Garfinkel PE, Lin E, Goering P, Spegg C, Goldbloom DS, Kennedy S, Kaplan AS,
Woodside DB. Bulimia nervosa in a Canadian community sample: prevalence and
comparison of subgroups. Am J Psychiatry 1995; 152:1052–1058.
75. Dansky BS, Brewerton TD, Kilpatrick DG, O’Neil PM. The National Women’s
Study: relationship of victimization and posttraumatic stress disorder to bulimia
76. Soundy TJ, Lucas AR, Suman VJ, Melton LJ III. Bulimia nervosa in Rochester,
Minnesota from 1980 to 1990. Psychol Med 1995; 25:1065–1071.
77. World Health Organization. The ICD-10 Classification of Mental and Behavioural
Disorders: Clinial Descriptions and Diagnostic Guidelines. Geneva, 1992.
78. World Health Organization. The ICD-9 Classification of Mental and Behavioral
Disorders. Clinical Descriptions and Diagnostic Guidelines. Geneva, 1979.
79. Russell G. Bulimia nervosa: an ominous variant of anorexia nervosa. Psychol Med
1979; 9:429–448.
80. Pagsberg AK, Wang AR. Epidemiology of anorexia nervosa and bulimia nervosa in
Bornholm County, Denmark, 1970–1989. Acta Psychiatr Scand 1994; 90:
259–265.
81. Schwitzer AM, Rodriguez LE, Thomas C, Salimi L. The eating disorders NOS
diagnostic profile among college women. J Am Coll Health 2001; 49:157–166.
EPIDEMIOLOGY 95
82. Striegel-Moore RH, Cachelin FM, Dohm FA, Pike KM, Wilfley DE, Fairburn CG.
Comparisonof binge eating disorder and bulimia nervosa in a community sample.
Int J Eat Disord 2001; 29:157–165.
83. Hay P, Fairburn C. The validity of the DSM-IV scheme for classifying bulimic
eating disorders. Int J Eat Disord 1998; 23:7–15.
84. Cotrufo P, Barretta V, Monteleone P, Maj M. Full-syndrome, partialsyndrome and
subclinical eating disorders: an epidemiological study of female students in
Southern Italy. Acta Psychiatr Scand 1998; 98:112–115.
85. Hay P. The epidemiology of eating disorder behaviors: an Australian community-
based survey. Int J Eat Disord 1998; 23:371–382.
86. Fairburn CG, Cooper Z. The Eating Disorder Examination. 12th ed. In: Fairburn
87. Westenhoefer J. Prevalence of eating disorders and weight control practices in
Germany in 1990 and 1997. Int J Eat Disord 2001; 29:477–481.
88. Dohrenwend BP, Levav I, Shrout PE, Schwartz S, Naveh G, Link BG, Skodol AE,
Stueve A. Socioeconomic status and psychiatric disorders: the causationselection
issue. Science 1992; 255:946–952.
89. Gard MCE, Freeman CP. The dismantling of a myth: a review of eating disorders
and socioeconomic status. Int J Eat Disord 1996; 20:1–12.
90. McClelland L, Crisp A. Anorexia nervosa and social class. Int J Eat Disord 2001;
29:150–156.
91. Råstam M, Gillberg C. The family background in anorexia nervosa: a population-
based study. J Am Acad Child Adol Psychiatry 1991; 30:283–289.
92. Tsai G. Eating disorders in the Far East. Eat Weight Disord 2000; 5:183–197.
93. Baska T, Straka S, Madar R. Smoking and some life-style changes in medical
students—Slovakia, 1995–1999. Cent Eur J Public Health 2001; 9:147–149.
94. Abdollahi P, Mann T. Eating disorder symptoms and body image concerns in Iran:
comparisons between Iranian women in Iran and in America. Int J Eat Disord
2001; 30:259–268.
95. Nasser M. Screening for abnormal eating attitudes in a population of Egyptian
secondary school girls. Soc Psychiatry Psychiatr Epidemiol 1994; 29:25–30.
96. Nasser M. Comparative study of the prevalence of abnormal eating attitudes among
Arab female students of both London and Cairo Universities. Psychol Med 1986;
16:621–625.
97. Dolan B, Lacey JH, Evans C. Eating behaviour and attitudes to weight and shape in
British women from three ethnic groups. Br J Psychiatry 1990; 157:523–528.
98. Nasser M. Culture and Weight Consciousness. London: Routledge, 1997.
99. Nasser M, Katzman MA, Gordon RA, eds. Eating disorders and cultures in
transition. London: Routledge, 2001.
100. Hoek HW, van Harten PN, van Hoeken D, Susser E. Lack of relation between
culture and anorexia nervosa: results of an incidence study on Curaçao. N Engl J
Med 1998; 338:1231–1232.
96 RÅSTAM ET AL.
101. Vandereycken W, Hoek HW. Are eating disorders culture-bound syndromes? In:
Halmi KA, ed. Psychobiology and Treatment of Anorexia Nervosa and Bulimia
Nervosa. Washington, DC: American Psychiatric Press, 1993:19–36.
102. Santonastaso P, Mondini S, Favaro A. Are fashion models a group at risk for eating
disorders and substance abuse? Psychother Psychosom 2002; 71:168–172.
103. Smolak L, Murnen SK, Ruble AE. Female athletes and eating problems: a meta-
analysis. Int J Eat Disord 2000; 27:371–380.
104. Hulley AJ, Hill AJ. Eating disorders and health in elite women distance runners.
Int J Eat Disord 2001; 30:312–317.
5
Long-Term Outcome, Course of Illness and
Mortality in Anorexia Nervosa, Bulimia
Pamela K.Keel
Harvard University
Cambridge, Massachusetts, U.S.A.
David B.Herzog
Massachusetts General Hospital
Boston, Massachusetts, U.S.A.
Research on the long-term course and outcome of eating disorders (anorexia

nervosa, bulimia nervosa, and binge eating disorder) has the potential to
improve the treatment, diagnosis, and understanding of these disorders. In
order to evaluate the efficacy of treatment interventions it is important to
understand both the natural course of illness and the course following
intervention. Distinctions between disorders can be revealed by differential
treatment response as well as differences in course and outcome over time.
Finally, accurate distinctions between disorders will ultimately contribute to
revealing their etiology. For example, attempts to reveal the underlying genetic
bases of eating disorders require the accurate distinction of phenotypes. The
purpose of this chapter is to review and compare data concerning the long-term
course and outcome of anorexia nervosa, bulimia nervosa, and binge eating
disorder across the following outcome domains: mortality, recovery, relapse,
cross-over, prognostic variables, and treatment utilization. Because of
differences in when specific eating disorders have been introduced to the
psychiatric nomenclature (ranging from 1873 to 1994), significant discrepancies
exist in the amount of data available to describe outcome for anorexia nervosa,
bulimia nervosa, and binge eating disorder.
98 KEEL AND HERZOG
ANOREXIA NERVOSA
Mortality
Anorexia nervosa has been associated with one of the highest risks of premature
death among psychiatric disorders (1). Estimates of crude mortality rates across
outcome studies suggest that 5–5.9% of patients diagnosed with anorexia
nervosa will suffer a fatal outcome (2,3), with a 5.6% crude mortality rate per
decade (4). Standardized mortality ratios (representing the ratio of the observed
number of deaths to the expected number of deaths in a matched population)
have ranged from 1.32 (4b) to 12.82 (4c) with most studies reporting
significantly elevated risk of premature death. Earlier studies tended to report
starvation as the cause of death; conversely, more recent studies have reported
suicide as a leading cause of death in anorexia nervosa (5). This shift may reflect
several coinciding secular trends. First, the diagnostic criteria for anorexia
nervosa shifted from requiring a loss of 25% of prior body weight to weight 15%
below that expected for age and height. For most individuals who develop
anorexia nervosa during adolescence, the shift effectively decreased the weight
loss required to be diagnosed with anorexia nervosa. Similarly, epidemiological
data have demonstrated that weight of patients presenting for treatment of
anorexia nervosa has increased over time (6), reflecting both the change in
diagnostic criteria and the possibility of earlier intervention. Finally, there may
have been secular improvements in techniques employed to refeed
undernourished patients. Few predictors of fatal outcome have been revealed.
These have included low weight (7,8), poor psychosocial function (9,10),
longer duration of follow-up (2,10), and severity of alcohol use disorders (10).
Recovery
Although longer duration of follow-up has been associated with increased risk of
death, it is also associated with higher rates of recovery (2,11). Prospective,
longitudinal analyses of anorexia nervosa course suggest that recovery increases
at a slow and steady rate, with continued recovery occurring years after
intake (11) or treatment (12). Collapsing recovery rates across studies of
varying durations of follow-up, Steinhausen (2) reported that approximately
46% of patients recover, 33% improve (but remain symptomatic), and 20%
remain chronically ill.
LONG-TERM OUTCOME 99
Relapse
Unlike mortality and recovery, which can be investigated in outcome studies,
relapse can only be assessed by studies of eating disorder course. Thus, fewer
studies have elucidated this important variable. Morgan and Russell (13)
reported that 51 % of patients hospitalized for anorexia nervosa required
readmission over the course of follow-up. Herzog and colleagues (11) reported
that 40% of women with anorexia nervosa who achieve full recovery later
relapsed. Finally, Strober et al. (12) reported that approximately 30% of
women who had achieved weight recovery during hospitalization relapsed after
discharge, and they recorded even lower rates of relapse among women
considered partially recovered (9.8%) or fully recovered (0%) over follow-up.
These latter data indicate that the definition of recovery (or remission) has a
significant effect on the likelihood of relapse. A common pattern among patients
with anorexia nervosa is to experience improvement in weight (a sign of
recovery) as a consequence of the development of binge eating episodes (a sign
of cross-over).
Cross-over
A recent report from a prospective longitudinal study of eating disorders
indicated that the majority of women with the restricting subtype of anorexia
nervosa develop symptoms of binge eating and purging over time (14). These
findings are consistent with the restraint hypothesis (15), which posits that
dietary restriction increases susceptibility to develop binge eating episodes. For
some patients with anorexia nervosa, binge eating and purging occur at low
weight and the person continues to suffer from anorexia nervosa. For others,
binge eating results in weight gain and the binge-purge behaviors are maintained
at normal weight representing cross-over from anorexia nervosa to bulimia
nervosa. Across follow-up studies, approximately 10–15% of individuals
presenting with anorexia nervosa cross over and develop bulimia nervosa(16).
Prognostic Factors
Table 1 presents prognostic factors adapted from a recent review of outcome in
anorexia nervosa (2). As can be seen from Table 1, no prognostic factor has
been unambiguously associated with prognosis in anorexia nervosa. However,
there is a clear trend in the nature of the contradictory results. Studies finding a
specific direction of association are most often contradicted by studies finding
no significant association rather than an equal number of studies finding an
association in the opposite direction. For example, 12 studies have suggested
100 KEEL AND HERZOG
TABLE 1 Prognostic Indicators of Favorable Course or Outcome in Anorexia Nervosa
Note: Data are adapted from Steinhausen (2) to allow comparison with data from Keel
and Mitchell (19) updated in Table 2.
that an older age of onset is associated with worse prognosis in anorexia

nervosa, and 14 studies suggest no significant association. Conversely, only 2
suggest that an older age of onset is associated with favorable prognosis in
anorexia nervosa. Similarly, 14 studies suggest that a longer duration of
symptoms prior to presentation is associated with poor prognosis in anorexia
nervosa, and 7 studies suggest no significant association. Conversely, no studies
suggest that a longer duration of symptoms is associated with favorable
prognosis in anorexia nervosa. This pattern suggests that contradictory findings
are most likely attributable to differences in statistical power among studies.
One seeming contradiction among prognostic indicators is the association
between age of onset and duration of symptoms. If one develops an eating
disorder at a younger age, then one may have better prognosis. However, if the
duration of symptoms prior to presentation is longer, then this is associated with
worse prognosis. These findings may reveal the impact of delay in seeking help.
That is, an early age of onset is predictive of better course and outcome only if
there is a short delay before seeking treatment.
Treatment Utilization
In Germany, Fichter and colleagues (17) reported that 63% of women received
further inpatient treatment following baseline inpatient care for anorexia
nervosa. Levels of outpatient care were also high, with approximately 89% of
women receiving some form of outpatient care during follow-up (among a
subset of women for whom data were available). This treatment comprised an
average of 69.3 individuals sessions, 5.0 group sessions, and 0.4 family session.
In the United States, Striegel-Moore et al. (18) reported that female patients
with anorexia nervosa utilized an average of 26.0 days of inpatient care and
17.0 days of outpatient care during one year. Male patients utilized an average of
15.6 and 9.2 days of inpatient and outpatient care per year, respectively.
BULIMIA NERVOSA
Mortality
A crude mortality rate of 0.3% has been reported across follow-up studies of
bulimia nervosa (19). However, this value might represent an underestimate
given the relatively short duration of follow-up across studies and the low
ascertainment rates for several studies. Few studies have calculated standardized
mortality ratios for bulimia nervosa, and no original report has found a
standardized mortality ratio that differed significantly from 1. In a recent meta-
analysis of mortality in eating disorders, Nielsen (4) reported a significantly
elevated SMR for bulimia nervosa. However, this analysis included studies
reporting mortality among “bulimic patients” comprising both normal-weight
bulimia nervosa and the binge-purge subtype of anorexia nervosa. In a recent
examination of mortality in eating disorders, Keel et al. (10) found that
mortality was similarly elevated for both the restricting and binge-purge
subtypes of anorexia nervosa but was not elevated in bulimia nervosa. As the
ascertainment methods and duration of follow-up were uniform across
diagnostic categories, this study provides the strongest evidence that bulimia
nervosa is not associated with increased risk of premature death. Causes of
premature death have primarily included suicide and automobile accidents (19).
Because risk of death does not appear to be elevated in bulimia nervosa, no
predictors of fatal outcome have been revealed.
Recovery
Prospective, longitudinal analyses of bulimia nervosa course (11,20) suggest
that the majority of patients recover from their eating disorder at some point
during the course of follow-up. Combining results across outcome studies of
varying durations of follow-up, Keel and Mitchell (19) reported that
approximately 50% of individuals presenting with bulimia nervosa recover and
maintain their recovery, 30% are improved but maintain partial syndromes, and
20% continue to meet full criteria for bulimia nervosa. Follow-up studies of
longer duration (21,22) suggest that rates of full bulimia nervosa drop to 10%
by 10 years follow-up.
102 KEEL AND HERZOG
Relapse
Rates of relapse have ranged from 26% to 50% across follow-up studies of
bulimia nervosa (19,23,24), with definitions of “recovery” explaining the
majority of variance in these estimates. Similar to the patterns observed with
anorexia nervosa, more stringent definitions of recovery are associated with lower
relapse rates than less stringent definitions (25,26). This pattern raises the
question of whether studies are assessing relapse or simple symptom
fluctuation. Despite considerable differences in methodology across studies,
relapse rates converge around 30% (19), suggesting that approximately onethird
of women who initially recover from their eating disorder suffer a resurgence of
their bulimic symptoms.
Crossover
Across follow-up studies, between 0% and 7% crossover from bulimia nervosa
to anorexia nervosa (19,24). However, it remains unclear whether these
represent new-onset cases of anorexia nervosa or relapse from previous
episodes of anorexia nervosa that had crossed over to bulimia nervosa.
Crossover to binge eating disorder also seems low, ranging from 0% to
1.1% (19,21,24). A recent paper concerning the predictive validity of bulimia
nervosa as a diagnostic category (27) reported that over time women with bulimia
nervosa were more likely to continue to suffer from bulimia nervosa than the
other recognized eating disorders of the DSM. Although approximately 10%
reported a brief history of binge eating in the absence of inappropriate
compensatory behavior, all but one of these women experienced full remission
of her eating disorder, suggesting that this symptom pattern represented a phase
of recovery rather than development of a distinct eating disorder. Conversely,
approximately 10% of women suffered from a purging disorder, characterized
by recurrent purging in the absence of objectively large binge episodes at normal
weight. The likelihood of suffering from a purging disorder did not differ
significantly from the likelihood of suffering from bulimia nervosa at long-term
follow-up and was significantly more likely than suffering from binge eating
disorder (27). Thus, to the extent that cross-over occurs in bulimia nervosa, it
seems to involve crossing over to a purging disorder.
Prognostic Factors
Table 2 presents prognostic factors updated from a recent review of outcome in
bulimia nervosa (19). As can be seen from the Table 2, few prognostic factors
have been replicated across studies. Similar to the pattern observed in Table 1,
TABLE 2 Prognostic Indicators of Favorable Course or Outcome in Bulimia Nervosa
Note: Data are updated from Keel and Mitchell (19) with findings from studies of follow-
up duration 5 years (21,24,43).
most often findings diverge along the lines of a significant association in one
direction versus no significant association. For example, six studies reported that
longer duration of symptoms prior to presentation was associated with poor
prognosis in bulimia nervosa, and three studies found no significant association.
However, no studies reported that a longer duration of symptoms was
associated with favorable prognosis in bulimia nervosa. Similarly, six studies
suggested that personality disturbance was associated with poor prognosis, and
four studies found no significant association, but no studies reported that
personality disturbances were associated with favorable outcome. Despite this
pattern for duration of symptoms and personality disturbance, the number of
studies suggesting no significant associations is greater than the number of
studies suggesting significant associations for the remaining variables, including
the impact of treatment. However, a recent investigation (20) has demonstrated
an association between baseline treatment and long-term course in bulimia
nervosa, suggesting that cognitive behavioral therapy may speed recovery (28).
Treatment Utilization
Although most women who suffer from bulimia nervosa may never seek
treatment (23,29), treatment utilization is high among those who do. In
Germany, Fichter et al. (24) reported that 67% of women received further
inpatient treatment following baseline inpatient care for bulimia nervosa. Levels
of outpatient care were also high, with approximately 83% of women receiving
some form of outpatient care during follow-up (among a subset of women for
whom data were available). This treatment comprised an average of 59.2
individuals sessions, 10.9 group sessions, and 0.7 family session (24). In the
United States, Striegel-Moore et al. (18) reported that female patients with
bulimia nervosa utilized an average of 14.7 days of inpatient care and 15.6 days
104 KEEL AND HERZOG
of outpatient care during one year. Male patients utilized an average of 21.7 and
9.1 days of inpatient and outpatient care per year, respectively. Although
inpatient care did not differ between female and male patients, women received
significantly more outpatient care (18).
BINGE EATING DISORDER

Since the introduction of binge eating disorder in the DSM-IV (30), a number of
treatment studies have been published and several have provided short-term and
intermediate follow-up data (31–36). However, relatively little is known
concerning long-term course and outcome. At this point, two studies (23,37)
have characterized course and outcome for this disorder at 5 or more years
following presentation. Because of the limited number of long-term follow-up
studies, the results of these two studies will be reviewed individually below.
Fichter et al. (37) reported one death among 68 patients treated for binge
eating disorder at 6 year follow-up for a crude mortality rate of 1.5%. Fairburn
et al. (23) reported no deaths in his cohort of 48 individuals with binge eating
disorder recruited from the community. Rates of full binge eating disorder at
follow-up were 4% (23) and 5.9% (37), with 57.4% (37) to 82% (23) reported
as improved or recovered. Fairburn et al. (23) reported that relapse occurred in
4–10% of recovered individuals across various points of follow-up. Cross-over
to anorexia nervosa did not occur, and cross-over to bulimia nervosa was rare,
ranging from 3% (23) to 7.4% (37). Neither article reported on predictors of
long-term course or outcome. Fichter and colleagues (37) reported that 66% of
the binge eating disorder patients received further inpatient treatment following
baseline inpatient care for binge eating disorder. Conversely, Fairburn and
colleagues (23) reported that as few as 3% of individuals with binge eating
disorder received treatment for an eating disorder during the course of
follow-up, resulting in a total of 8% ever having received treatment.
COMPARISON OF EATING DISORDERS
Anorexia Nervosa and Bulimia Nervosa

Few studies have been designed to evaluate the course or outcome of more than
one eating disorder. Thus, comparisons of anorexia nervosa and bulimia nervosa
are significantly limited by methodological differences across studies. However,
certain similarities and differences emerge despite this limitation.
Both within studies (10,17) and across studies (2,19), anorexia nervosa
appears to be associated with significantly increased risk of premature death and
bulimia nervosa is not. Suicide has been noted as a common cause of death for
both disorders. Interestingly, rates and lethality of suicide attempts have not
differed between women with anorexia nervosa versus bulimia nervosa (38).
Thus, suicide attempts appear to be more likely to result in death among
women suffering from anorexia nervosa. One possible explanation for this
difference is that starvation compromises the health of individuals with anorexia
nervosa, leaving them less likely to survive a suicide attempt.
Within studies (11,17), anorexia nervosa has been associated with both lower
and slower rates of recovery in comparison with bulimia nervosa. However,
collapsing results across studies suggests a similar distribution of recovery,
improvement, and chronicity across syndromes (2,19). However, this latter
comparison is flawed because follow-up studies for anorexia nervosa have a
significantly longer duration of follow-up in comparison with follow-up studies
of bulimia nervosa, and longer duration of follow-up is associated with higher rates
of recovery for both syndromes (2,21). Thus, the apparent similarity in results
between disorders across studies is likely an artifact of differences in duration of
follow-up.
Despite differing rates of recovery, relapse rates were surprisingly similar
between anorexia nervosa and bulimia nervosa within one study of longterm
course (21). However, the relationship between relapse and chronicity likely
differs between disorders. Because a higher percentage of women with bulimia
nervosa than women with anorexia nervosa recover at some point during the
course of follow-up, a higher percentage of women who are ill at follow-up in
bulimia nervosa outcome studies represent women who have recovered and
relapsed. Thus, for women with bulimia nervosa, chronicity appears to be
characterized by considerable symptom fluctuation. Conversely, a smaller
percentage of women recover from anorexia nervosa; thus, a smaller
percentage of women who remain ill are women who achieved even brief
periods of remission. Thus, chronicity within anorexia nervosa is marked by a
more steady course of illness. This difference may also be reflected in the nature
of the core features of these syndromes. For anorexia nervosa, the core
symptom represents the absence of a behavior—eating. Conversely, for bulimia
nervosa, the core symptoms represent the presence of a set of behaviors—binge
eating and purging. Thus, in order to develop anorexia nervosa an individual
must demonstrate persistence in not eating. Conversely, a woman can develop
bulimia nervosa by having relatively normal eating behaviors interrupted by
episodes of abnormal eating behaviors. For these reasons, the nature of a
chronic course may differ markedly between these disorders.
Patterns of cross-over also differ markedly between anorexia nervosa and
bulimia nervosa. Approximately 10–50% of women with anorexia nervosa
cross over and develop bulimia nervosa, and approximately 30% of women with
106 KEEL AND HERZOG
bulimia nervosa report a prior history of anorexia nervosa. Conversely, only

0–7% of women with bulimia nervosa have been reported to cross over and
develop anorexia nervosa. This pattern likely reflects both the psychological and
physiological consequences of the core features of each disorder. Specifically,
intense dietary restriction and weight loss likely increase risk for developing
binge eating episodes. Conversely, the presence of objectively large binge eating
episodes likely protects against low weight. Indeed, comparisons of the subtypes
of anorexia nervosa suggest that binge eating is associated with higher weight.
We found better evidence for prognostic factors for anorexia nervosa than
for bulimia nervosa. However, this is likely a function of the difference in the
number and size of outcome studies for the two disorders. An early age of onset
was associated with improved prognosis in anorexia nervosa but was
unassociated with prognosis in bulimia nervosa. This may reflect the differences
in age of onset between the syndromes. Currently, the most effective
treatments for anorexia nervosa involve family interventions. This is likely to be
more useful for young adolescent patients who reside with their parents. Thus,
little is known about the effective treatment of older patients with anorexia
nervosa. Conversely, bulimia nervosa develops during young adulthood, and
efficacious treatments have focused on outpatient cognitive behavioral therapy
and antidepressant medications that are likely to work well across the
developmental range of late adolescence to middle age.
For both anorexia nervosa and bulimia nervosa, longer duration of symptoms
prior to presentation predicts worse outcome. However, as has been discussed
before (19), this finding is somewhat tautological as it essentially reveals that
chronicity predicts itself rather than elucidating what specific factors contribute
to a chronic course.
Severity of symptoms appears to be associated with prognosis in anorexia
nervosa but not bulimia nervosa. This result may reflect our supposition that
chronicity in anorexia nervosa is associated with a steady course whereas
chronicity in bulimia nervosa is associated with considerable symptom
fluctuation. If this is true, then a single measure of symptom severity in bulimia
nervosa would have relatively low reliability and thus would have limited ability
to predict prognosis.
Personality disturbances have been associated with worse prognosis for both
anorexia nervosa and bulimia nervosa. However, the nature of personality
disturbances has differed. For anorexia nervosa, obsessive-compulsive
personality features are associated with worse prognosis. Conversely, for
bulimia nervosa, borderline and marked impulsiveness have been associated
with worse prognosis. These likely reflect differences in the core symptoms of
these disorders. Anorexia nervosa is marked by rigid adherence to dietary
restriction enabling maintenance of weight significantly below what would
be expected for height and age. Conversely, bulimia nervosa is marked by

recurrent loss of control over eating. As described above, these differences are
also reflected in course among those who remain chronically ill.
The association between treatment and long-term course and outcome has
differed between anorexia nervosa and bulimia nervosa. Although treatment has
not demonstrated significant associations with long-term outcome in bulimia
nervosa (21), it has been associated with differential course (20). Thus,
treatments that have demonstrated efficacy for bulimia nervosa appear to speed
recovery and be associated with improved long-term course. Conversely, few
treatments have demonstrated efficacy for anorexia nervosa largely due to a
relative dearth of controlled-treatment outcome studies in anorexia
nervosa (39). Controlled-treatment studies have been initiated for anorexia
nervosa; however, insufficient time has passed to evaluate the longterm impact
of treatment on course or outcome.
Our analyses of data reported by Striegel-Moore and colleagues (18) reveal
that women with anorexia nervosa spent significantly more days and money in
inpatient treatment than women with bulimia nervosa, 26.0 days and 17,384
dollars versus 14.7 days and 9088 dollars, respectively [t(196) = 3.55;
p < 0.001 for inpatient days, and t(96) = 3.22, p < 0.01 for inpatient cost].
However, no significant differences in days or cost of outpatient treatment
existed between women with anorexia nervosa and those with bulimia nervosa
17.0 days and 2344 dollars versus 15.6 days and 1882 dollars, respectively [t
(1229) = 1.17; p > 0.05 for outpatient days, and t(1229) = 1.95; p > 0.05 for
outpatient cost]. However, these data were limited by insurance company
coverage and do not necessarily indicate clinical need. Compared to women
with bulimia nervosa, women with anorexia nervosa utilized significantly more
inpatient treatment during the course of follow-up but not individual outpatient
treatment or treatment with antidepressants or anxiolytics (40). Across women
with eating disorders, treatment utilization was predicted by the following
variables in a multivariate analysis: lower Global Assessment of Function scores
(measuring both greater symptom severity and worse psychosocial function),
personality disorder, and comorbid lifetime mood disorders (40).
Notably, in the National Insurance Claims Database study (18), the number
of individuals receiving eating disorder treatment was quite small relative to the
number of individuals in the database. Only 0.06% of female patients in the
database (1194 of 2,005,760) were treated for anorexia nervosa, bulimia
nervosa, or both. Considering the prevalence rates for anorexia nervosa and
bulimia nervosa in the DSM-IV, this suggests that the majority of women with
these eating disorders do not receive treatment for their eating disorder over
the course of a year. Similarly, Fairburn et al. (23) reported that only 28% of
108 KEEL AND HERZOG
individuals with bulimia nervosa received treatment during the course of follow-
up, resulting in a total of 40% ever receiving treatment for an eating disorder.
Bulimia Nervosa and Binge Eating Disorder

Inconsistent results for the comparison of bulimia nervosa and binge eating
disorder outcome have been presented in the literature. Fairburn and
colleagues (23) found that rates of recovery were higher in women diagnosed
with binge eating disorder than women with bulimia nervosa. However, Fichter
and colleagues (37) failed to find differences in outcome between these two
syndromes. Variance in sample ascertainment may account for the contradictory
findings between these two studies. One study recruited women from
treatment (37), and the other recruited women from the community (23).
CONCLUSIONS
In the introduction, we noted that data on long-term course and outcome could
improve our treatment, diagnosis, and understanding of eating disorders. Such
data support the predictive validity of distinguishing between anorexia nervosa
and bulimia nervosa. These disorders appear to be associated with different
rates of mortality, recovery, relapse, cross-over, and distinct sets of prognostic
indicators (including differential treatment response). While promising
treatments have been identified for bulimia nervosa, significant work remains for
anorexia nervosa. Current controlled treatment outcome studies for anorexia
nervosa should be designed to facilitate repeated follow-up assessments.
While few studies have assessed the long-term outcome of bulimia nervosa
and binge eating disorder, even fewer have examined the natural history of
these disorders at long-term follow-up. Almost all longitudinal studies of eating
disorders (including those on anorexia nervosa) are based on treatment seeking
samples, even though the majority of women with eating disorders may not seek
treatment (23,29). The reliance on treatment seeking samples introduces
certain biases in study results. For example, Fairburn et al. (29) reported higher
rates of comorbid disorders in women seeking treatment for bulimia nervosa
than in women with bulimia nervosa from the community. This is referred to as
Berkson’s bias (41) and represents the increased likelihood of comorbidity
among treatment-seeking individuals compared to individuals who do not seek
treatment. Indeed, this bias likely explains the prospective treatment utilization
was predicted by greater severity of eating disorder symptoms, comorbid mood
disorders, and comorbid personality disorders (40). Thus, follow-up studies of
treatment seeking samples may present a more dire description of outcome due
to Berkson’s bias. Furthermore, relying on treatment seeking samples prevents
description of disorders’ natural course as treatment may increase rates of

recovery (42) or speed time to recovery (20,28).
The distinct personality disturbances associated with prognosis in anorexia
versus bulimia nervosa could be examined for their etiological significance.
Specifically, symptoms that appear to be associated with the maintenance of
eating disorder symptoms may play a role in their initiation. Most risk factor
research has failed to elucidate specific risk factors for eating disorders that are
distinct from risk factors for other forms of psychopathological conditions.
Evidence that obsessive-compulsive features are associated with anorexia
nervosa prognosis whereas impulsive-borderline features are associated with
bulimia nervosa prognosis offers the opportunity to reveal risk factors that may
be specific to type of eating disorder. Improved understanding of the etiologic
development of eating disorders holds the promise of their prevention.
REFERENCES
Note’. The following references were included in the original reports of
prognostic indicators for anorexia nervosa and bulimia nervosa presented in
Tables 1 and2: (2,11–13,17,19,22,26,42,44–124).
1. Harris EC, Barraclough B. Excess mortality in mental disorder. Brit J Psychiatry
1998; 17:11–53.
2. Steinhausen H-C. The Outcome of Anorexia Nervosa in the 20th Century. Am J
Psychiatry 2002; 159(8): 1284–1293.
3. Sullivan PF. Mortality in anorexia nervosa. Am J Psychiatry 1995; 152:1073–1074.
4a. Nielsen S. Epidemiology and mortality in eating disorders. Eating Disorders 2001;
24:201–214.
4b. Crisp AH, Callander JS, Halek C, Hsu LKG. Long-term mortality in anorexia
nervosa. Brit J Psychiatry 1992; 161:104–107.
4c. Eckert E, Halmi KA, Marchi P, Grove W, Crosby R. Ten-year follow-up study of
anorexia nervosa: clinical course and outcome. Psychol Med 1995; 25:143–156.
5. Nielsen S, Moller-Madsen S, Isager T, Jorgensen J, Pagsberg K, Theander S.
Standardized mortality in eating disorders: a quantitative summary of previously
published and new evidence. J Psychosom Res 1998; 44:413–434.
6. Eagles JM, Johnston MI, Hunter D, Lobban M, Millar HR. Increasing incidence of
anorexia nervosa in the female population of northeast Scotland. Am J Psychiatry
1995; 152:1266–1271.
7. Patton GC. Mortality in eating disorders. Psychol Med 1988; 18:947–951.
8. Herzog W, Deter HC, Fiehn W, Petzold E. Medical findings and predictors
of long-term physical outcome in anorexia nervosa: a prospective, 12-year
followup study. Psychol Med 1997; 27:269–279.
9. Engel K, Wittern M, Hentze M, Meyer AE. Long-term stability of anorexia
nervosa treatments: Follow-up study of 218 patients. Psychiatric Dev 1989; 4:
395–407.
110 KEEL AND HERZOG
10. Keel PK, Dorer DJ, Eddy KT, Franko DL, Charatan DL, Herzog DB, Predictors of
mortality in eating disorders. Arch Gen Psychiatry 2003; 60:179–183.
11. Herzog D, Dorer DJ, Keel PK, Selwyn SE, Ekeblad ER, Flores A, Greenwood
DN, Burwell RA, Keller MB. Recovery and relapse in anorexia and bulimia
nervosa: a 7.5-year follow-up study. J Am Acad Child Adol Psychiatry 1999; 38:
829–837.
12. Strober M, Freeman R, Morrell W. The long-term course of servere anorexia
nervosa in adolescents: survival analysis of recovery, relapse, and outcome
predictors over 10–15 years in a prospective study. Int J Eat Disord 1997; 22:
339–360.
13. Morgan H, Russell GFM. Value of family background and clinical features as
predictors of long-term outcome in anorexia nervosa: four-year follow-up study of
41 patients. Psychol Med 1975; 5:355–371.
14. Eddy KT, Keel PK, Dorer DJ, Delinsky SS, Franko DL, Herzog DB. Longitudinal
comparison of anorexia nervosa subtypes. Int J Eat Disord 2002; 31:191–201.
15. Polivy J, Herman CP. Dieting and binging: a causal analysis. Am Psychologist
1985; 40:193–201.
16. Herzog DB, Keller MB, Lavori PW. Outcome in anorexia nervosa and bulimia
nervosa: a review of the literature. J Nerv Ment Dis 1988; 176:131–143.
17. Fichter M, Quadlieg N. Six-year course and outcome of anorexia nervosa. Int J
Eating Disord 1999; 26:359–385.
18. Striegel-Moore RH, Leslie D, Petrill SA, Garvin V, Rosenheck RA. One-year use
and cost of inpatient and outpatient services among female and male patients with
an eating disorder: evidence from a national database of health insurance claims. Int
J Eat Disord 2000; 27:381–389.
19. Keel PK, Mitchell JE. Outcome in bulimia nervosa. Am J Psychiatry 1997; 154(3):
313–321.
20. Miller KB, Keel PK, Crow SJ, Thuras P, Mitchell JE: Treatment and treatment
response predict the long-term course of bulimia nervosa. Submitted.
21. Keel PK, Mitchell JE, Miller KB, Davis TL, Crow SJ. Long-term outcome of bulimia
nervosa. Arch Gen Psychiatry 1999; 56:63–69.
22. Collings S, King M. Ten-year follow-up of 50 patients with bulimia nervosa. Br J
Psychiatry 1994; 164:80–87.
23. Fairburn CG, Cooper Z, Doll HA, Norman P, O’Conner M. The natural course of
bulimia nervosa and binge eating disorder in young women. Arch Gen Psychiatry
2000; 57:659–665.
24. Fichter MM, Quadflieg N. Six-year course of bulimia nervosa. Int J Eat Disord
1997; 22:361–384.
25. Field AE, Herzog DB, Keller MB, West J, Nussbaum K, Colditz GA.
Distinguishing recovery from remission in a cohort of bulimic women: how should
asymptomatic periods be described? J Clin Epidemiol 1997; 50:1339–1345.
26. Olmsted M, Kaplan AS, Rockert W. Rate and prediction of relapse in bulimia
nervosa. Am J Psychiatry 1994; 151:738–743.
27. Keel PK, Mitchell JE, Miller KB, Davis TL, Crow SJ. Predictive validity of bulimia
28. Keel PK, Mitchell JE, Davis TL, Crow SJ. The long-term impact of treatment in
women diagnosed with bulimia nervosa. Int J Eat Disord 2002; 31:151–158.
29. Fairburn CG, Welch SL, Norman PA, O’Connor ME, Doll HA. Bias and bulimia
nervosa: how typical are clinic cases? Am J Psychiatry 1996; 153:386–391.
30. American Psychiatric Association. Diagnostic and Statistical Manual for Mental
31. Cachelin FM, Striegel-Moore RH, Elder KA, Pike KM, Wilfley DE, Fairburn CG.
Natural course of a community sample of women with binge eating disorder. Int J
Eat Disord 1999; 25:45–54.
32. Agras WS, Telch CF, Arnow B, Eldredge K, Marnell M. One-year follow-up of
cognitive-behavioral therapy for obese individuals with binge eating disorder.
33. Devlin MJ, Goldfein JA, Carino JS, Wolk SL. Open treatment of overweight binge
eaters with phentermine and fluoxetine as an adjunct to cognitivebehavioral
therapy. Int J Eat Disord 2000; 28:325–332.
34. Peterson CB, Mitchell JE, Engbloom S, Nugent S, Pederson MM, Crow SJ, Thuras
P. Self-help versus therapist-led group cognitive-behavioral treatment of binge
eating disorder and follow-up. Int J Eat Disord 2001; 30:363–374.
35. Ricca V, Mannucci E, Mezzani B, Moretti S, Di Bernardo M, Bertelli M, Rotella
CM, Faravelli C. Fluoxetine and fluvoxamine combined with individual cognitive-
behaviour therapy in binge eating disorder: a one-year follow-up study. Psychother
Psychosom 2001; 70:298–306.
36. Ciano R, Rocco PL, Angarano A, Biasin E, Balestrieri M. Group-analytic and
psychoeducational therapies for binge-eating disorder: an exploratory study on
efficacy and persistence of effects. Psychother Res 2002; 12:231–239.
37. Fichter MM, Quadflieg N, Gnutzman A. Binge eating disorder: treatment outcome
over a 6-year course. J Psychosom Res 1998; 44:385–405.
38. Franko DL, Keel PK, Dorer DJ, Renn R, Eddy KT, Herzog DB. Suicidality in
eating disorders: data from a 12-year longitudinal study. Submitted.
39. Peterson CB, Mitchell JE. Psychological and pharmacological treatment of eating
disorders: a review of research findings. J Clin Psychol 1999; 55:685–697.
40. Keel PK, Dorer DJ, Eddy KT, Delinsky SS, Franko DL, Blais MB, Keller MB,
Herzog DB. Predictors of treatment utilization among women with anorexia and
bulimia nervosa. Am J Psychiatry 2002; 159:140–142.
41. Berkson J. Limitations of the application of fourfold table analysis to hospital data.
Biometrics 1946; 2:47–53.
42. Fairburn C, Norman PA, Welch SL, O’Connor ME, Doll HA, Peveler RC. A
prospective study of outcome in bulimia nervosa and the long-term effects of three
psychological treatments. Arch Gen Psychiatry 1995; 52:304–312.
43. Reas D, Williamson DA, Martin CK, Zucker NL. Duration of illness predicts
outcome for bulimia nervosa: a long-term follow-up study. Int J Eat Disord 2000;
27:428–434.
112 KEEL AND HERZOG
44. Abraham S, Mira M, Llewellyn-Jones D. Bulimia: a study of outcome. Int J Eat

Disord 1983; 2:175–180.
45. Agras W, Rossiter EM, Arnow B, Telch CF, Raeburn SD, Bruce B, Koran LM.
One-year follow-up of psychosocial and pharmacologic treatments for bulimia
nervosa. J Clin Psychiatry 1994; 55:179–183.
46. Beaumont P, George GCW, Smart DE. “Dieters” and “vomiters and purgers” sin
47. Blitzer J, Rollins N, Blackwell A. Children who starve themselves: anorexia
nervosa. Pcychosom Med 1961; 23:369–383.
48. Brotman A, Herzog DB, Hamburg P. Long-term course in 14 bulimic patients
treated with psychotherapy. J Clin Psychiatry 1988; 49:157–160.
49. Browning C, Miller SI. Anorexia nervosa: a study in prognosis and management.
Am J Psychiatry 1968; 124:1128–1132.
50. Bryant-Waugh R, Knibbs J, Fosson A, Kaminski Z, Lask B. Long-term follow-up
of patients with early onset anorexia nervosa. Arch Dis Child 1988; 63:5–9.
51. Burns T, Crisp AH. Outcome of anorexia nervosa in males. Br J Psychiatry 1984;
145:319–325.
52. Casper R, Jabine LN. An eight-year follow-up: outcome from adolescent compared
to adult-onset anorexia nervosa. J Youth Adol 1996; 25:499–517.
53. Dally P, Sargant W. Treatment and outcome of anorexia nervosa. Br Med J 1966;
2:793–795.
54. Dally P. Anorexia Nervosa. London: Heinemann, 1969.
55. Deter H, Herzog W, Petzold E. The Heidelberg-Mannheim study: long-term
follow-up of anorexia nervosa patients at the University Medical Center—
background and preliminary result. In: Herzog W, Deter HC, Vandereycken W,
eds. The Course of Eating Disorders: Long-Term Follow-up Studies of Anorexia
and Bulimia Nervosa. Berlin: Springer-Verlag, 1992:71–84.
56. Edelstein C, Yager J, Gitlin M, Landsverk J. A clinical study of anti-depressant
medications in the treatment of bulimia. Psychiatr Med 1989; 7:111–121.
57. Fahy T, Eisler I, Russell GFM. Personality disorder and treatment response in
bulimia nervosa. Br J Psychiatry 1993; 162:765–770.
58. Fahy T, Eisler I. Impusivity and eating disorders. Br J Psychiatry 1993; 162:
193–197.
59. Fairburn C, Peveler RC, Jones R, Hope RA, Doll HA. Predictors of 12-month
outcome in bulimia nervosa and the influence of attitudes to shape and weight. J
Consult Clin Psychol 1993; 61:686–698.
60. Fairburn C, Kirk J, O’Connor M, Cooper PJ. A comparison of two psychological
treatments for bulimia nervosa. Behav Res Ther 1986; 24:629–643.
61. Fairburn C, Jones R, Peveler RC, Hope RA, O’Connor M. Psychotherapy and
bulimia nervosa: longer-term effects of interpersonal psychotherapy, behavior
therapy, and cognitive behavior therapy. Arch Gen Psychiatry 1993; 50:419–428.
62. Fallon B, Walsh T, Sadik C, Saoud JB, Lukasik V. Outcome and clinical course in
inpatient bulimic women: a 2-to 9-year follow-up study. J Clin Psychiatry 1991;
52:272–278.
63. Fichter MM, Quadflieg N, Rief W. Longer-term course (6-year) course of bulimia
nervosa. Neuropsychopharmacology 1994; 10:7728.
64. Fichter M, Quadfleig N, Rief W. Course of multi-impulsive bulimia. Psychol Med
1994; 24:591–604.
65. Frazier S. Anorexia nervosa. Dis Nerv Syst 1965; 26:155–159.
66. Garfinkel P, Moldofsky H, Garner DM. The outcome of anorexia nervosa:
significance of clinical features, body image and behavior modification. In: Vigersky
R, ed. Anorexia Nervosa. New York: Raven Press, 1977:315–329.
67. Giles T, Young RR, Young DE. Behavioral treatment of severe bulimia. Behav
Ther 1985; 16:393–405.
68. Goetz P, Succop RA, Reinhart JB, Miller A. Anorexia nervosa in children: a follow-
up study. Am J Orthopsychiatry 1977; 47:597–603.
69. Greenfeld D, Anyan WR, Hobart M, Quinlan DM, Plantes M. Insight into illness
and outcome in anorexia nervosa. Int J Eat Disord 1991; 10:101–109.
70. Hall A, Slim E, Hawker F, Salmond C. Anorexia nervosa—long-term outcome in
50 female patients. Br J Psychiatry 1984; 145:407–413.
71. Halmi K, Brodland G, Loney J. Prognosis in anorexia nervosa. Ann Intern Med
1973; 78:907–909.
72. Halmi K, Bordland G, Rigas C. A follow-up study of seventy-nine patients with
anorexia nervosa: an evaluation of prognostic factors and diagnostic criteria. In:
Winokur G, Roff M, Wirt RD, eds. Life History Research in Psychopathology.
Vol. 4. Minneapolis: University of Minnesota Press, 1976:290–300.
73. Hawley R. The outcome of anorexia nervosa in younger subjects. Br J Psychiatry
1985; 146:657–660.
74. Herpetz-Dahlmann B, Wewetzer C, Schulz E, Remschmidt H. Course and
outcome in adolescent anorexia nervosa. Int J Eat Disord 1996; 19:335–345.
75. Herzog D, Sacks NR, Keller MB, Lavori PW, von Ranson KB, Gray HM. Patterns
and predictors of recovery in anorexia nervosa and bulimia nervosa. J Am Acad
76. Herzog T, Hartmann A, Sandholz A, Stammer H. Prognostic factors in outpatient
psychotherapy of bulimia. Psychother Psychosom 1991; 56:48–55.
77. Higgs J, Goodyer IM, Birch J. Anorexia nervosa and food avoidance emotional
disorder. Arch Dis Child 1989; 64:346–351.
78. Hsu L, Holder D. Bulimia nervosa: treatment and short-term outcome. Psychol
Med 1986; 16:65–70.
79. Hsu L, Crisp AH, Harding B. Outcome of anorexia nervosa. Lancet 1979; 1:
61–65.
80. Hudson J, Pope HG, Keck PE, McElroy SL. Treatment of bulimia nervosa with
trazodone: short-term response and long-term follow-up. Clin Neuropharmacol
1989; 12(suppl 1):S38-S46.
81. Jarman F, Rickards WS, Hudson L. Late adolescent outcome of early-onset
anorexia nervosa. J Paediatr Child Health 1991; 27:221–227.
114 KEEL AND HERZOG
82. Johnson C, Tobin DL, Dennis A. Differences in treatment outcome between

borderline and nonborderline bulimics at one-year follow-up. Int J Eat Disord
1990; 9:617–627.
83. Johnson-Sabine E, Reiss D, Dayson D. Bulimia nervosa: a 5-year follow-up study.
Psychol Med 1992; 22:951–959.
84. Kalucy R, Crisp AH, Harding B. A study of 56 families with anorexia nervosa. Br J
Med Psychol 1977; 50:381–395.
85. Kay D, Shapira K. The prognosis in anorexia nervosa. In: Meyer J, Feldman H,
eds. Anorexia Nervosa. Stuttgart: Thieme-Vedag, 1965:113–127.
86. Keller M, Herzog DB, Lavori PW, Ott IL, Bradburn IS, Mahoney EM. High rates
of chronicity and rapidity of relapse in patients with bulimia nervosa and depression.
87. Kreipe R, Churchill BH, Strauss J. Long-term outcome of adolescents with
anorexia nervosa. Am J Dis Child 1989; 143:1322–1327.
88. Lacey J. Bulimia nervosa, binge eating, and psychogenic vomiting: a controlled
treatment study and long-term outcome. Br Med J 1983; 286:1609–1613.
89. Laessle R, Beumont PJV, Butow P, Lennerts W, O’Connor M, Pirke KM, Touyz
SW, Waadt S. A comparison of nutritional management with stress management in
the treatment of bulimia nervosa. Br J Psychiatry 1991; 159: 250–261.
90. Lesser L, Ashenden BJ, Debushey M, Eisenberg L. Anorexia nervosa in children.
Am J Orthopsychiatry 1960; 30:572–580.
91. Martin F. The treatment and outcome of anorexia nervosa in adolescents: a
prospective study and five-year follow-up. J Psychiatr Res 1985; 19:509–514.
92. Mitchell J, Davis L, Goff G, Pyle R. A follow-up study of patients with bulimia. Int
J Eating Disord 1986; 5:441–450.
93. Mitchell J, Pyle RL, Hatsukami D, Goff G, Glotter D, Harper J. A 2–5 year follow-
up study of patients treated for bulimia. Int J Eating Disord 1988; 8:157–165.
94. Morgan H, Purgold J, Welbourne J. Management and outcome in anorexia nervosa
—a standardized prognostic study. Br J Psychiatry 1983; 143:282–287.
95. Nash E, Colborn AL. Outcome of hospitalized anorexics and bulimics in Cape Town,
1979–1989. S Afr Med J 1994; 84:74–79.
96. Niederhoff H, Wiesler B, Kuenzer W. Somatisch orientierte Behandlung der
Anorexia nervosa. Monatsschr Kinderheilkd 1975; 123:343–344.
97. Norman D, Herzog DB, Chauncey S. A one-year outcome study of
bulimia: psychological and eating symptom changes in a treatment and non-
treatment group. Int J Eating Disord 1986; 5:47–57.
98. Nussbaum M, Shenner IR, Baird D, Saravay S. Follow-up investigation in patients
with anorexia nervosa. J Pediatrics 1985; 106:835–840.
99. Pyle R, Mitchell JE, Eckert ED, Hatsukami D, Pomeroy C, Zimmerman R.
Maintenance treatment and 6-month outcome for bulimic patients who respond to
initial treatment. Am J Psychiatry 1990; 147:871–875.
100. Ratnasuriya R, Eisler I, Szmukler GI, Russell GFM. Anorexia nervosa: outcome
and prognostic factors after 20 years. Br J Psychiatry 1991; 158:465–502.
101. Remschmidt H, Wienand F, Wewetzer C. Der Langzeitverlauf der Anorexia

nervosa. Monatsschr Kinderheilkd 1988; 136:726–731.
102. Rollins N, Blackwell A. The treatment of anorexia nervosa in children and
adolescents: stage 1. J Child Psychol Psychiatry 1968; 9:81–91.
103. Rosenvinge J, Mouland SO. Outcome and prognosis of anorexia nervosa: a
retrospective study of 41 subjects. Br J Psychiatry 1990; 156:92–97.
104. Rossiter E, Agras WS, Telch CF, Schneider JA. Cluster B personality disorder
characteristics predict outcome in the treatment of bulimia nervosa. Int J Eating
Disord 1993; 13:349–357.
105. Saccomani L, Savoini M, Cirrincione M, Ravera G. Long-term outcome of children
and adolescents with anorexia nervosa: study of comorbidity. J Psychosom Res
1998; 44:565–571.
106. Santonastaso P, Favaretto G, Canton G. Anorexia nervosa in Italy: clinical features
and outcome in a long-term follow-up study. Psychopathology 1987; 20:8–17.
107. Santonastaso P, Pantano M, Panarotto L, Silvestri A. A follow-up study on
anorexia nervosa: clinical features and diagnostic outcome. Eur Psychiatry 1991; 6:
177–185.
108. Schulze U, Neudorfli A, Krill A, Warnke A, Remschmidt H, HerpertzDahlmann
B. Verlauf und Heilungserfolg der fruhen Anorexia nervosa. Z Kinder
Jugendpsychiatr 1997; 25:5–16.
109. Seidensticker J, Tzagournis M. Anorexia nervosa—clinical features and longterm
follow-up. J Chronic Dis 1980; 21:366–367.
110. Steiner H, Mazer C, Litt IF. Compliance and outcome in anorexia nervosa. West
JMed 1990; 157:133–139.
111. Steinhausen H, Glanville K. Retrospective and prospective follow-up of adolescent
anorexia nervosa. Acta Psychiatr Scand 1983; 68:1–10.
112. Steinhausen H, Seidel R. Outcome in adolescent eating disorders. Int J Eat Disord
1993; 14:487–496.
113. Surzenberger S, Cantwell PD, Burroughs J, Salkin B, Green JK. A follow-up study
of adolescent psychiatric inpatients with anorexia nervosa. J Am Acad Child
Psychiatry 1977; 16:703–715.
114. Swift W, Kalin NH, Wamboldt FS, Kaslow N, Ritholz M. Depression in bulimia at
2- to 5-year follow-up. Psychiatry Res 1985; 16:111–122.
115. Thackwray D, Smith MC, Bodfish JW, Meyers AW. A comparison of be havioral
and cognitive-behavioral interventions for bulimia nervosa. J Consult Clin Psychol
1993; 61:639–645.
116. Theander S. Anorexia nervosa: a psychiatric investigation of 94 female patients.
Acta Psychiatr Scand Suppl 1970; 214:1–194.
117. Thiel A, Zuger M, Jacoby GE, Schussler G. Thirty-month outcome in patients with
anorexia or bulimia nervosa and concomitant obsessive-compulsive disorder. Am J
Psychiatry 1998; 155:244–249.
118. Toner B, Garfield PE, Garner DM. Long-term follow-up of anorexia nervosa.
Psychosom Med 1986; 48:520–529.
116 KEEL AND HERZOG
119. Walford G, McCune N. Long-term outcome in early-onset anorexia nervosa. Br J

Psychiatry 1991; 159:383–389.
120. Walsh B, Hadigan CM, Devlin MJ, Gladis M, Roose SP. Long-term outcome of
antidepressant treatment for bulimia nervosa. Am J Psychiatry 1991; 148:
1206–1212.
121. Warren W. A study of anorexia nervosa in young girls. J Child Psychol Psychiatry
1968; 9:27–40.
122. Willi J, Hagemann R. Langzeitverlaufe von Anorexia nervosa. Schweizerische Med
Wochenschrift 1976; 106:1459–1465.
123. Willi J, Limacher B, Nussbaum P. Zehnjahres-Katamnese der 1973–75 im Kanton
Zurich erstmals hospitalisierten Anorexien. Schweizerische Med Wochenschrift
1989; 119:147–159.
124. Wonderlich S, Fullerton D, Swift W, Klein MH. Five-year outcome from eating
disorders relevance of personality disorders. Int J Eating Disord 1994; 15:
233–243.
6
An Overview of Risk Factors for Anorexia
Nervosa, Bulimia Nervosa, and Binge Eating
Disorder
Corinna Jacobi
University of Trier
Trier, Germany
Lisette Morris
University of Göttingen
Göttingen, Germany
Martina de Zwaan
Friedrich Alexander University Erlangen-Nuremberg
Erlangen, Germany
In the field of eating disorders, more than 30 variables have been reported as
risk factors for the development of an eating disorder (Table 1). Among these
variables, risk has been assessed from different perspectives: reported risk
factors include social, familial, psychological, developmental, and biological
factors. For many of these factors, the question of whether they occurred prior
to the onset of the eating disorder has not yet been definitively answered.
Therefore, it is impossible to determine whether they are symptoms,
maintaining factors, consequences, or “scars” of the disorder.
The inconsistent use of the term risk factor as well as the lack of precise
definitions led Kraemer and coworkers to set out a conceptual basis for a
typology of risk factors (1,2). This theoretical framework forms the basis of our
overview on risk factors for eating disorders.
In the past, risk and etiological factors for eating disorders have either been
proposed from a specific theoretical perspective (e.g., biological, cognitive-
behavioral, psychodynamic) or from an integrative perspective (e.g.,
biopsychosocial model). Factors in these models can be based on both clinical
experience and empirical data, but the empirical foundation can vary from very
strong to very weak and can comprise a wide range of methods and definitions
for risk and etiologic factors. In contrast, the risk factor approach applied here
118 JACOBI ET AL.
TABLE 1 Potential Risk Factors for Eating Disorders
to risk factors for eating disorders is primarily descriptive and data-oriented,

refrains from making dynamic assumptions about interactions of factors, and
should be regarded as a precursor to a more complex model.
The purpose of this chapter is to summarize findings on risk factors for eating
disorders (anorexia nervosa, bulimia nervosa, binge eating disorder) on the basis
of the risk factor approach outlined by Kraemer et al. (1). More specifically, we
will compare potential risk factors from longitudinal and cross-sectional studies,
establish the status of each of the potential risk factors discussed in the literature
RISK FACTORS 119
for the different eating disorder syndromes, and discuss additional risk factor
characteristics (i.e., specificity, potency).
OVERVIEW OF BASIC RISK FACTOR CONCEPTS

In the conceptual papers previously noted (1,2), a risk factor is considered to be
a measurable characteristic of each subject in a specified population, which
precedes the outcome of interest and which can be shown to divide the
population into a high- and a low-risk group. The probability of the outcome in
the high-risk group must be shown to be greater than in the lowrisk group.
Beyond a statistically significant association between risk factor and outcome,
the clinical significance of a risk factor should be indicated by the magnitude of
the association, called the “potency” of the factor.
If, in the defined population, an association between the factor and the
outcome can be shown, the factor is called a correlate (see Table 2 for a summary
of definitions and identification methods). Any correlate measured
concomitantly with or after the outcome may potentially be a concomitant or
consequence of the outcome (i.e., a symptom or scar). Only if a correlate can
be demonstrated to precede the outcome is the term “risk factor” justified. To
establish a factor’s status as a risk factor, precedence is a crucial criterion. A risk
factor that can be shown to change spontaneously within a subject (e.g., age,
weight) or to be changed by intervention (such as medication or psychotherapy)
is called a variable risk factor. A risk factor that cannot be demonstrated to change
or be changed (e.g., race, gender, year of birth) is called a fixed marker. A
variable risk factor for which it can be shown that manipulation changes the risk
of the outcome (e.g., onset of disorder) is called a causal risk factor. If that
cannot be shown, it is called a variable marker.
Because precedence is a crucial criterion of a risk factor, the majority of risk
factors can only be assessed in longitudinal studies. Exceptions are fixed
markers taken from medical records or birth registers documented before the
onset of the eating disorder. On the other hand, the majority of factors assessed
in cross-sectional studies (e.g., epidemiological studies, casecontrol studies,
family studies) are correlates. Exceptions are, again, some fixed markers, such
as race or gender. The status of variable markers and causal risk factors can
only be established in randomized clinical trials (prevention or intervention
studies) which confirm that the modification of the factors leads to a change in
the risk of the outcome (e.g., onset of the disorder).
In spite of their limitations, cross-sectional studies may constitute an
important first step on the way to testing specific risk (or causal) factors in
longitudinal studies (3,4). Especially cross-sectional studies that assess putative
risk factors retrospectively may lead to different hypotheses compared to those
120 JACOBI ET AL.
TABLE 2 Risk Factor Typology and Identification Method
that do not. Although retrospective assessment of risk factor information is

problematic because of retrospective recall or memory biases, we decided to
differentiate between cross-sectional studies with retrospective risk factor
assessment and those without. For the former category the additional term
retrospective correlate was coined for factors shown to predate the onset of the
disorder according to subjects’ self-report. This procedure allowed us to
compare these results with the results of studies in which risk factor
terminology is optimally realized.
METHOD
The methodology of our approach has been described in detail in Jacobi
et al. (5). Therefore, only the most important methodological issues are
summarized here.
RISK FACTORS 121
For the identification of studies, a detailed computerized and manual

literature search of empirical studies of potential risk factors for the onset of
eating disorders was conducted through April 2002. Approximately 5000
abstracts were screened; about 300 (cross-sectional and longitudinal) studies
were finally selected for inclusion in the review.
The following inclusion/exclusion criteria were imposed: For the majority of
the factors examined, only studies with a control group (normal or unaffected
control group) were included in the review; studies comparing different eating
disorder groups alone were excluded. To establish specificity an additional
clinical group was required. The minimal sample size in the studies was
10 subjects per cell. The follow-up interval for longitudinal studies had to be at
least 1 year to allow enough time for symptoms of eating disturbances or
disorders to change or emerge. Because the focus of our overview was placed on
risk factors for eating disorder syndromes, longitudinal studies solely addressing
(dimensional) disturbances or symptoms via questionnaires (e.g., Eating
Disorders Inventory, EDI) were not included.
Factors from the included studies were first classified according to the Kraemer
et al. (1) typology as outlined above. Factors were also classified with regard to
outcome status, i.e., they were categorized according to whether they were
associated with or predicted (a) full-syndrome anorexia nervosa, bulimia
nervosa, or binge eating disorder, or (b) partial-syndrome eating disorders,
otherwise defined “eating disturbances,” or mixed outcomes of syndromes and
symptoms.
Finally, a third level of classification was applied to establish additional risk
factor characteristics (e.g., specificity, potency):
We classified a factor as a specific risk factor for one of the three diagnostic
categories of eating disorders if it predicted the outcome only for this group but
not for the clinical comparison group.
Risk factor potency was determined on the basis of the following indicators
generally reported in the original studies: effect sizes based on Cohen’s delta
(ordinal risk factors) and the proportion of subjects with the risk factor in the
low- vs. high-risk group (binary risk factors) (6). These were then transformed
into one common potency indicator, the area under the curve (AUC) statistic.
The AUC estimates the probability that a person with an eating disorder will
have a higher value on an ordinal risk factor or be more likely to have a binary
risk factor than a person without an eating disorder. The standards used to
categorize the AUC reflect Cohen’s standards for effect sizes (ES).
122 JACOBI ET AL.
CHARACTERISTICS OF LONGITUDINAL RISK

FACTOR STUDIES FOR EATING DISORDERS
Because longitudinal studies represent the “gold standard” of risk factor research,
the main characteristics of the included studies are described here in more
detail. In the following sections, we will then summarize the results of both
longitudinal and cross-sectional risk factor research according to the diagnostic
syndromes anorexia nervosa, bulimia nervosa, and binge eating disorder. For
anorexia and bulimia nervosa, these results will also be presented graphically by
ordering risk factors and retrospective correlates along a timeline according to
their chronological emergence (Figs. 1 and 2).
Twenty-eight longitudinal studies were found. Thirteen of these were
excluded from our overview because of too few cases, insufficient information
on risk factors and reported methodology, lack of controlling for initial eating
disorder symptoms, too broad outcome measures, or too short follow-up
intervals.
In spite of large samples sizes, the majority of longitudinal studies have only
identified risk factors for a mixture offull DSM syndromes of anorexia nervosa and
bulimia nervosa and/or partial syndromes or eating disorders not otherwise
specified (ED-NOS; 7–15). In six studies, the outcome samples were high-risk
samples usually defined as scoring above the Eating Attitudes Test 26 (EAT-26)
cutoff (16–21).
A closer examination of the outcomes in longitudinal studies reveals that the
focus is on bulimic and binge eating syndromes while reports of anorexic
syndromes as outcomes were much less common. Taken together, only 27
cases of anorexia nervosa out of a total of 12,776 subjects in the 15 studies
emerged during the follow-up periods.
Samples in the studies consist mostly of adolescents between 12 and 15 years;
three studies assess infants or younger children (8,11,12), two studies young
adults (17,15). In nine studies the samples included females only, in six the
samples consisted of both males and females. The duration of followups varied
from 1 to 18 years. Due to differences in the primary aims of the studies, the
number and broadness of included potential risk factors, and the definitions and
assessment of risk status or caseness (symptomatic/asymptomatic; cases/
noncases; high/moderate/low risk) varies significantly: seven studies assess
eating disorder symptoms and syndromes with structured diagnostic interviews,
eight studies rely solely on questionnaire cutoffs or score combinations.
FIGURE 1 Risk factors, fixed and variable markers, and retrospective correlates for anorexia nervosa.
RISK FACTORS 123
124 JACOBI ET AL.
FIGURE 2 Risk factors, fixed and variable markers, and retrospective correlates for bulimia nervosa.
RISK FACTORS 125
RISK FACTORS FOR ANOREXIA NERVOSA
Psychosocial Risk Factors
Gender
In both clinical and nonclinical samples, anorexia and bulimia nervosa have been
observed to occur predominantly in females (14,15,22–25): populationbased
studies estimate a female-to-male ratio of 10:1 (26,27). Since gender is an
immutable characteristic and asymmetrical gender distributions have been found
for other mental disorders, e.g., (28), female status is classifiable as a
nonspecific fixed marker for both anorexia and bulimia nervosa.
Ethnicity
Though eating disorders have been traditionally regarded as a predominantly
Caucasian issue [see review by Striegel-Moore and Smolak (29)], a literature
survey from Crago et al. (30) revealed a more complex ethnic distribution
pattern: native Americans were noted as having higher rates of eating disorders
than Caucasians, whereas equal rates were reported for Hispanics and lower
rates for Asians and blacks. The results from two further studies also point to
lower rates of anorexia and bulimia nervosa among Asians (31,32). Therefore,
non-Asian ethnicity can be classified as a fixed marker of currently unknown
specificity.
Acculturation
An association between the level of acculturation within ethnic minorities and
the presence of eating disorder symptoms has been reported in a number of
cross-sectional studies (33–35). On the basis of the assumption that
acculturation predates onset of eating disorders symptoms, it is classified as a
retrospective correlate of unknown specificity.
Age
In both clinical and population-based surveys, the peak incidence of eating
disorders has been found in the age range from adolescence to early
adulthood (36). The magnitude of the age-associated increase in risk depends on
which age groups are compared. As yet, direct comparisons of eating disorder
rates in latency age samples with later adolescent samples have not been made.
126 JACOBI ET AL.
In the Kraemer typology age is classified as a variable risk factor and, because of
its relationship to other psychiatric disorders, age is categorized as nonspecific.
Weight concerns/Dieting/Negative Body Image

Cross-Sectional Evidence. The association between dieting and eating disorders is
probably one of the most often quoted in theories on the etiology of eating
disorders. Because the relationship between dieting and eating disorders seems
to be particularly strong for patients who binge, i.e., patients with anorexia
nervosa bingeing-purging subtype and bulimia nervosa, and less strong for
restricting anorexic patients, the results will be summarized in the respective
section of psychosocial risk factors for bulimia nervosa. Based on cross-sectional
studies, dieting behavior appears to be a retrospective correlate of the bingeing-
purging subtype of anorexia nervosa. Its specificity is unclear, since comparisons
with other clinical groups have not been reported.
Longitudinal Evidence. Because most of the newly emerging cases in
longitudinal studies were bulimic or partial bulimic cases, sometimes mixed
with a few anorexic cases, longitudinal evidence for a relationship between
weight concerns or dieting and anorexia nervosa is scarce. Only Marchi and
Cohen (12) found that elevations on a measure of anorexic symptoms during
childhood predicted adolescent anorexia nervosa or symptoms of anorexia.
Accordingly, weight concerns, negative body image, and dieting are tentatively
classified as variable risk factors for anorexia nervosa, which need replication.
The specificity of these factors has not yet been addressed.
General Psychiatric Disturbance/Negative Emotionality

Cross-Sectional Evidence. Previous theories on the development of eating
disorders as well as a large body of research on comorbidity have stressed the
role of other psychiatric disorders (e.g., affective disorders, substance abuse
disorders, anxiety disorders, personality disorders) as underlying and/or
associated conditions in clinical samples of anorexic and bulimic patients [for
reviews, see, for example, (37,38)]. Chronology of the onset of the comorbid
disorders, however, has only been explicitly addressed in two crosssectional
comorbidity studies.
Rastam (39) examined premorbid physical and developmental problems and
premorbid personality disorders obtained via a semistructured interview carried
out with the mothers of an adolescent sample of anorexic patients and control
probands. The total number of premorbid personality disorders was
significantly higher in the anorexic patients (66% vs. 27%). Of the individual
personality disorder diagnoses, only premorbid obsessivecompulsive personality
RISK FACTORS 127
disorder (OCPD) occurred significantly more often in anorexic patients than in

the control probands (35% vs. 4%).
Bulik et al. (40) compared prevalence and onset of comorbid childhood
anxiety disorders in two eating-disordered groups (women with anorexia and
bulimia nervosa), a clinical control group (women with major depression), and
a healthy control group N = 98. Rates of lifetime anxiety disorders were highest
in the three clinical groups (48–60%) but they were also found in controls (30%).
The prevalence of obsessive-compulsive disorder (OCD) and separation anxiety
disorder was highest in the anorexic group, overanxious disorder highest in the
anorexic and bulimic groups, and panic disorder highest in the anorexia nervosa
and major depression groups. Ninety percent of anorexic women and 71% of
depressed women with lifetime anxiety disorders reported that the onset of
their anxiety disorder predated the onset of their other disorder.
Based on logistic regressions, the authors interpret their findings as
supporting the hypothesis that overanxious disorder of childhood is a
nonspecific risk factor for later psychopathology (eating or affective disorders)
and that the relationship with anorexia nervosa is particularly strong. It has also
been shown that OCD may be a specific risk factor for anorexia nervosa (40).
No information could be obtained as to the rates of the specific anxiety
disorders preceding the eating disorders in this study. When the rates of any
lifetime anxiety disorder were compared in the three patient groups, the rates in
anorexic and bulimic versus depressed patients were almost twice as high: 54%
of anorexic patients, 53% of bulimic patients, and 34% of depressed patients
endorsed the presence of an anxiety disorder predating the onset of their other
disorder.
Taken together, cross-sectional evidence for comorbid disorders predating
the onset of anorexia nervosa is limited to childhood overanxious disorder,
OCD, and OCPD. These disorders are therefore classified a retrospective
correlates for anorexia. As the evidence for the correlate status of these anxiety
disorders is limited to one study, additional studies are needed to replicate these
results and also to examine the role of other comorbid disorders (e.g., affective
and substance use disorders).
Longitudinal Evidence. Seven longitudinal studies included general psychiatric
morbidity, psychopathology, or, more specifically, negative emotionality as
potential predictors of eating disorders (10,13,14,16,19–21). In only one of
these studies (21) was anorexia detected at follow-up; only one case was
identified. It therefore seems premature to classify psychiatric morbidity or
negative emotionality as a risk factor for the full syndrome of anorexia nervosa.
128 JACOBI ET AL.
Sexual Abuse
Cross-Sectional Evidence. Sexual abuse in childhood, adolescence, or adulthood
has been discussed as a risk factor for eating disorders in many studies as well as
in early (41) and more recent reviews (38). All but one study are cross-
sectional in nature. The studies included here were not restricted to childhood
abuse but included sexual abuse at a later age as well. This procedure was
elected, as a valid distinction between childhood and adolescent sexual abuse
often proved impossible due to methodological inconsistencies and divergent
definitions in the cross-sectional studies.
In general, cross-sectional studies comparing homogeneous diagnostic
groups of anorexic patients to controls are rare. The few controlled studies
primarily comprise a mixture of anorexic and bulimic clinical samples. Whereas
some studies found higher rates of sexual abuse in these mixed samples, in
others elevated rates seemed to be linked to binge eating-type anorexic
patients (42–44). However, no information was given in these studies on the onset
of the eating disorder and the first occurrence of sexual abuse.
In the study by Webster and Palmer (45), on set of eating disorders occurred
in 74% of the eating-disordered subjects after age 16 years. However, no
differences were found in the rates of sexual abuse occurring before age 16
when comparing four clinical groups [anorexia nervosa (AN), bulimia nervosa
(BN), mixed AN/BN, major depression] to a nonmorbid comparison group.
One study with clinical samples explicitly assessed the precedence of the abuse
relative to the eating disorders (46). Comparing three patient groups (AN, BN,
and depressive patients) and a control group, they found that the rates
of abuse did not differ significantly between the three patient groups
(42.9%/27.1%/40.0%), but that all of the clinical groups had significantly
higher rates of sexual abuse than the control group (6.7%).
Taken together, the evidence for the classification of sexual abuse as a
retrospective correlate of anorexia nervosa is based on one study only with
retrospective assessment. Replication studies are therefore needed urgently.
Longitudinal Evidence. To date, only one longitudinal study has addressed the
role of sexual abuse as a risk factor for eating disorders (8). Because only one
patient developed anorexia nervosa at follow-up, the results will be reported in
the respective section of risk factors for bulimia.
Adverse Life Events

Cross-Sectional Evidence. Only a few studies have investigated the relationship
between stressful life events and the onset of an eating disorder. With one
exception, these studies are again all cross-sectional in nature. Among studies
RISK FACTORS 129
comparing eating-disordered patients solely to normal controls, the results are

not consistent: the percentages of (recovered) anorexic and bulimic patients,
who had experienced a severe event before the onset of their eating disorder,
did not differ markedly from the percentage of normal controls with severe
events in the study by Troop and Treasure (47).
Schmidt et al. (48), comparing anorexic and bulimic patients, and
community controls, found no differences in the proportion of patients with at
least one severe event, but found that significantly more anorexic and bulimic
patients had experienced what was called a “major difficulty”; in addition,
“pudicity” problems were found more frequently in anorexic patients than in the
other groups. Also, Rastam and Gillberg, (49) comparing a mixed clinical and
population sample of 51 anorexic patients with a matched control group, did not
find differences with respect to chronic life events but rather with respect to
recent (3 months before onset of the disorder) major events such as loss of first-
degree relatives (14% of anorexic vs. 0% of controls).
Among studies that also included a psychiatric control group, the studies by
Horesh et al. found that adverse life events differentiated significantly between
anorexic patients and healthy controls (50) and between a mixed group of
anorexic and bulimic patients and healthy controls (51). More distinct
differences in comparison to psychiatric controls could only be found for
inappropriate parental pressures (51). Gowers et al. found an intermediate rate
of negative life events for anorexic patients between that of psychiatric patients
and community controls (52).
Taken together, there is some evidence that anorexic patients experience
more severe life events before the onset of their eating disorder than healthy
control subjects, although the evidence is not as consistent as for the role of
sexual abuse. There is also some evidence that this relationship is not specific to
patients with eating disorders but applies to psychiatric patients in general. In all
of the studies investigating the role of life events in precipitating eating
disorders, the events were reported to have occurred prior to the onset of the
eating disorders. Therefore, general adverse life events are classified as
nonspecific retrospective correlates.
Longitudinal Evidence. No longitudinal evidence for a relationship between
adverse life events and the subsequent development of anorexia nervosa exists.
Early Childhood Eating and Digestive Problems

Cross-Sectional Evidence. One case-control study found retrospectively assessed
early feeding and severe gastrointestinal problems to be extremely frequent in
anorectic patients (found in 90%). Rates were almost two times higher than in
the matched control group (49). Early childhood feeding and digestive
130 JACOBI ET AL.
problems are therefore classified as retrospective correlates of unclear

specificity.
Longitudinal Evidence. The first longitudinal study including a sample of
younger children found digestive problems and picky eating to be prospectively
related to subsequent anorexic symptoms and anorexic symptoms to be related
to later full diagnoses of anorexia nervosa (12).
The second longitudinal study also found a range of early childhood eating
problems, assessed between ages 1 and 10, to predict eating disorder diagnoses
in early and late adolescence and young adulthood: eating conflicts, struggles
around meals, and unpleasant meals in childhood predicted later diagnoses of
anorexia nervosa (11). On the basis of these studies, picky eating, anorexic
symptoms in childhood, digestive and other early eating-related problems as
well as eating conflicts, struggles around meals, and unpleasant meals are
classified as variable risk factors for anorexia nervosa or anorexic symptoms.
The specificity of these eating and digestive problems has not yet been tested.
Given the small number of studies, replication studies are needed.
Family Interaction/Family Functioning/Attachment Styles

Cross-Sectional Evidence. Historically, the role of dysfunctional family
interaction styles was put forward in theories on the development of eating
disorders (53,54). Characteristics of the patient-family relationships of eating
disordered patients include problematical family structures, interaction, or
communication styles (e.g., overprotection, enmeshment) and attachment
styles.
In the majority of studies, e.g., (55–59), both anorexic and bulimic patients
describe different aspects of their family structure (interaction, communication,
cohesion, affective expression, attachment disturbances, etc.) as more disturbed,
conflictual, pathological, or dysfunctional than normal controls across different
family assessment measures. However, the issue of precedence to the eating
disorders was not addressed in any of these studies. Furthermore, similar
patterns of family interaction or family functioning have been found in
connection with other disorders, indicating a possibly nonspecific nature of
these disturbances.
To date, family variables for the time span prior to onset of the eating
disorders have been assessed retrospectively in only three studies: childhood and
family background as well as the specificity of these possible characteristics were
addressed retrospectively in the study by Webster and Palmer (45). Three
groups of eating-disordered patients (AN, BN, mixed AN/BN) were compared
with a group of women with major depression and a nonmorbid comparison
group using a semistructured interview measure of childhood care before
RISK FACTORS 131
age 17. When compared with the nonmorbid group, patients with anorexia
nervosa showed no significant differences on any of the variables studied (e.g.,
parental indifference, control and care, antipathy, discord in family).
Shoebridge and Gowers (60) addressed overprotection or high-concern
parenting in anorexia nervosa as part of a cross-sectional case-control study.
High-concern parenting was assessed by a structured clinical interview carried
out with the mothers covering the first 5 years of the child’s life. Two or more
high-concern attitudes and behaviors were almost four times more common in
mothers of anorexic patients than in mothers of controls. In addition, infant
sleep pattern difficulties were significantly more common in anorexic patients
than in controls. In the same sample, a more than threefold higher frequency of
obstetric losses was found prior to the birth of the future anorexic patients
compared with controls. The findings concerning parenting may therefore also
reflect an adaptive parental behavior to a severe life event and may not be
generalizable to a different sample or different (e.g., adolescent) age group. The
presence of two or more variables of high-concern parenting is classified as
retrospective correlate of unclear specificity.
The community-based studies by Fairburn et al. (61–63) compared three
groups of eating-disordered patients (patients with a history of anorexia
nervosa, with bulimia nervosa, and with binge eating disorder) with general
psychiatric patients and healthy controls. A large number of putative risk factors
(including variables of parental interaction) were assessed by an interview
focusing on the period before the onset of the eating disorders. For most of the
retrospectively assessed variables of parental interaction (arguments, criticism,
high expectations, underinvolvement, minimal affection, low care, and high
overprotection) the (previously) anorexic patients showed elevated rates when
compared with normal controls (63). However, in comparison with the
psychiatric controls, none of these variables turned out to be specifically
predictive of anorexia nervosa.
These factors together with high-concern parenting are therefore classified as
nonspecific retrospective correlates.
Longitudinal Evidence. As anorexia nervosa was not included as an outcome in
any of the longitudinal studies assessing variables of family interaction, no
evidence for these variables as risk factors for anorexia nervosa exists.
Family History/Family Psychopathology

Cross-Sectional Evidence. To date, a large body of cross-sectional research has
examined the role of familial psychiatric disorders as a risk factor for eating
disorders utilizing both the family history and the family study methods. The
results of eight cross-sectional studies comparing the familial psychopathology
132 JACOBI ET AL.
of anorexic patients with relatives of either healthy or psychiatric controls can

be summarized as follows [for details, see Jacobi et al. (5)].
The majority of these studies suggest elevated rates of psychiatric disorders in
first-degree relatives of anorexic patients. The evidence is particularly strong for
rates of eating disorders (AN and BN), affective disorders, some anxiety
disorders (panic, generalized anxiety disorder, OCD), and OCPD, and rather
weak for rates of substance abuse disorders. However, the chronology of the
onset of anorexia nervosa in relation to the respective psychiatric disorders in the
families was not addressed in any of these studies.
Selected parental psychiatric disorders (parental depression, alcoholism, and
drug abuse) occurring prior to the onset of the child’s anorexia nervosa were
assessed retrospectively in the risk factor study by Fairburn et al. (63) as part of
the environmental domain risk factor. For none of these parental psychiatric
disorders were significantly elevated rates found in the previously anorexic
patients in comparison with healthy and psychiatric controls. These variables
therefore remain correlates.
Longitudinal Evidence. No longitudinal evidence for the role of family
psychopathology as a risk factor for anorexia nervosa exists.
Low Self-Esteem/Negative Self-Concept/lneffectiveness

Cross-Sectional Evidence. Low self-esteem, a negative self-concept, or
“ineffectiveness” has assumed a central role in many clinically derived theories
of eating disorders, e.g., (64). Constructs related to low self-esteem, especially
ineffectiveness (operationalized by the EDI subscale Ineffectiveness), have been
examined in many cross-sectional studies. Anorexic patients have consistently
been found to exhibit lower self-esteem, a more negative self-concept, or
higher levels of ineffectiveness than normal controls (5,65). The retrospective
assessment of self-concept and its temporal relation to onset of the eating
disorder has only been considered in one cross-sectional study: Fairburn and
coworkers (63) found for the (previously) anorexic patients that negative self-
evaluation prior to the onset of the eating disorder was more common than in
healthy or, more specifically, psychiatric controls. Negative self-evaluation can
therefore be considered a specific retrospective correlate.
Longitudinal Evidence. Because longitudinal evidence on the role of low self-
esteem or negative self-evaluation is, for the most part, based on studies with
bulimia nervosa or binge-related syndromes as outcome, it cannot be classified
as a risk factor for anorexia nervosa.
RISK FACTORS 133
Perfectionism
Cross-Sectional Evidence. From a clinical point of view it is well recognized that
anorexic patients often display rigid, stereotypical, ritualistic, or perfectionistic
behaviors. Most recently, these characteristics have been examined in a
psychobiological light, connecting perfectionistic traits with alterations in
serotonin activity in a number of cross-sectional studies. These studies found
elevated scores for perfectionism in remitted anorexic and bulimic patients,
even when using different measures of perfectionism, e.g., (66–68). Because
none of the studies addressed premorbid perfectionism, the status of the factor
perfectionism based on these studies remains a correlate.
To our knowledge, again only Fairburn et al.’s risk factor study (63) included
perfectionism in their retrospective assessment of putative risk factors. They
found that in (recovered) anorexic patients premorbid perfectionism was more
common than in psychiatric and healthy controls. Perfectionism in these studies
is therefore classified as a specific, retrospective correlate.
Longitudinal Evidence. The basis of longitudinal studies including cases of
anorexia nervosa as outcome is currently too weak to classify perfectionism as a
risk factor for anorexia nervosa. It therefore remains a correlate.
Athletic Competition/Participation in Weight-Related

Subculture/Exercise
Members of professions that overemphasize a certain (low) weight or shape and
athletes from certain sport disciplines (ballet dancers, gymnasts, wrestlers,
swimmers, jockeys, etc.) were originally proposed as high-risk groups for the
development of eating disorders more than 20 years ago (69). Interest in the
past few years has focused primarily on the examination of eating disorder
symptoms in athletes and a phenomenon called the “female athlete triad” (eating
disorders, amenorrhea, osteoporosis), which a considerable number of elite
athletes seem to develop, e.g., (70,71). Another major focus has been on the role
of physical activity or high-level exercise in general in the development and
maintenance of eating disorders (72,73).
Cross-Sectional Evidence. The few controlled studies with ballet dancers
generally report that a higher percentage of dancers score in a clinically
symptomatic EAT range (69,74–76) show elevations on EDI subscales (77) or
on eating disorder-related symptoms and behaviors (75,77,78,86). However,
rates for the full syndromes of eating disorders (anorexia and bulimia nervosa)
are, when assessed, not higher than in the control groups.
The results of studies with elite athletes are less consistent. In one of
the most comprehensive surveys covering six different sport disciplines,
134 JACOBI ET AL.
Sundgot-Borgen (79) surveyed 522 female elite athletes. 1.3% of the sample
met DSM-IIIR criteria for anorexia nervosa and 8.2% for a category defined as
“anorexia athletica.”
More conservative results were obtained from a recent national survey of
male and female student athletes (80). None of the student athletes met
stringent criteria for anorexia nervosa. Using less stringent criteria, 2.85% of
the females and none of the males were classified as having subclinical anorexia.
However, the precedence of participation in weight-related subcultures to the
eating disorders was not addressed in any of these studies.
Of all the studies examining the impact of physical activity or exercise in a
noncompetitive context on the development of eating disorders, only one
retrospectively assessed the amount of physical activity before the onset of the
eating disorder. Davis et al. (81) compared a mixed eating-disordered sample of
anorexic, previously anorexic, and bulimic patients with control subjects on the
amount of physical activity predating the eating disorder using a structured
interview. The results indicated that the eating-disordered patients were
significantly more physically active than controls from adolescence (age 13)
onward and also prior to the onset of the disorder.
On the basis of cross-sectional studies, athletic competition and participation
in a competitive weight-related subculture are classified as correlates. However,
high level of exercise is classified as a retrospective correlate.
Longitudinal Evidence. Participation in weight-related subcultures and high-
level exercise have not been addressed in longitudinal studies as yet.
Other Factors
Cross-Sectional Evidence. In addition to the specific retrospective correlates
reported above, in Fairburn et al.’s study (63) a greater level of exposure to all
three risk domains (personal, environmental, dieting) was found in the
previously anorexic patients when compared to healthy controls but not in
comparison with psychiatric controls. The exposure to the three risk domains
thus seems to be a nonspecific effect.
Body dysmorphic disorder was assessed retrospectively for a period of at
least 6 months prior to the onset of the eating disorder in samples of anorexic
and bulimic patients as well as in controls. Body dysmorphic disorder (BDD)
belongs to the somatoform disorder spectrum according to both ICD and
DSM-IV, but similarities to OCD are also stressed by some authors. Symptoms
of BDD were found to be more than three times higher in anorexic patients than
in controls and were not observed at all in bulimic patients. BDD is therefore
classified as a retrospective correlate of unclear specificity.
RISK FACTORS 135
Longitudinal Evidence. Additional factors were only reported in longitudinal

studies with primarily bulimic or binge eating disorder cases. The results will
therefore be reported in the bulimia section.
Biological Risk Factors
Genetic Factors
It must be kept in mind that genes, environment, and neurobiology interact and
are inseparable. The prevailing hypothesis assumes that an accumulation of
various genes of small effect together with adverse environmental factors
increases the risk of developing an eating disorder in those carrying the greatest
genetic and environmental loading. At the moment, relevant genetic factors
must be seen as fixed markers because they precede the onset of the eating
disorder but cannot be manipulated.
Family studies provide a necessary first step in determining whether a
disorder is genetic by establishing whether it clusters among biologically related
individuals. These studies generally have found an increased rate of eating
disorders in relatives of anorexics compared with controls. However, given that
first-degree relatives share genes and environments, these studies cannot
differentiate genetic versus environmental causes for family clusters. In twin
studies, the contribution of additive genetic effects to liability has been
estimated to be between 58% and 88% for anorexia nervosa (82–17). Even
though these numbers are intriguing, the small number of studies precludes
definite conclusions. In addition, in one study (88) the concordance rates of the
dizygotic twins were higher than those of the monozygotic twins.
Researchers have begun directly examining genetic influences through
candidate genes and linkage studies. Thus far, only a few polymorphisms have
been found to be associated with anorexia nervosa. However, the literature in
this area is growing rapidly, and studies are under way that include performance
of genome-wide screenings in affected families [see (5)].
CNS Serotonin Activity

From a neurobiological perspective, research suggests that the (serotonin)
(5-HT) system may play an important role in the pathophysiology and perhaps
etiopathogenesis of eating disorders. The serotonergic system is not the only
neurotransmitter system potentially implicated in the physiopathology of eating
disorders but it is the most extensively studied system. Disturbances of brain
serotonin activity have been described in acutely ill as well as long-term
136 JACOBI ET AL.
recovered patients. Research methods for the assessment of serotonin activity

include neuroendocrine and behavioral responses to pharmacological challenge
studies (using methyl-chlorophenylpiperazine [mcpp], fenfluramine, or
tryptophan), tryptophan depletion, tryptophan availability in plasma,
serotonergic parameters in platelets (3H-imipramine binding, uptake of
serotonin, serotonin-amplified platelet aggregation, sero tonin-induced platelet
calcium mobilization), and release and metabolism of serotonin in the CNS as
reflected by levels of the serotonin metabolite 5-hydroxyindoleacetic acid
(5-HIAA) in cerebrospinal fluid (CSF). More recently, brain imaging studies
with single proton emission computed tomography (SPECT) and position
emission tomography (PET) were used to characterize binding of serotonin
transporters and 5-HT2A receptors (89–92). All of these studies have used a
cross-sectional case-control design. Because neurobiological abnormalities
cannot be assessed retrospectively, some of the studies included a recovered study
design. The rationale in these studies is that individuals who have achieved stable
remission from a disorder provide an opportunity to identify trait-related
characteristics that may be risk factors for the development of the disorder.
Nevertheless, it is not possible to determine whether alterations are the cause of
a disorder, a reversible consequence of the disorder (which might contribute to
the perpetuation of the disorder), or a “scar” caused by the abnormal behavior
using this design.
In anorexia nervosa, CNS serotonergic responsiveness is substantially
reduced in low-weight patients, which could be secondary to a diet-induced
reduction of availability of the amino acid, tryptophan, the precursor of 5HT,
e.g., (93–98). There is some evidence that reduced brain serotonin activity
persists after short-term weight recovery (92,99). In contrast, neuroendocrine
responses to 5-HT-stimulating drugs have been reported to be normalized (100)
and CSF concentrations of 5-HIAA even to be elevated (101) in women who
were long-term weight recovered from anorexia nervosa, suggesting that
hyperserotonergic function might be a trait marker in the disorder. In addition,
increased CNS serotonin has been shown to be associated with obsessive-
compulsive symptomatology. Obsessive personality traits frequently persist
after the patient has recovered from anorexia nervosa. It has been hypothesized
that increased serotonin activity might contribute to symptoms such as anxiety,
perfectionism, obsessions with symmetry and exactness, and harm avoidance,
which can, when coupled with psychosocial influences, make people vulnerable
to developing anorexia nervosa.
RISK FACTORS 137
Pregnancy and Perinatal Complications

Cross-Sectional Evidence. Although associations between pregnancy and perinatal
complications and schizophrenia have been reported repeatedly, e.g., (102),
their association with eating disorders has received relatively little attention.
Pregnancy and perinatal complications were examined in a large population-
based sample of female twins (103). Information was obtained retrospectively
by an interview carried out with the parents. Shorter gestational age and
pregnancy complications were associated with an in creased risk for anorexia
nervosa. Because the risk for alcoholism, major depression, and most anxiety
disorders was not significantly associated with pregnancy complications or
gestational age, pregnancy complications and gestational age are classified as
specific retrospective correlates of anorexia nervosa.
Cnattingius et al. (104) reported a prospectively assessed, more than three
times increased risk of very preterm birth in a large case-register sample of
anorexic patients. Also the risk (odds ratio, OR) of severe birth trauma
(cephalhematoma) was between two- and threefold higher in anorexic patients.
An elevation of these obstetrical complications was not found in a related study
with patients with schizophrenia, or with affective or reactive psychosis (105).
Preterm birth and birth trauma are therefore classified as specific fixed markers
for anorexia nervosa.
Longitudinal Evidence. Pregnancy and perinatal complications have not been
addressed in longitudinal studies as yet.
Pubertal Timing
Cross-Sectional Evidence. Pubertal timing effects (i.e., the age of occurrence of a
pubertal event) are often confounded with pubertal status effects in cross-
sectional studies limited to one age or grade. Although all girls pass through
puberty, the age at which puberty begins varies considerably. An association
between early pubertal timing and eating disorder symptoms or diagnosis has
been observed in girls in pubertal transition as well as retrospectively after
puberty completion (106,107). Whether the association between eating
disorders and early sexual maturation is a function of increasing body mass index
(BMI) or other aspects of puberty is not clear. Early pubertal timing as assessed
in the cross-sectional studies is a nonspecific fixed marker.
Longitudinal Evidence. In none of the longitudinal studies, in which pubertal
timing was addressed, did cases of anorexia nervosa emerge during the follow-
up periods.
138 JACOBI ET AL.
RISK FACTORS FOR BULIMIA NERVOSA
Psychosocial Risk Factors
Gender, Ethnicity, Acculturation, Age

The cross-sectional evidence for these factors has been presented in full in the
anorexia nervosa section above. Their classification according to the Kraemer
et al. (1) taxonomy can be summarized as follows: Female status is classifiable as
a nonspecific fixed marker and non-Asian ethnicity as a fixed marker of
currently unknown specificity. Furthermore, acculturation is classified as a
retrospective correlate of as-yet-unknown specificity and age as a nonspecific
variable risk factor.
Weight Concerns/Dieting/Negative Body Image

Cross-Sectional Evidence. Dieting has long been considered an important
precursor, if not cause, of eating disorders. Clinical studies addressing the
chronology of dieting and bingeing are few and the majority date back about
20 years. Onset of bulimia (i.e., onset of binge eating) has been observed
in these studies to occur for the vast majority of the afflicted (73–91%)
either during a period of voluntary dieting (108,109) or following weight
loss (110,111). While the studies by Mitchell et al. (108) and Pyle et al. (109)
remain rather vague in their description of the temporal sequence of the two
behaviors, recent studies have corroborated the temporal precedence of dieting
in bulimic subjects (112–114).
Experimental and laboratory-based studies on restrained eating and diet-
induced binge eating yield further evidence for the relevance of this factor. One
of the earliest, albeit uncontrolled, studies demonstrating the effects of
prolonged dieting in a nonclinical population was reported by Keys et al. (115).
Besides a variety of emotional, physical, and social changes following a period of
semistarvation and average weight loss of 25%, “bingeing” was one of the
behavioral changes reported, which had never been observed in the subjects
prior to the experiment. In addition, so-called ‘counterregulation’ or behavioral
disinhibition of “restrained eaters” under laboratory conditions has been
regarded as an experimental analog of binge eating (116–118). Although the
concepts of restrained eating and disinhibition following conditions that disrupt
the dieter’s restrained eating pattern (e.g., experimental preload, anxiety or
stress induction, alcohol consumption) closely resemble the behavioral pattern
RISK FACTORS 139
displayed by bulimic patients, their generalizability to clinical populations

has—to our knowledge—not been tested.
Taken together, the cross-sectional research provides strong evidence of the
temporal sequence dieting—binge eating. Accordingly, dieting behavior can be
classified as a retrospective correlate of bulimia nervosa.
Longitudinal Evidence. A factor best labeled as “weight concerns,” consisting of
fear of weight gain, dieting behavior, negative body image, and specific eating
disorder symptoms or attitudes (e.g., bulimic behavior), has been assessed quite
often in eating disorder research (12 of the total of 15 longitudinal studies). It
predicted the development of eating disturbances and caseness (mixed bulimic
and partial syndromes) in 9 of the 10 studies controlling for initial eating
symptomatology (7,9,10,13–16,19–21). In the study by Patton et al. (13),
subjects initially classified as “dieters” were found to have an almost eightfold
higher risk of becoming “cases” than those initially classified as nondieters.
Weight concerns negative body image, and dieting can be classified as
variable risk factors on the basis of the longitudinal studies. However, the
specificity of weight concerns remains unclear, as none of the longitudinal
studies included other psychiatric outcomes.
General Psychiatric Disturbance/Negative Emotionality

Cross-Sectional Evidence. Psychiatric disorders have been postulated to have a role
as “primary underlying conditions” in the development of eating disorders in
general (see anorexia nervosa section). However, chronology of disorder onset
has only been addressed in a single cross-sectional study.
Bulik et al. (40) compared prevalence and onset of comorbid childhood
anxiety disorders in a norexics, bulimics, and clinical as well as healthy controls
(see anorexia nervosa section for more detail). They found the prevalence of
social phobia to be highest in the bulimic group, and overanxious disorder
highest in both the bulimic and anorexic groups. In regard to precedence, 94%
of bulimic women with lifetime anxiety disorders reported that the onset of
their anxiety disorder predated the onset of their eating disorder.
Based on logistic regressions, the authors interpret their findings as
supporting the hypotheses that social phobia and overanxious disorder of
childhood are nonspecific risk factors for later psychopathology (eating and
affective disorders), although the association between social phobia and bulimia
nervosa was particularly strong (40). Although no information was given on the
rates of specific anxiety disorders preceding the eating disorders in this study,
almost twice as many bulimics compared to depressed patients (53% vs. 34%)
endorsed the presence of any lifetime anxiety disorder predating the onset of
their further disorder.
140 JACOBI ET AL.
In an attempt to explore mood- and anxiety-related prodromal symptoms of

bulimia nervosa, Raffi et al. (119) found the following symptoms to be
significantly more common in bulimics than controls 6 months prior to onset of
the disorder: low self-esteem (see below), depressed mood, anhedonia,
irritability, impaired work performance, generalized anxiety, reactivity, phobic
avoidance, guilt, pessimism, and strict dieting. Because the exact sequence of
the individual prodromal symptoms was not assessed, both the mood- and
anxiety-related symptoms observed before the onset of the bulimia could be
confounded with strict dieting. Although specificity was not tested in this study,
it seems unlikely that these prodromal symptoms are specific to bulimia nervosa
as they also typically occur as prodromal symptoms of depressive disorders.
Taken together, cross-sectional evidence for comorbid psychiatric disorders
is limited to childhood overanxious disorder and social phobia. These disorders
are therefore classified as retrospective correlates of bulimia nervosa. As the
evidence for the correlate status of these disorders is limited to a single study,
further studies are needed to replicate and extend these results. Prodromal
mood- and anxiety-related symptoms are also classified as retrospective
correlates of bulimia nervosa. However, their exact interaction with more the
specific prodromal symptoms of eating disorders remains to be clarified.
Longitudinal Evidence. General psychiatric morbidity, psychopathology, and—
more specifically—negative emotionality have been investigated as potential
predictors of eating disturbances and disorders in seven longitudinal
studies (10,13,14,16,19–21).
The change score in general psychiatric morbidity in the study by Patton
et al. (13) turned out to be the only predictor of bulimia nervosa caseness.
Psychiatric morbidity also was found to predict the onset of eating disorders
(including partial syndromes) independent of dieting status in a further study by
Patton et al. (14). Subjects in the highest psychiatric morbidity category
exhibited an almost sevenfold risk of developing an eating disorder.
Leon et al. (21) found negative affectivity to be the only significant (though
moderate) predictor of eating disorder risk measured 3–4 years later. In four
other studies (10,16,19,20), negative emotionality or psychopathology did not
predict the outcome. However, two of the temperament scales (distress, fear)
used in the study by Killen et al. (10) discriminated asymptomatic from
symptomatic girls.
Accordingly, prior psychiatric morbidity is classified as a variable risk factor
for bulimia nervosa. Although Patton et al. (13) did not address the question of
specificity, longitudinal studies in other fields indicate that premorbid anxiety
disorders and negative affectivity are risk factors for the development
of other psychiatric conditions, including affective disorders and substance
abuse (120–122). Therefore, they should be considered nonspecific risk factors.
RISK FACTORS 141
Sexual Abuse
Cross-Sectional Evidence. Sexual abuse, especially during childhood, has been
discussed as a risk factor for bulimia nervosa in many studies and reviews,
e.g., (41,123). The majority of the cross-sectional studies on abuse have been
conducted using clinical samples. Higher rates of childhood sexual abuse have
been found for eating-disordered subjects (mixed anorexia and bulimia) in
comparison with general-practice and nonmorbid controls (42), but not in
comparison with psychiatric controls (43). Several studies addressing bulimia
nervosa more specifically yielded similar results (44,46). In others no evidence
of elevated rates of sexual abuse before age 16 (45) or during childhood (124) was
found. In the latter study, however, higher rates of rape, sexual harassment, and
molestation after age 17 were reported. In another study, significantly elevated
rates of sexual abuse before age 13 were found only in bulimic patients with a
comorbid borderline personality disorder (125). Only one study with a clinical
sample explicitly assessed temporal precedence of abuse (46). However, in the
studies by Steiger et al. (125) and Webster and Palmer (45), it seems quite
likely that abuse predated eating disorder onset.
Three further studies investigating sexual abuse and bulimia nervosa were
conducted using community samples. Dansky et al. (126) found significantly
higher rates of rape, sexual molestation, aggravated assault, direct
victimization, and posttraumatic stress disorder in bulimic respondents than in
controls. Bulimics in a large epidemiological study were found to have
experienced childhood sexual abuse approximately three times more often than
comparison subjects (127). Sexual abuse was also found to be significantly more
common in bulimics than in nonmorbid subjects (126). However, no difference
was found between bulimics and general psychiatric patients. Though Garfinkel
et al. (127) failed to report age of first sexual abuse and age of bulimia onset,
the temporal sequence of abusive experiences predating eating disorder was
established in the studies by Dansky et al. (126) and Welch and Fairburn(128).
Taken together, strong evidence of elevated rates of sexual abuse prior to
onset of bulimia nervosa has been reported or is highly probable in five studies
(two with community samples, three with clinical samples). The studies failing
to address precedence also yield evidence of elevated rates of sexual abuse. No
differences were found when comparing bulimics with psychiatric controls. Thus,
sexual abuse is classified as a nonspecific, retrospective correlate of bulimia
nervosa.
Longitudinal Evidence. The association between childhood adversities including
sexual abuse and later eating- or weight-related problems has only been
investigated in one longitudinal study (8). In the large community-based sample
of mothers and their offspring, individuals who had experienced sexual abuse or
142 JACOBI ET AL.
physical neglect during childhood were at elevated risk for subsequent eating
disorders and eating problems. Information on sexual abuse and physical neglect
were obtained from a central registry and—for a subgroup of the sample—from
maternal interviews. Sexual abuse and physical neglect are classified as
nonspecific, variable risk factors on the basis of this study. However, additional
prospective replication studies are needed.
Adverse Life Events

Cross-Sectional Evidence. The relationship between stressful life events and the
onset of an eating disorder has been investigated in several cross-sectional studies
with inconsistent results. Bulimic patients in the population-based study by
Welch et al. (129) reported more life events in the year preceding the onset of
their disorder than age-matched healthy controls (18% vs. 5% for three or more
events). Adverse life events were also found to differentiate significantly
between eating-disordered patients (mixed anorexic and bulimic) and healthy
and psychiatric controls (51) (see anorexia section). However, bulimic patients
were not found to differ significantly from normal controls regarding severe
events in the studies by Troop and Treasure (47) and Schmidt et al. (48), both
described above in detail, though significantly more bulimic patients had
experienced a “major difficulty” and “pudicity” problems (48).
Taken together, there is some evidence that bulimic patients experience
more severe life events than healthy controls. However, this appears to be the
case for psychiatric patients in general. Since the reported life events predated
the onset of the eating disorders in all the studies, general adverse life events are
classified as nonspecific retrospective correlates.
Longitudinal Evidence. No longitudinal evidence for a relationship between
adverse life events and subsequent bulimia nervosa exists to date.
BMI and Other Weight-Related Variables

Cross-Sectional Evidence. Childhood obesity, one of the many risk factors assessed
retrospectively by Fairburn et al. (61), was shown to differentiate bulimic
patients from psychiatric controls: approximately 40% of the bulimics vs. 13%
of psychiatric and 15% of normal controls reported childhood obesity. Parental
obesity was also reported more frequently by the bulimic subjects than either of
the control groups. On the basis of this study, childhood and parental obesity
are classified as specific retrospective correlates of bulimia nervosa.
Longitudinal Evidence. The results from longitudinal studies investigating BMI
are inconsistent. In half of the studies without limitations, higher BMI or
body fat was found to be predictive of eating problems (9), weight control,
RISK FACTORS 143
caseness (13), or a partial diagnosis of “binge eating” (15). In the remaining

three studies without limitations, BMI or percentage of body fat at time 1 was
not related to subsequent eating disturbances or caseness (10,14,19).
Accordingly, higher BMI is classified as a variable risk factor. However, as the
number of studies supporting versus not supporting BMI as a risk factor are
equal, replication studies are needed. The specificity of BMI has not been
addressed in longitudinal studies.
Early Childhood Eating and Digestive Problems

Cross-Sectional Evidence. Early childhood eating and digestive problems have not
yet been investigated retrospectively in bulimic patients.
Longitudinal Evidence. Two longitudinal studies have addressed the association
between early childhood eating problems and bulimia nervosa. Marchi and
Cohen (12) found pica, early digestive problems, and reducing efforts to be
related to later bulimic symptoms. The risk of bulimia nervosa was found to be
almost seven times higher in individuals with a history of pica in early
childhood. Eating problems occurring typically between ages 1 and 10 were
assessed in a second longitudinal study in order to predict eating disorder
diagnoses in the time span from early adolescence to young adulthood. Only
eating too little was predictive of future bulimia nervosa; pica, digestive
problems, not eating, not being interested in food, picky eating, and eating too
slowly were not (11). The results of the longitudinal studies are inconclusive
and the studies few; therefore, further research is necessary.
Family Interaction/Family Functioning/Attachment Styles

Cross-Sectional Evidence. In the majority of the cross-sectional studies on family
interaction, functioning, or attachment style, bulimic (and anorexic) patients
describe different aspects of their family structure as more disturbed or
dysfunctional than normal controls (see anorexia section for more detail). To
date, family variables have only been assessed retrospectively, for the time span
prior to the onset of bulimia nervosa, in two studies.
Bulimics reported significantly more indifference, discord, lack of care, and
overall adversity than anorexics and healthy controls, but did not differ from
psychiatric controls (major depression) in the study by Webster and
Palmer (45), which is described in detail in the anorexia section. These variables
of troubled family background are therefore classified as nonspecific
retrospective correlates.
In a community-based study by Fairburn et al. (61), parental variables such as
high expectations, low contact, and critical comments about weight and shape
144 JACOBI ET AL.
by the family were found to be specifically predictive of bulimia nervosa as

compared with psychiatric controls, and are therefore classified as specific
retrospective correlates of bulimia nervosa.
Longitudinal Evidence. As none of the longitudinal studies assessing variables of
family interaction included bulimia nervosa as an outcome, no evidence for
these variables as risk factors for the disorder exists to date.
Family History/Family Psychopathology

Cross-Sectional Evidence. The majority of these studies suggest elevated rates of
certain psychiatric disorders in first-degree relatives of bulimic patients: eating
disorders, affective disorders, anxiety disorders, substance use disorders, and
cluster B personality disorders are found more frequently among bulimics’
relatives than relatives of control probands. Unfortunately, the temporal
relation of the psychiatric disorders of bulimics’ family members in relation to
the onset of bulimia nervosa in the patient was not addressed in any of these
studies.
Selected parental psychiatric disorders occurring prior to the onset of the
child’s bulimia nervosa were assessed retrospectively in one of the risk factor
studies by Fairburn et al. (61) as part of the environmental domain risk factors.
Parental depression, parental alcoholism, and parental drug abuse predating the
onset of the eating disorder were significantly more common in bulimic patients
than healthy controls. In addition, parental alcoholism was more common than
in psychiatric controls. Based on this study, parental alcoholism is classified as a
specific retrospective correlate, and parental depression and drug abuse are
classified as nonspecific retrospective correlates, of bulimia nervosa.
Longitudinal Evidence. No longitudinal evidence for the role of family
psychopathology as a risk factor for bulimia nervosa exists.
Low Self-Esteem/Negative Self-Concept/lneffectiveness

Cross-Sectional Evidence. Self-esteem, self-concept, and ineffectiveness have been
examined in relation to bulimia nervosa in numerous crosssectional studies.
Therein, bulimic patients have consistently been found to possess a more
negative self-concept than normal controls (65,130). Selfconcept deficits, as
assessed in these studies, are therefore classified as correlates of bulimia
nervosa.
The retrospective assessment of self-concept and its temporal relation to
onset of the bulimia nervosa has been considered in two cross-sectional studies.
Fairburn et al. (61) found negative self-evaluation prior to onset of eating
disorder to be more common in bulimic subjects than in healthy and psychiatric
RISK FACTORS 145
controls. Low self-esteem was also reported by Raffi et al. (119) to be one of
several prodromal symptoms of bulimic patients in comparison with controls.
Although low self-esteem would have to be considered a highly potent
retrospective correlate, the most potent prodromal symptom reported was
anorexia or strict dieting. It remains unclear as to how the two prodromal
symptoms were interrelated before the onset of bulimia.
Taken together, negative self-evaluation as assessed in the studies by Fairburn
et al. (61) and Raffi et al. (119) can be classified as a specific retrospective
correlate.
Longitudinal Evidence. Measures of self-concept have been included in four
longitudinal studies (7,17,18,20). In the studies by Leon et al. (20) and Calam
and Waller (19) they did not prove important in risk prediction or disordered
eating. On the other hand, low self-esteem predicted elevated EAT scores 4
years later in the study by Button et al. (17). Girls in the lowest self-esteem
range had an eightfold increased risk for a high EAT (≥20) compared to those
with high self-esteem. Similarly, Ghaderi and Scott (7) reported significantly
lower self-esteem at time 1 for the incidence group that developed an eating
disorder 2 years later. The EDI-Ineffectiveness subscale was included in four of
the studies at baseline (9,10,20,21), but turned out to be predictive of disturbed
eating patterns or caseness in only one of the multivariate analyses as part of the
latent variable negative affectivity (21). However, significant differences were
found in the univariate comparisons of the subsequent symptomatic and
asymptomatic groups (9,10).
Based on longitudinal assessment, there seems to be a slight superiority of
studies confirming the presence of a negative self-concept, low self-esteem, or
higher ineffectiveness prior to the onset of an eating disorder. Exclusion of the
two longitudinal studies with limitations (17,18) did not affect these results.
Low self-esteem and ineffectiveness are therefore classified as variable risk
factors. Based on existing studies, their specificity is unclear although it seems
reasonable to assume that they are not highly specific for eating disorders.
Replication studies are needed for further confirmation of risk factor status.
Perfectionism
Cross-Sectional Evidence. Elevated scores for perfectionism have been found in
eating-disordered patients in a number of cross-sectional studies. The results of
the studies, which warrant the classification of perfectionism as a correlate, have
been presented in full in the anorexia nervosa section. However, premorbid
perfectionism was not addressed in these studies. Only Fairburn et al.’s risk
factor study (61) included perfectionism in their retrospective assessment of
putative risk factors. Premorbid perfectionism in bulimic patients was elevated
146 JACOBI ET AL.
compared to healthy controls but did not differ from that of the psychiatric
control group. On the basis of this study, perfectionism can be classified as a
nonspecific retrospective correlate of bulimia nervosa.
Longitudinal Evidence. Perfectionism has been assessed using the corresponding
EDI subscale in four longitudinal studies (9,10,20,21). In the studies by
Killen et al. (9,10), perfectionism at time 1 was not found to be related to
subsequent eating disturbances in multivariate analyses, but differentiated
between symptomatic and asymptomatic girls at baseline in univariate
comparisons (9). In both studies by Leon et al. (20,21), perfectionism did not
turn out to be predictive in multivariate comparisons. No other measures of
perfectionism have been employed in any of the longitudinal studies. On the
basis of these results, perfectionism is classified as a correlate.
Athletic Competition/Participation in Weight-Related

Subculture/Exercise
Cross-Sectional Evidence. The majority of the cross-sectional evidence for an
association between athletic competition, participation in weight-related
subculture, and exercise has been presented in the anorexia nervosa section
above.
Results more specific to bulimia nervosa were reported in a comprehensive
survey of 522 female elite athletes from six sport disciplines (79). Of the
surveyed athletes, 8% met DSM-IIIR criteria for bulimia nervosa. Slightly
more conservative results were obtained in a recent national survey by Johnson
et al. (80), in which 1.1% of the female versus none of the male athletes met the
full DSM-IV criteria for bulimia nervosa and 9.2% of the female versus 0.005%
of the male athletes were classified as having subclinical bulimia. The question
of temporal precedence was not addressed in these studies.
On the basis of current cross-sectional evidence, athletic competition and
participation in a competitive weight-related subculture is classified as
correlates of bulimia nervosa.
High level of exercise has been classified as a retrospective correlate in a
mixed sample of eating-disordered patients [(81); see anorexia nervosa section].
However, there seems to be a preponderance of anorexic patients in this
sample, making it unclear to what extent the results are generalizable to the
diagnosis of bulimia nervosa.
Longitudinal Evidence. A longitudinal investigation of the role of athletic
competition, participation in weight-related subculture, or exercise in bulimia
nervosa has not yet been conducted.
RISK FACTORS 147
Other Factors
Cross-Sectional Studies. In addition to the specific retrospective correlates
reported above, Fairburn et al. (103) also found a greater level of exposure to
all three assessed risk domains (personal, environmental, dieting) in bulimic
patients compared with healthy controls. Exposure to the three risk domains is
therefore classified as a nonspecific retrospective correlate.
Longitudinal Evidence. Girls who later turned out to be symptomatic in the
study by Killen et al. (9) showed elevations on subscales of Aggressive and
Unpopular in a personality inventory when compared with asymptomatic girls.
In the study by Killen et al. (10), girls who developed a partial syndrome had
higher 30-day prevalence of alcohol consumption. These factors thus are
classified as nonspecific variable risk factors.
High use of escape-avoidance coping as well as low perceived social support
were found to be prospective risk factors for subsequent eating disorders
(primarily bulimia nervosa and binge eating disorder) in the study by Ghaderi
and Scott (7). Accordingly, these are classified as variable risk factors but are in
need of replication.
Genetic Factors
Similarly as in anorexia nervosa, there is an increased rate of eating disorders in
bulimia nervosa relatives compared to control relatives. In twin studies, the
contribution of additive genetic effects to liability to bulimia nervosa has been
found to vary between 28% and 83%. The remaining variance is explained mainly
by unique environmental factors. The magnitude of the contribution of shared
environment is less clear, but it appears to be less prominent than of additive
genetic factors (85). However, in one study (130), an equal environment
assumption (EEA) violation (twin resemblance) accounted for all of the variance
otherwise attributed to additive genetic effects. The EEA is central to the twin
method and assumes that monozygotic and dizygotic twins are exposed to
equivalent environmental influences of etiological importance. Only a small
number of molecular genetic studies have been conducted for bulimia nervosa,
with mostly negative results.

In bulimia nervosa, there is also considerable evidence for an impaired
serotonergic responsiveness during the acute illness state, e.g., (66,132–142).
148 JACOBI ET AL.
Studies have shown an inverse relationship between symptom severity and

measures of serotonergic responsiveness (143,144). In addition, there is
evidence for an association between self-destructiveness, a history of sexual
abuse, and impulsivity and reduced serotonin function (145–147). In longterm
recovered patients with bulimia nervosa, 5-HIAA levels were elevated in CSF in
comparison with those of controls (66). The prolactin response to fenfluramine
was significantly larger than in patients with current bulimia nervosa but not
compared to healthy controls (148). Moreover, recovered individuals
experienced a transient return of eating disorder-related symptoms after
tryptophan depletion (140). These results in recovered bulimic patients suggest
that dysregulation in some CNS serotonergic pathways may persist after
recovery from bulimia nervosa. Given the results in recovered patients with
bulimia nervosa, it could be hypothesized that a disturbance of 5-HT activity
may create vulnerability for the development of an eating disorder. However,
as mentioned earlier, it cannot be ruled out that the findings related to
serotonergic functions in recovered eating-disordered patients may be long-term
consequences of the eating disorder rather than a premorbidly existing trait.
In addition to affecting eating behavior directly, alterations in CNS serotonin
function may contribute to other psychological symptoms associated with
bulimia nervosa. The diminished CNS serotonin could play a role in the high
prevalence of depressive disorders in patients with bulimia nervosa. An
impulsive-aggressive behavioral style, which is frequently seen in bulimic
patients, may also be associated with diminished CNS serotonin function.
Taken together, the available data cannot prove whether disturbances of
serotonergic function, as described above, predate the onset of eating disorder
symptoms, or stem from dietary abnormalities or other changes characteristic
of the disorders. Consequently, they can only be regarded as correlates. Studies of
high-risk individuals prior to the development of an eating disorder may be
clarifying.
Pregnancy and Perinatal Complications

Cross-Sectional Evidence. In a large population-based study by Foley et al. [(103);
see anorexia section], pregnancy complications were also associated with an
increased risk for bulimia nervosa. Because these complications were not
associated with the other psychiatric disorders they are classified as specific
Longitudinal Evidence. Pregnancy and perinatal complications have not yet
been addressed in longitudinal studies.
RISK FACTORS 149
Pubertal Timing
Cross-Sectional Evidence. Based on evidence from the studies by Hayward
et al. (106) and Graber et al. (107), described in detail in the anorexia nervosa
section above, early pubertal timing can be regarded as a nonspecific fixed
marker of bulimia nervosa.
Longitudinal Evidence. Indicators of pubertal timing were also assessed in five
longitudinal studies, in which neither an association with subsequent eating
disturbance (9,16,19,20) nor predictive status in a structural model (21) could
be found. In the study by Graber et al. (19), timing of pubertal maturation was
related to eating disturbances present at study begin, with the girls in the “chronic”
group evidencing earliest age of menarche. Taken together, there is no
longitudinal basis for the classification of pubertal timing as a risk factor or a
correlate.
RISK FACTORS FOR BINGE EATING DISORDER
Psychosocial Factors
BED is not a distinct diagnostic category like anorexia and bulimia nervosa but
put part of eating disorders not otherwise specified (ED-NOS). Because the
research criteria have only been put forward in the latest revision of the DSM
(DSM-IV), the number of risk factor studies explicitly including the proposed
BED criteria are very small. As the outcome of the longitudinal studies is often a
mixture of bulimic or binge eating syndromes, it can be assumed that some of
the risk factors listed in the bulimia section are also relevant for BED.
Unfortunately, the majority of longitudinal studies do not permit a strict
differentiation between bulimic versus binge eating-related syndromes.
Therefore, only longitudinal and cross-sectional studies with explicit reliance on
the research criteria for BED will be covered in this section.
Cross-Sectional Evidence. Binge eating disorder has been observed to occur
predominately in females, as was also the case for anorexia and bulimia nervosa.
However, the gender distribution is not quite as asymmetrical for BED:
Preliminary estimates suggest a 2.5 female-to-male ratio (149). Accordingly,
gender is classified as a nonspecific fixed marker for BED.
Eating disorders have generally been seen as afflictions of Caucasian females
[see (29) and anorexia nervosa section]. However, though lower rates of body
dissatisfaction and weight concerns have been found among African Americans,
the rates of bingeing behavior are equal or elevated in comparison to
Caucasians (150). African American and Caucasian ethnicity can therefore be
classified as fixed markers for BED of currently unknown specificity.
150 JACOBI ET AL.
Overall, only two cross-sectional studies (62,151) retrospectively assessed

potential risk factors in BED patients compared to healthy controls: In the study
by Fairburn et al. (62), subjects with BED reported greater levels of exposure to
the following personal vulnerability factors than healthy controls: negative self-
evaluation, major depression, marked conduct problems, and deliberate self-
harm. With regard to environmental factors, they re ported greater levels of
exposure to parental criticism, high expectations, minimal affection, parental
underinvolvement, and maternal low care and high overprotection. In addition,
they were more likely to report sexual abuse; repeated severe physical abuse;
bullying; critical comments by family about shape, weight, or eating; and
teasing about shape, weight, eating, or appearance. When compared to the
psychiatric controls, subjects with BED showed a higher level of exposure to the
following factors: low parental contact; critical comments about shape, weight,
or eating; and childhood obesity. The latter variables represent specific
retrospective correlates of BED, whereas the former variables are nonspecific
Women’s perception of acceptance versus rejection by their parents during
childhood was the focus of the study by Dominy et al. (151). Women with BED
were compared with obese and nonobese women without eating disorders on
their perception of parents. BED women reported greater paternal (not
maternal) neglect and rejetion than nonobese women did, while the obese non-
BED women did not differ significantly from the other two groups. Perceived
paternal neglect and rejection are therefore classified as retrospective correlates.
Because no clinical control groups were included, the specificity of these
correlates is not known.
Longitudinal Evidence. Two longitudinal studies (7,8) were able to identify
BED cases based on the DSM-IV research criteria: Johnson et al. (8), in a large
community-based sample of mothers and offspring, found that individuals who
had experienced sexual abuse or physical neglect during childhood were at
elevated risk for eating disorders (primarily bulimia nervosa and BED), and
some eating problems during adolescence or early adulthood. Information on
sexual abuse ad physical neglect had been obtained from a central registry and—
for a subgroup of the sample—from maternal interviews. Sexual abuse and
physical neglect in this study are classified as nonspecific, variable risk factors
for BED.
Low self-esteem, high body concern, high use of escape-avoidance coping, as
well as low perceived social support were found to be prospective risk factors
for subsequent eating disorders in the study by Ghaderi and Scott (7). Of the
28 subjects who developed an eating disorder during follow-up, 13 were BED
cases. Therefore, these factors are classified as variable risk factors.
RISK FACTORS 151
Genetic Factors
There are no twin studies on BED. However, because of the low lifetime
prevalence of eating disorders and the restricted power of the statistical
analyses, attempts have been made to quantify the contribution of genes
and environment to individual symptoms of eating disorders (e.g., binge
eating) and to continuous measures of disordered eating and related
attitudes (152–154). Studies with population-based twin samples suggest a
genetic risk for the development of binge eating (86) with heritability estimates
of 46% (155).
There are even fewer molecular genetic studies for BED than bulimia
nervosa, Burnetet al. [(156), 5-HT2C receptor] and Ricca et al. [(157), 5-HT2A
receptor] did not find a difference in genotype and allele frequencies between
subjects with BED and controls.

At present, the possible contribution of central 5-HT dysfunction to the
pathophysiology of BED has not been established. Only one study has employed
a neuroendocrine challenge test with D-fenfluramine in patients with
BED (144). As opposed to patients with anorexia nervosa and severe bulimia
nervosa, the prolactin response was normal in BED women. Others found
reduced 5-HT transporter binding in the midbrain in obese bingeeating women
compared with obese control women using β-CIT SPECT (91).
As in bulimia nervosa, the satisfactory response to treatment with selective
serotonin reuptake inhibitors provides indirect evidence of a possible
disturbance of 5-HT transmission in these subjects. Several double-blind,
placebo-controlled studies of medications that influence serotonin have been
found efficacious in reducing the binge eating frequency in obese patients
with BED (158–160). However, there are also studies that did not find a
difference between fluoxetine and placebo in the reduction of binge eating
frequency (161,162).
ADDITIONAL RISK FACTOR CHARACTERISTICS

As expected, a wide range of different indicators of potency for risk factors was
found, depending on the methodology used to discriminate between highand
low-risk groups in the respective studies. Based on the indicators used, the
152 JACOBI ET AL.
potency of risk factors and retrospective correlates can be categorized as

follows:
For anorexia nervosa, female gender and a high level of exercise are the only
high-potency factors. Feeding and gastrointestinal problems, infant sleep
difficulties, a high-concern parenting style, OCPD, perfectionism, and negative
self-evaluation are retrospective correlates of medium potency. Sexual abuse
and social support are variable risk factors of medium potency. Birth related
perinatal complications and premature delivery represent fixed markers of low
potency. The remaining factors are of low or unclear potency.
Factors that have been found only once and are, therefore, in need of replication
are early sleeping, feeding, and gastrointestinal problems, highconcern
parenting, childhood anxiety disorders, and high level of exercise.
For bulimia nervosa, female gender and weight concerns/dieting represent the
best confirmed, most potent risk factors from longitudinal and cross-sectional
studies. Sexual abuse and escape-avoidance coping are variable risk factors of
medium potency; negative self-evaluation is a retrospective correlate also of
medium potency. The remaining variable risk factors and retrospective
correlates are of low or unclear potency. Perceived low social support, though
very potent, has only been confirmed in a single sample of higher age (18–30
years) than most of the other samples. Its generalizability to early adolescence is
therefore unclear.
Female gender, ethnicity, high BMI, low interoception, increased weight
concerns, negative self-evaluation, and psychiatric morbidity are factors
confirmed in more than one study, whereas the remaining factors are in need of
replication.
SUMMARY AND CONCLUSIONS

In spite of a large number of potential risk factor studies for eating disorders,
previous research is characterized by a number of limitations, which become
evident when a more rigorous risk factor terminology is applied to the putative
risk factors for eating disorders. The majority of so-called risk factor studies are
cross-sectional, thus only allowing for the identification of correlates. However,
a subset of these addressed precedence by retrospectively assessing the putative
risk factor before onset of the eating disorder. Factors consistently confirmed as
risk factors by longitudinal studies and as retrospective correlates and fixed
markers by cross-sectional studies are gender, ethnicity, higher BMI, childhood
eating problems, sexual abuse, psychiatric morbidity, low self-esteem, and
weight concerns/dieting. The status of other factors (e.g., family interaction,
perfectionism), frequently designated risk factors in previous reviews, could not
be confirmed. The specificity of potential risk factors has only been addressed
RISK FACTORS 153
for a few retrospective correlates of anorexia and bulimia nervosa. The only
well-supported high-potency factors are gender and weight concerns. Because
many of the risk factors may be population specific, replication studies are
needed.
Although the majority of longitudinal studies included sample sizes of several
hundred subjects, the samples are still too small for consistent and meaninful
risk factor detection of full syndromes of eating disorders. Given the low
prevalence of eating disorders, especially of anorexia nervosa, this is not
surprising. Consequently, longitudinal evidence on risk factors is much stronger
for bulimia nervosa and binge-related syndromes, whereas our knowledge on
risk factors for anorexia nervosa is still very limited. When attrition rates of
longitudinal studies are also taken into account, one may question whether risk
factor identification for full syndromes of anorexia nervosa is possible at all.
Furthermore, because of the overlap of the syndromes, especially partial
syndromes, previous research does not permit a valid differentiation of risk
factors for bulimia nervosa versus binge eating disorder.
The majority of factors are variable risk factors, theoretically amenable to
modification. No causal factors have been found, i.e., factors preceding the
onset of eating disorders that change the probability of the risk of the outcome.
For many of the factors (e.g., dieting, perfectionism), an experimental
manipulation would seem unethical. Although randomized controlled CBT
trials for bulimia nervosa can be considered a way of manipulating a variable risk
factor, i.e., weight and shape concerns, the similar reductions in weight and
shape concerns and dieting, achieved by such divergent treatment approaches as
CBT, IPT, and pharmacotherapy, raise the question about the mechanisms of
change in general.
To date, probably the most realistic way to manipulate a potential causal
factor would be to target the most potent variable risk factors in a highrisk
group within a preventive approach and to show that by reducing exposure to
the risk factor the risk of developing the disorder can be decreased in
comparison with a low-risk group. Preliminary studies (163,164) indicate that
such targeted approaches may result in stronger effects than those found in
previous, more universal (e.g., school-based) preventive approaches. Large-
scale studies testing the causal status of variable risk factors in highrisk samples
are currently under way and will help to further clarify the status of specific
factors.
Finally, as pointed out above, some of the risk factors, such as maternal
eating disorders, parental weight, and early feeding and gastrointestinal
problems, are already present at birth or during the first years of a child’s life.
To manipulate these, preventive interventions would have to target parental
attitudes and behaviors. To our knowledge, none of the prevention programs to
154 JACOBI ET AL.
date have focused on or included parents in order to reduce the risk for their
children. However, the identification of risk factors during different
developmental periods could result in differential recommendations for
preventive intervention in the future.
REFERENCES
1. Kraemer HC, Kazdin AE, Offord DR, Kessler RC, Jensen PS, Kupfer DJ. Coming
to terms with the terms of risk. Arch Gen Psychiatry 1997; 54:337–343.
2. Kazdin AE, Kraemer HC, Kessler RC, Kupfer DJ, Offord DR. Contributions of
risk factor research to developmental psychopathology. Clin Psychol Rev 1997; 17:
375–406.
3. Kazdin AE. Research Designs in Clinical Psychology. 3d ed. Boston: Allyn and
Bacon,1998.
4. Rutter M. Beyond longitudinal data: causes, consequences, changes, and continuity.
5. Jacobi C, Hayward C, de Zwaan M, Kraemer H, Agras WS. Coming to terms with
risk factors for eating disorders: application of risk terminology and suggestions for
a general taxonomy. Psychol Bull. In press.
6. Kraemer HC, Kazdin AE, Offord DR, Kessler RC, Jensen PS, Kupfer DJ.
Measuring the potency of risk factors for clinical or policy significance. Psychol
Meth 1999; 4:257–271.
7. Ghaderi A, Scott B. Prevalence, incidence and prospective risk factors for eating
disorders. Acta Psychiatrica Scand 2001; 104:122–130.
8. Johnson JG, Cohen P, Kasen S, Brook JS. Childhood adversities associated with
risk for eating disorders or weight problems during adolescence or early adulthood.
Am J Psychiatry 2002; 159:394–400.
9. Killen JD, Taylor CB, Hayward C, Wilson DM, Haydel KF, Hammer LD,
Simmonds B, Robinson TN, Litt I, Varady A, Kraemer H. Pursuit of thinness and
onset of eating disorder symptoms in a community sample of adolescent girls: a
three year prospective analysis. Int J Eat Disord 1994; 16:227–238.
10. Killen JD, Taylor CB, Hayward C, Haydel KF, Wilson DM, Hammer LD,
Kraemer HC, Blair-Greiner A, Strachowski D. Weight concerns influence the
development of eating disorders: a 4-year prospective study. J Consul Clin Psychol
1996; 64:936–940.
11. Kotler LA, Cohen P, Davis M, Pine DS, Walsh BT. Longitudinal relationships
between childhood, adolescent, and adult eating disorders. J Am Acad Child Adol
Psychiatry 2001; 40:1424–1440.
12. Marchi M, Cohen P. Early childhood eating behaviors and adolescent eating
disorders. J Am Acad Child Adol Psychiatry 1990; 29:112–117.
13. Patton GC, Johnson-Sabine E, Wood K, Mann AH, Wakeling A. Psychol Med
1990; 20:383–394.
RISK FACTORS 155
14. Patton GC, Selzer R, Coffey C, Carlin JB, Wolfe R. Onset of adolescent eating
disorders: population based cohort study over 3 years. Br Med J 1999; 318:
765–768.
15. Vollrath M, Koch R, Angst J. Binge eating and weight concerns among young
adults. Results from the Zurich Cohort Study. B J Psychiatry 1992; 160:498–503.
16. Attie I, Brooks-Gunn J. Development of eating problems in adolescent girls: a
longitudinal study. Dev Psychol 1989; 25:70–79.
17. Button EJ, Sonuga-Barke EJS, Davies J, Thompson M. A prospective study of self
esteem in the prediction of eating problems in adolescent schoolgirls: questionnaire
findings. Br J Clin Psychol 1996; 35:193–203.
18. Calam R, Waller G. Are eating and psychosocial characteristics in early teen age
years useful predictors of eating characteristics in early adulthood? A 7-year
longitudinal study. Int J Eat Disord 1998; 24:351–362.
19. Graber JA, Brooks-Gunn J, Paikoff RL, Warren MP. Prediction of eating
problems: an 8-year study of adolescent girls. Dev Psychol 1994; 30:823–834.
20. Leon GR, Fulkerson JA, Perry CL, Early-Zald MB. Prospective analysis of
personality and behavioral influences in the later development of disordered eating.
J Abnormal Psychol 1995; 104:140–149.
21. Leon GR, Fulkerson JA, Perry CL, Keel PK, Klump KL. Three to four year
prospective evaluation of personality and behavioral risk factors for later disordered
eating in adolescent girls and boys. J Youth Adol 1999; 28:181–196.
22. Lewinsohn PM, Hops H, Roberts RE, Seeley JR, Andrews JA. Adolescent
psychopathology: I. Prevalence and incidence of depression and other DSMIII-R
disorders in high school students. J Abnormal Psychol 1993; 102:133–144.
23. Nielsen S. The epidemiology of anorexia nervosa in Denmark from 1973 to 1987: a
nationwide register study of psychiatric admission. Acta Psychiatrica Scand 1990;
81:507–514.
24. Schotte DE, Stunkard AJ. Bulimia vs. bulimic behaviors on a college campus. JA
1987; 258:1213–1215.
25. Whitaker AH, Johnson J, Shaffer D, Rapoport JL, Kalikow K, Walsh BT, Davis M,
Braimann S, Dolinsky A. Uncommon troubles in young people: prevalence
estimates of selected psychiatric disorders in a nonreferred adolescent population.
26. Hsu LKG. Epidemiology of the eating disorders. Psychiatric Clin North Am 1996;
19:681–700.
28. Blazer DG, Kessler RC, McGonagle KA, Swartz MS. The prevalence and
distribution of major depression in a national community sample: the National
Comorbidity Survey. Am J Psychiatry 1994; 151:979–986.
29. Striegel-Moore R, Smolak L. The role of race in the development of eating
disorders. In: Smolak L, Levine MP, Striegel-Moore R, eds. The Develomental
Psychopathology of Eating Disorders. Mahwah, NJ: Lawrence Erlbaum Associates,
1996:259–284.
156 JACOBI ET AL.
30. Crago M, Shisslak CM, Estes LS. Eating disturbances among American minority
groups: a review. Int J Eat Disord 1996; 19:239–248.
31. Chen CN, Wong J, Lee N, Chan-Ho MW, Lau JT, Fung M. The Shatin
Community mental health survey in Hong Kong. II. Major findings. Arch Gen
Psychiatry 1993; 50:125–133.
32. Ohzeki T, Hanaki K, Motozumi H, Ishitani N, Matsuda-Ohtahara H, Sunaguchi M,
Shiraki K. Prevalence of obesity, leanness and anorexia nervosa in Japanese boys
and girls aged 12–14 years. Ann Nutr Metab 1990; 34:208–212.
33. Davis C, Katzman MA. Perfection as acculturation: psychological correlates of
eating problems in Chinese male and female students living in the United States. Int
J Eat Disord 1999; 25:65–70.
34. Gowen LK, Hayward C, Killen JD, Robinson TN, Taylor CB. Acculturation and
eating disorder symptoms in adolescent girls. J Res Adol 1999; 9:67–83.
35. Hooper MSH, Garner DM. Application of the eating disorders inventory to a
sample of black, white and mixed race schoolgirls in Zimbabwe. Int J Eat Disord
1986; 5:161–168.
36. Woodside DB, Garfinkel P. Age of onset in eating disorders. Int J Eat Disord 1992;
12:31–36.
37. Mitchell JE, Specker SM, de Zwaan M. Comorbidity and medical complications of
bulimia nervosa. J Clin Psychiatry 1991; 52(Suppl 10): 13–20.
38. Wonderlich SA, Mitchell JE. Eating disorders and comorbidity: Empirical,
conceptual, and clinical implications. Psychopharmacol Bull 1997; 33:381–390.
39. Rastam M. Anorexia nervosa in 51 Swedish adolescents: premorbid problems and
40. Bulik CM, Sullivan PF, Fear JL, Joyce PR. Eating disorders and antecedent anxiety
disorders: a controlled study. Acta Psychiatrica Scand 1997; 96:101–107.
41. Pope HG, Hudson JI. Is childhood sexual abuse a risk factor for bulimia nervosa?
Am J Psychiatry 1992; 149:455–463.
42. Brown L, Russell J, Thornton C, Dunn S. Experiences of physical and sexual abuse
in Australian general practice attenders and an eating disordered population. Aust N
Z J Psychiatry 1997; 31:398–404.
43. Folsom V, Krahn D, Nairn K, Gold L, Demitrack MA, Silk KR. The impact of
sexual and physical abuse on eating disordered and psychiatric symptoms: a
comparison of eating disordered and psychiatric inpatients. Int J Eat Disord 1993;
13:249–257.
44. Steiger H, Zanko M. Sexual traumata among eating-disordered, psychiatric, and
normal female groups. J Interpers Viol 1990; 5:74–86.
45. Webster JJ, Palmer RL. The childhood and family background of women with
clinical eating disorders: a comparison with women with major depression and
women without psychiatric disorder. Psychol Med 2000; 30:53–60.
46. Vize CM, Cooper PJ. Sexual abuse in patients with eating disorder, patients with
depression and normal controls. A comparative study. Br J Psychiatry 1995; 167:
80–85.
RISK FACTORS 157
47. Troop NA, Treasure JL. Psychosocial factors in the onset of eating disorders:
responses to life events and difficulties. Br J Med Psychol 1997; 70:373–385.
48. Schmidt U, Tiller J, Blanchard M, Andrews B, Treasure J. Is there a specific
trauma precipitating anorexia nervosa? Psychol Med 1997; 27:523–530.
49. Rastam M, Gillberg C. Background factors in anorexia nervosa. A controlled study
of 51 teenage cases including a population sample. Eur Child Adol Psychiatry 1992;
1:54–65.
50. Horesh N, Apter A, Lepkifker E, Ratzoni G, Weizman R, Tyrano S. Life events
and severe anorexia nervosa in adolescence. Acta Psychiatrica Scand 1995; 91:5–9.
51. Horesh N, Apter A, Ishai J, Danziger Y, Miculincer M, Stein D, Lepkifker
E, Minouni M. Abnormal psychosocial situations and eating disorders in
adolescence. J Am Acad Child Adol Psychiatry 1996; 35:921–927.
52. Gowers SG, North CD, Byram V. Life event precipitants of adolescent anorexia
nervosa. J Child Psychol Psychiatry 1996; 37:469–477.
53. Minuchin S, Rosman BL, Baker L. Psychosomatic Families: Anorexia Nervosa in
Context. Cambridge, MA: Harvard University Press, 1978.
54. Bruch H. Eating Disorders: Obesity, Anorexia Nervosa, and the Person Within.
New York: Basic Books, 1973.
55. Cole-Detke H, Kobak R. Attachment processes in eating disorder and depression.
J Consul Clin Psychol 1996; 64:282–290.
56. Friedmann MA, Wilfley DE, Welch RR, Kunce JT. Self-directed hostility and
family functioning in normal-weight bulimics and overweight binge eaters. Addict
Behav 1997; 22:367–375.
57. McNamara K, Loveman C. Differences in family functioning among bulimics,
repeat dieters, and nondieters. J Clin Psychol 1990; 46:518–523.
58. Shisslak CM, McKeon RT, Crago M. Family dysfunction in normal weight bulimic
and bulimic anorexic families. J Clin Psychol 1990; 46:185–189.
59. Strober M, Humphrey LL. Familial contributions to the etiology and course of
anorexia nervosa and bulimia. J Consult Clin Psychol 1987; 55:654–659.
60. Shoebridge P, Gowers SG. Parental high concern and adolescent-onset anorexia
nervosa. Br J Psychiatry 2000; 176:132–137.
61. Fairburn CG, Welch SL, Doll HA, Davies BA, O’Connor ME. Risk factors for
bulimia nervosa. A community-based case-control study. Arch Gen Psychiatry
1997; 54:509–517.
for binge-eating disorder: a community-based case-control study. Arch Gen
Psychiatry 1998; 55:425–432.
63. Fairburn CG, Cooper Z, Doll HA, Welch SL. Risk factors for anorexia nervosa.
Three integrated case-control comparisons. Arch Gen Psychiatry 1999; 56:
468–476.
64. Bruch H. Perceptual and conceptual disturbances in anorexia nervosa. Psychosom
Med 1962; 14(2): 187–194.
65. Jacobi C. Zur Spezifitat und Veranderbarkeit von Beeinträchtigungen
desSelbstkonzepts bei Essstörungen. Regensburg: S. Roderer Verlag, 1999.
158 JACOBI ET AL.
66. Kaye WH, Greeno CG, Moss H, Fernstrom J, Fernstrom M, Lilenfeld LR,
Weltzin TE, Mann JJ. Alterations in serotonin activity and psychiatric symptoms
after recovery from bulimia nervosa. Arch Gen Psychiatry 1998; 55:927–935.
67. Srinivasagam NM, Kaye WH, Plotnicov KH, Greeno C, Weltzin TE, Rao R.
Persistent perfectionism, symmetry, and exactness after long-term recovery from
anorexia nervosa. Am J Psychiatry 1995; 152:1630–1634.
68. Bastiani AM, Rao R, Weltzin T, Kaye WH. Perfectionism in anorexia nervosa. Int
J Eat Disord 1995; 17:147–152.
69. Garner DM, Garfinkel PE. Sociocultural factors in anorexia nervosa. Lancet 1978;
2:674.
70. Brownell KD. Eating disorders in athletes. In: Brownell KD, Fairburn CG, eds.
Eating Disorders and Obesity. A Comprehensive Handbook. New York: Guilford
Press, 1995:191–196.
71. Putukian M. The female athlete triad. Clin Sports Med 1998; 17:675–696.
72. Davis C, Fox J, Cowles M, Hastings P, Schwass K. The functional role of exercise
in the development of weight and diet concerns in women. J Psychosom Res 1990;
34:563–574.
73. Davis C, Katzman DK, Kaptein S, Kirsh C, Brewer H, Kalmbach K, Olmstedt MP,
Woodside DB, Kaplan AS. The prevalence of high-level exercise in the eating
disorders: etiological implications. Comprehen Psychiatry 1997; 38:321–326.
74. Joseph A, Wood IK, Goldberg SC. Determining populations at risk for developing
anorexia nervosa based on selection of college major. Psychiatry Res 1982; 7:
53–58.
75. Abraham S. Eating and weight controlling behaviors of young ballet dancers.
Psychopathology 1996a; 29:218–222.
76. Abraham S. Chracteristics of eating disorders among young ballet dancers.
Psychopathology 1996b; 29:223–229.
77. Kurtzman FD, Yager J, Landsverk J, Wiesmeier E, Bodurka DC. Eating disorders
among selected female student populations at UCLA. J Am Diet Assoc 1989; 89:
45–53.
78. Braistedt JR, Mellin L, Gong EJ, Irwin CE Jr. The adolescent ballet dancer.
Nutritional practices and characteristics associated with anorexia nervosa. J Adol
Health Care 1985; 6:365–371.
79. Sundgot-Borgen J. Risk and trigger factors for the development of eating disorders
in female elite athletes. Med Sci Sports Exercise 1994; 26:414–419.
80. Johnson C, Powers PS, Dick R. Athletes and eating disorders: the National
Collegiate Athletic Association study. Int J Eat Disord 1999; 26:179–188.
81. Davis C, Kennedy SH, Ravelski E, Dionne M. The role of physical activity in the
development and maintenance of eating disorders. Psychol Med 1994; 24:957–967.
82. Holland AJ, Hall A, Murray R, Russell GFM, Crisp AH. Anorexia nervosa: a study
of 34 twin pairs and one set of triplets. Br J Psychiatry 1984; 145:414–419.
83. Holland AJ, Sicotte N, Treasure J. Anorexia nervosa: evidence for a genetic basis.
J Psychosom Res 1988; 32:561–571.
RISK FACTORS 159
84. Treasure J, Holland A. Genetic vulnerability to eating disorders: evidence from

twin and family studies. In: Remschmidt H, Schmidt MH, eds. Child and Youth
Psychiatry: European Perspectives. New York: Hogrefe and Huber, 1989:59–68.
85. Bulik CM, Sullivan PF, Wade TD, Kendler KS. Twin studies of eating disorders: a
review. Int J Eat Disord 2000; 27:1–20.
86. Wade TD, Bulik CM, Neale M, Kendler KS. Anorexia nervosa and major
depression: Shared genetic and environmental risk factors. Am J Psychiatry 2000;
157:469–471.
87. Klump KL, Miller KB, Keel PK, McGue M, Iacono WG. Genetic
and environmental influences on anoexia nervosa syndromes in a population-based
twin sample. Psychol Med 2001; 31:737–740.
88. Walters EE, Kendler KS. Anorexia nervosa and anorexic-like syndromes in a
population-based female twin sample. Am J Psychiatry 1995; 152:64–71.
89. Kaye WH, Frank GKW, Meltzer CC, Price JC, McConaha CW, Crossan PJ, Klump
KL, Rhodes L. Altered serotonin 2A receptor activity in women who have
recovered from bulimia nervosa. Am J Psychiatry 2001; 158:1151–1154.
90. Tauscher J, Pirker W, Willeit M, de Zwaan M, Bailer U, Neumeister A, Asenbaum
S, Lennkh C, Praschak-Rieder N, Briicke T, Kasper S. [123]-?s-CIT and single
photon emission computed tomography reveal reduced brain serotonin transporter
availability in bulimia nervosa. Biol Psychiatry 2001; 49:326–332.
91. Kuikka JT, Tammela L, Karhunen L, Rissanen A, Bergstroem KA, Naukkarinen H,
Vanninen E, Karhu J, Lappalainen R, Repo-Tiihonen E, Tiihonen J, Uusitupa M.
Reduced serotonin transporter binding in binge eating women.
Psychopharmacology 2001; 155:310–314.
92. Frank GK, Kaye WH, Meltzer CC, Price JC, Greer P, McConaha C, Skovira K.
Reduced 5-HT2A receptor binding after recovery from anorexia nervosa. Biol
Psychiatry 2002; 52:896–906.
93. Brewerton TD, Jimerson DC. Studies of serotonin function in anorexia nervosa.
94. Hadigan CM, Walsh BT, Buttinger C, Hollander E. Behavioural and
neuroendocrine responses to meta-CPP in anorexia nervosa. Biol Psychiatry 1995;
37:504–511.
95. Kaye WH, Ebert MH, Raleigh M, Lake R. Abnormalities in CNS monoamine
metabolism in anorexia nervosa. Arch Gen Psychiatry 1984; 41:350–355.
96. Kaye WH, Gwirtsman HE, George DT, Jimerson DC, Ebert MH. CSF 5HIAA
concentrations in anorexia nervosa: reduced values in underweight subjects
normalise after weight restoration. Biol Psychiatry 1988; 23:102–105.
97. Monteleone P, Brambilla F, Bortolotti F, La Rocca A, Maj M. Prolactin response to
d-fenfluramine is blunted in people with anorexia nervosa. B J Psychiatry 1998b;
172:438–442.
98. Ward A, Brown N, Lightman S, Campbell IC, Treasure J. Neuroendocrine,
appetitive and behavioural responses to d-fenfluramine in women recovered from
anorexia nervosa. Br J Psychiatry 1998; 172:351–358.
160 JACOBI ET AL.
99. Frank GK, Kaye WH, Weltzin TE, Perel J, Moss H, McConaha C, Pollice C.
Altered response to meta-chlorophenylpiperazine in anorexia nervosa: support for
a persistent alteration of serotonin activity after short-term weight restoration. Int
J Eat Disord 2001; 30:57–68.
100. O’Dwyer AM, Lucey JV, Russell GMF. Serotonin activity in anorexia nervosa after
long-term weight restoration: response to D-fenfluramine challenge. PsycholMed
1996; 26:353–359.
101. Kaye WH, Gwirtsman, El H, George DT, Ebert MH. Altered serotonin activity in
anorexia nervosa after long-term weight restoration. Does elevated cerebrospinal
fluid 5-hydroxyindoleacetic acid level correlate with rigid and obsessive behavior?
102. Geddes JR, Lawrie SM. Obstetric complications and schizophrenia: a metaanalysis.
Br J Psychiatry 1995; 167:86–93.
103. Foley DL, Thacker LR, Aggen SH, Neale MC, Kendler KS. Pregnancy and
perinatal complications associated with risks for common psychiatric disorders in a
population-based sample of female twins. Am J Med Genet 2001; 105:426–431.
104. Cnattingius S, Hultman CM, Dahl M, Sparén P. Very preterm birth, birth trauma,
and the risk of anorexia nervosa among girls. Arch Gen Psychiatry 1999; 56:
634–638.
105. Hultman CM, Sparén P, Takei N, Murray RM, Cnattingius S. Prenatal and
perinatal risk factors for schizophrenia, affective psychosis, and reactive psychosis
of an early onset: case control study. Br Med J 1999; 318:421–426.
106. Hayward C, Killen JD, Wilson DM, Hammer LD, Litt IF, Kraemer HC, Haydel F,
Varady A, Taylor CB. Psychiatric risk associated with early puberty in adolescent
girls. J Am Acad Child Adol Psychiatry 1997; 36:255–262.
107. Graber JA, Lewinsohn PM, Seeley JR, Brooks-Gunn J. Is psychopathology
associated with the timing of pubertal development? J Am Acad Child Adol
Psychiatry 1997; 36:1768–1776.
108. Mitchell JE, Hatsukami D, Pyle RL, Eckert ED. The bulimia syndrome: course of
the illness and associated problems. Comprehens Psychiatry 1986; 27:165–170.
109. Pyle RL, Mitchell MD, Eckert ED. Bulimia: a report of 34 cases. J Clin Psychiatry
1981; 42:60–64.
110. Russell GF. Bulimia nervosa: an ominous variant of anorexia nervosa. Psychol Med
1979; 9:429–448.
111. Garfmkel, Modolfsky, Garner. The heterogeneity of anorexia nervosa: Bulimia as a
distinct subgroup. Arch Gen Psychiatry 1980; 37:1036–1040.
112. Brewerton TD, Dansky BS, Kilpatrick DG, O’Neil PM. Which comes first in the
pathogenensis of bulimia nervosa, dieting or bingeing ? Int J Eat Disord 2000; 28:
259–264.
113. Mussell MP, Mitchell JE, Fenna CJ, Crosby RD, Miller JP, Hoberman HM. A
comparison of onset of binge eating versus dieting in the development of bulimia
114. Haiman C, Devlin MJ. Binge eating before the onset of dieting: A distinct subgroup
of bulimia nervosa ? Int J Eat Disord 1999; 25:151–157.
RISK FACTORS 161
115. Keys A, Brozek J, Hentschel A, Mickelsen O, Taylor HL. The Biology of Human
Starvation. Minneapolis: University of Minnesota Press, 1950.
116. Polivy J, Herman PC. Dieting and bingeing. A causal analysis. Am Psychol 1985;
40:193–201.
117. Ruderman AJ. Dietary restraint: a theoretical and empirical review. Psychol Bull
1986; 99:247–262.
118. Tuschl RJ. From dietary restraint to binge eating: some theoretical considerations.
Appetite 1990; 14:105–109.
119. Raffi AR, Rondini M, Grandi S, Fava GA. Life events and prodromal symptoms in
bulimia nervosa. Psychol Med 2000; 30:727–731.
120. Hayward C, Killen JD, Kraemer HC, Taylor CB. Predictors of panic attacks in
adolescents. J Am Acad Child Adol Psychiatry 2000; 39:207–214.
121. Ingram RE, Price JM, Eds. Vulnerability to Psychopathology. New York:
Guildford Press, 2000.
122. Pine DS, Cohen P, Gurley D, Brook J, Ma Y. The risk for early adulthood anxiety
and depressive disorders in adolescents with anxiety and depressive disorders. Arch
Gen Psychiatry 1998; 55:56–64.
123. Wonderlich SA, Brewerton TD, Jocic Z, Dansky B, Abbott DW. Relationship of
childhood sexual abuse and eating disorders. J Am Academy of Child and
Adolescent Psychiatry 1997; 36:1107–1115.
124. Casper RC, Lubomirsky S. Individual psychopathology relative to reports of
unwanted sexual experiences as predictor of a bulimic eating pattern. Int J Eat
Disord 1997; 21:229–236.
125. Steiger H, Leonard S, Ng Ying Kin NMK, Ladouceur C, Ramdoyal D, Young SN.
Childhood abuse and platelet tritiated-paroxetine binding in bulimia nervosa:
implications of borderline personality disorder. J Clin Psychiatry 2000; 61:
428–435.
126. Dansky BS, Brewerton TD, Kilpatrick DG, O’Neal PM. The National women’s
study: Relationship of victimization and posttraumatic stress disorder to bulimia
127. Garfinkel PE, Lin E, Goering P, Spegg C, Goldbloom DS, Kennedy S, Kaplan AS,
128. Welch SL, Fairburn CG. Sexual abuse and bulimia nervosa: three integrated case-
control comparisons. Am J Psychiatry 1994; 151:402–407.
129. Welch SL, Doll HA, Fairburn CG. Life events and the onset of bulimia nervosa: A
controlled study. Psychol Med 1997; 27:515–522.
130. Jacobi C, Paul Th, de Zwaan M, Nutzinger DO, Dahme B. Specificity of
selfconcept disturbances in eating disorders. Int J Eat Disord. In press.
131. Hettema JM, Neale MC, Kendler KS. Physical similarity and the equalenvironment
assumption in twin studies of psychiatric disorders. Behav Gene 1995; 25:
327–335.
132. Brewerton TD, Mueller EA, Lesem MD, Brandt HA, Quearry B,
George DT, Murphy DL, Jimerson DC. Neuroendocrine responses to
162 JACOBI ET AL.
m-chlorophenylpiperazine and 1-tryptophan in bulimia. Arch Gen Psychiatry 1992;

49:852–861.
133. Goldbloom DS, Garfinkel PE, Katz R, Brown G. The hormonal response to
intravenous 5-hydroxytryptophan in bulimia nervosa. Psychosom Med 1990; 52:
225–226.
134. Jimerson DC, Wolfe BE, Metzger ED, Finkelstein DM, Cooper TB, Levine JM.
Decreased serotonin function in bulimia nervosa. Arch Gen Psychiatry 1997; 54:
529–534.
135. Kaye WH, Gendall KA, Fernstrom MH, Fernstrom JD, McConaha CW, Weltzin
TE. Effects of acute tryptophan depletion on mood in bulimia nervosa. Biol Psychiatry
2000; 4:151–157.
136. Levitan RD, Kaplan AS, Joffe RT, Levitt AJ, Brown GM. Hormonal and subjective
responses to intravenous meta-Chlorphenylpiperazine in bulimia nervosa. Arch
137. McBride PA, Anderson GM, Khait VD, Sunday SR, Halmi KA. Serotonergic
responsivity in eating disorders. Psychopharmacol Bull 1991; 27:365–372.
138. Monteleone P, Brambilla F, Bortolotti F, Ferraro C, Maj M. Plasma prolactin
response to D-fenfluramine is blunted in bulimic patients with frequent binge
episodes. Psychol Med 1998a; 28:975–983.
139. Oldman A, Walsh A, Salkovskis P, Fairburn CG, Cowen PJ. Biochemical and
behavioural effects of acute tryptophan depletion in abstinent bulimic subjects: a
pilot study. Psychol Med 1995; 25:995–1001.
140. Smith KA, Fairburn CG, Cowen PJ. Symptomatic relapse in bulimia nervosa
following acute tryptophan depletion. Arch Gen Psychiatry 1999; 56:171–176.
141. Weltzin TE, Fernstrom MH, Fernstrom JD, Neuberger SK, Kaye WH. Acute
tryptophan depletion and increased food intake and irritability in bulimia nervosa.
Am J Psychiatry 1995; 152:1668–1671.
142. Weltzin TE, Fernstrom JD, McConaha C, Kaye WH. Acute tryptophan depletion
in bulima: effects on large neutral amino acids. Biol Psychiatry 1994; 35: 388–397.
143. Jimerson DC, Lesem MD, Kaye WH, Brewerton TD. Low serotonin and
dopamine metabolite concentration in cerebrospinal fluid from bulimic patients
with frequent binge episodes. Arch Gen Psychiatry 1992; 49:132–138.
144. Monteleone P, Brambilla F, Bortolotti F, Maj M. Serotonergic dysfunction across
the eating disorders: relationship to eating behaviour, nutritional status and general
psychopathology. Psychol Med 2000; 30:1099–1110.
145. Steiger H, Gauvin L, Israel M, Koerner N, Ng Ying Kin NMK, Paris J, Young SN.
Association of serotonin and cortisol indices with childhood abuse in bulimia
nervosa. Arch Gen Psychiatry 2001 a; 58:837–843.
146. Steiger H, Koerner N, Engelberg MJ, Israel M, Ng Ying Kin NMK, Young SN.
Self-destructiveness and serotonin function in bulimia nervosa. Psychiatry
Res2001b; 103:15–26.
147. Steiger H, Young SN, Ng Ying Kin NMK, Koerner N, Israel M, Lageix P, Paris J.
Implications of impulsive and affective symptoms for serotonin function in bulimia
nervosa. Psychol Med 2001c; 31:85–95.
RISK FACTORS 163
148. Wolfe BE, Metzger ED, Levine JM, Finkelstein DM, Cooper TB, Jimerson DC.
Serotonin function following remission from bulimia nervosa.
Neuropsychopharmacology 2000; 22:257–263.
149. Spitzer RL, Devlin M, Walsh BT, Hasin D, Wing R, Marcus M, Stunkard A,
Wadden T, Yanovski S, Agras S, Mitchell J, Nonas C. Binge-eating disorder: A
multisite field trial of the diagnostic criteria. Int J Eat Disord 1992; 11:191–203.
150. Striegel-Moore RH, Schreiber GB, Lo A, Crawford P, Obarzanek E, Roding J.
Eating disorder symptoms in a cohort of 11 to 16-year-old black and white girls:
The NHLBI growth and health study. Int J Eat Disord 2000; 27:49–66.
151. Dominy NL, Johnson WB, Koch C. Perception of parental acceptance in women with
binge-eating disorder. J Psychol 2000; 134:23–36.
152. Rutherford J, McGuffin P, Katz RJ, Murray RM. Genetic influences on eating
attitudes in a normal female twin population. Psychol Med 1993; 23:425–436.
153. Klump KL, McGue M, Iacono WG. Age differences in genetic and environmental
influences on eating attitudes and behaviors in preadolescent and adolescent female
twins. Journal of Abnormal Psychology 2000; 109:239–251.
154. Wade T, Martin NG, Tiggemann M. Genetic and environmental risk factors for the
weight and shape concerns characteristic of bulimia nervosa. Psychological
Medicine 1998; 28:761–771.
155. Sullivan PF, Bulik CM, Kendler KS. Genetic epidemiology of binging and
vomiting. British Journal of Psychiatry 1998; 173:75–79.
156. Burnet PWJ, Smith KA, Cowen PJ, Harrison PJ. Allelic variation of the 5HT2c
receptor in bulimia nervosa and binge-eating disorder. Psychiatric Genetics 1999;
9:101–104.
157. Ricca V, Nacmias B, Cellini E, Di Bernardo M, Rotella CM, Sorbi S. 5-HT2A
receptor gene polymorphism and eating disorders. Neuroscience Letters 2002;
323:105–108.
158. Stunkard A, Berkowitz R, Tanrikut C, Reiss E, Young L. d-Fenfluramine
treatment of binge-eating disorder. Am J Psychiatry 1996; 153:1445–1449.
159. Hudson JI, McElroy SL, Raymond NC, Crow S, Keck PE, Carter WP, Mitchell JE,
Strakowski SM, Pope HG, Coleman B, Jonas JM. Fluvoxamine in the treatment of
binge-eating disorder. Am J Psychiatry 1998; 155:1756–1762.
160. McElroy SL, Casato LS, Nelson EB, Lake KA, Soutullo CA, Keck PE Jr, Hudson
JI. Placebo-controlled trial of sertraline in the treatment of binge eating disorder.
Am J Psychiatry 2000; 157:1004–1006.
161. Grilo CM, Masheb RM, Heninger G, Wilson GT. Psychotherapy and medication
for binge eating disorder, Abstract 095. International Conference on Eating
Disorders, Boston, April 25–28, 2002.
162. Devlin M. Psychotherapy and medication for binge eating disorder, Abstract
Plenary Session. International Conference on Eating Disorders, Boston,
April 25–28, 2002.
164 JACOBI ET AL.
163. Taylor CB, Altman T. Priorities in prevention research for eating disorders.
Psychopharmacol Bull 1997; 33:413–417.
164. Winzelberg AJ, Eppstein D, Eldredge KL, Wilfley D, Dasmahapatra R, Dev P,
Taylor CB. Effectiveness of an Internet-based program for reducing risk factors for
eating disorders. J Consult Clin Psychol 2000; 68:346–350.
7
Role of Genetics in Anorexia Nervosa,
Cynthia M.Bulik
University of North Carolina
Chapel Hill, North Carolina, U.S.A.
THE BURDEN OF PLAUSIBILITY

Research on the etiology of eating disorders has lagged behind other areas of
psychiatry, in part due to the imminent plausibility of sociocultural theories
about the illness. Perhaps more than any other psychiatric disorder, social
explanations that focus on Western societies’ drive for thinness and beauty
ideals as motivators for dieting and symbols of control provide a highly probable
explanation for why young girls engage in unhealthy weight loss practices. The
face validity of these explanations has inhibited our progress due to a burden of
plausibility. The sheer convenient believability of sociocultural explanations has
influenced research directions and hindered recognition of the seriousness of
eating disorders. The perception that anorexia nervosa (AN) and bulimia
nervosa (BN) are volitional disorders has obscured the facts regarding morbidity,
comorbidity, and mortality.
The tide is turning. Over the past decade there has been a concentrated
effort to advance our understanding of the influence of genes and environment
on eating disorders that has resulted in a number of centers conducting highly
sophisticated family, twin, and genetic studies. The mosaic of findings has
pointed us toward the conclusion that genetic effects play a moderate
to substantial role in liability to AN and BN. Less is known about the role of
genetics in binge eating disorder (BED), although results are not far behind. The
consistency of these findings across centers warrants a recommendation that all
practitioners in the field consider developing at least a passing familiarity with
their meaning and their implications for etiology, prevention, and management
of eating disorders.
This chapter reviews the value and methods associated with family and twin
studies, current findings relevant to eating disorders, and implications for
prevention and clinical practice.
166 BULIK
ROLE OF GENETICS IN ANOREXIA NERVOSA,

BULIMIA NERVOSA, AND BINGE EATING
DISORDER
Methods of Genetic Epidemiology: Family, Twin,

and Adoption Studies
There are three major research designs in genetic epidemiology that allow for
the unraveling of the relative contribution of genes and environment to the
etiology of complex behavioral traits or disorders.
Family Studies
The first step is to determine whether a trait or disorder runs in families. This
question can be addressed by the standard family study, which determines
whether there is a significantly greater lifetime risk of eating disorders in
biological relatives of individuals who have an eating disorder in comparison to
relatives of individuals without eating disorders. If risk to relatives is not
increased, then the probability of the disorder being influenced by genetic
factors is low. Family studies are limited in that they cannot tell you the extent
to which the familial aggregation is due to genes and to what extent it is due to
environment.
Adoption Studies
Two additional designs are possible that enable the disentangling of genetic and
environmental effects: adoption and twin designs. In an adoption paradigm, one
compares the degree of similarity between an adoptee and his or her biological
versus adoptive parents. A greater similarity to biological parents suggests genetic
effects whereas greater similarity to adoptive parents suggests environmental
effects. Although adoption studies are statistically powerful, adoption is rare and
many of the assumptions of the method are often not met (e.g., random
placement). Moreover, when studying relatively rare conditions such as eating
disorders, the prevalence of the disorders is often too low to allow meaningful
conclusions to be drawn from adoption studies.
Twin Studies
Twin studies are often the only practical design for teasing out the relative
contributions of genetic and environmental factors to complex traits.
Monozygotic (MZ) twins emerge from two genetically identical embryos.
ROLE OF GENETICS 167
Therefore, any differences between MZ twins provides strong evidence for the
role of environmental influences (1, pp. 171–172). Dizygotic (DZ) twins result
from the fertilization of two ova by different spermatozoa. DZ twins are no
more similar genetically than nontwin siblings and share—on average—half of
their genes identical by descent. Thus, differences between DZ twins can result
from genetic and/or environmental effects.
The classical twin study uses the similarities and differences between MZ and
DZ twin pairs to quantify and qualify genetic and environmental causes for a
particular trait. Using structural equation modeling techniques, one is able to
parse liability to a trait into three sources of variability: additive genetic
effects (a2), common or shared environmental effects (c2), and unique
environmental effects (e2).
Additive Genetic Effects (abbreviation “A”). Additive genetic effects result from
the cumulative impact of many individual genes each of small effect. The
presence of A is inferred when the correlation between MZ twins is greater than
the correlation between DZ twins. For example, the correlation between MZ
twins for Body Dissatisfaction from the Eating Disorders Inventory was 0.60 for
MZ twins and 0.13 for DZ twins in a study conducted by Klump et al. (2). The
fact that the MZ twin correlation was more than twice the DZ correlation
suggests that genes affect Body Dissatisfaction as measured by the EDI in this sample.
Common Environmental Effects (abbreviation “C”). Common environmental effects
result from etiological influences to which both members of a twin pair are
exposed regardless of zygosity. Thus, common environmental effects contribute
equally to the correlation between MZ and between DZ twins. For example, if
the correlation for body dissatisfaction had been 0.61 for MZ and 0.61 for DZ
twins, then we would infer that common environment was influencing the trait.
Individual-Specific Environmental Effects (abbreviation “E”). The second type of
environmental effect results from etiological influences to which one member
of a twin pair is exposed but not the other. Thus, individual-specific
environmental effects contribute to differences between members of a twin pair
and decrease the magnitude of the correlation between both MZ and DZ twin
pairs. In the simplest case, if the correlation between both MZ and DZ is 0, then
the trait is entirely determined by individual-specific environmental effects.
Examples include one member of a twin pair being exposed to a traumatic
experience not shared with the cotwin.
More sophisticated analyses allow one to quantify the relative contributions of
A, C, and E. The proportion of variance due to A (additive genetic effects) is a2
(also known as “heritability”). The proportion of variance due to C is c2 and the
proportion due to E is e2. The value of e2 also incorporates measurement error.
Finally, a2, c2, and e2 must add up to the total variance of 1.
168 BULIK
What Is Heritability? What Isn’t Heritability?

The methodology and terminology of twin studies can easily lead to
misinterpretation. Heritability estimates are often quoted with little
understanding of their meaning or of their limitations.
Allison and Faith (3) discuss a number of ways in which the concept of
heritability can be misinterpreted. For example, hypothesize for a moment that
a researcher reports in a scientific paper that the heritability of BN is 83% with a
95% confidence interval of 49–100%. Normally, the media are very interested
in such findings. Unfortunately, in their attempt to interpret the findings they
often lead readers astray. Here are a couple of errant headlines that highlight
common misinterpretations.
1. Researchers say 83% of the reason people develop bulimia is genetic.

2. Research says 83% of bulimia cases are caused by genes.
3. If you have the bulimia gene, there’s an 83% chance you’ll develop
bulimia.
Each of these grossly misrepresents what the heritability estimate tells us.
Even more challenging is helping patients incorporate this information into
their schema regarding the causes of their illness. The two concepts that are
keyto understanding heritability are confidence intervals and variability in
liability. Even though the research may come up with a point estimate like 83%
heritable, it is critical to pay attention to the confidence interval that bounds the
estimate. So it is more accurate to say that between 49% and 100% of the
variance in liability to BN is due to genetic effects. The second part of the
equation is variance. We all vary in terms of our risk for developing BN. What
these results are saying is that genes play a role in determining the extent to
which an individual is likely to develop BN (or whatever the relevant trait
may be).
Another critical point is that there is not one true heritability estimate for any
given trait or disorder. It is inaccurate to say “The heritability of bulimia nervosa
is 83%.” Heritability is a statistic that varies across populations and across time.
Kendler and Pedersen (4) illustrated beautifully how heritability estimates can
change for a trait over time. They explored the pattern of twin resemblance for
regular tobacco use in a population-based sample of Swedish twins. Results for
males suggested both genetic and rearing-environmental effects, which, in the
best-fit biometrical model, accounted for 61% and 20% of the variance in
liability to regular tobacco use, respectively. For women, the pattern differed
by birth cohort. In women born before 1925, rates of regular tobacco use were
low and twin resemblance was influenced primarily by environmental factors. In
later cohorts, rates of regular tobacco use in women increased substantially and
heritability estimates were on par with those seen in men (63%). Thus, the
genetic influences were only detectable in females once social constraints on
female tobacco use were relaxed and smoking became more prevalent.
Family and Twin Studies of Eating Disorders
Family Studies of Eating Disorders

There are now a series of large, well-controlled family studies of eating
disorders. The vast majority of controlled family studies (5–10) have found a
significantly greater lifetime prevalence of eating disorders among relatives of
eating-disordered individuals in comparison to relatives of controls. The relative
risk of anorexia in first-degree family members of individuals with anorexia has
been reported to be 11.3 (10). That means that relatives of individuals with AN
are 11.3 times more likely to have an AN than relatives of controls. For BN, the
relative risk in female relatives of probands with BN was reported to be
4.4 (10). Moreover, several studies have found increased rates of both AN and
BN, and eating disorder not otherwise specified (ED-NOS) (i.e., coaggregation)
in relatives of individuals with AN as well as individuals with BN, compared to
rates among relatives of controls (5–10), suggesting that the various eating
disorders share transmissible risk factors. Moreover, relatives of individuals
with AN and BN have also been found to have a significantly increased rate of
subthreshold eating disorders compared to relatives of controls (9,10),
suggesting that the eating disorders do not “breed true” but are expressed in
families as a broad spectrum of eating-related pathology. Anorexia may have a
slightly stronger tendency to aggregate specifically in families; Woodside
et al. (11) reported a tendency for anorexia to cluster more in families of
probands with anorexia than in probands with bulimia, possibly suggesting some
specificity of clustering for anorexia.
Much less is known about the familiality of BED. Brody et al. (12) noted that
binge eating is more commonly reported in family members of individuals with
BED than in family members of individuals without BED. A small study by
Fowler and Bulik (13) found that the percentage of BED individuals who
reported at least one first-degree relative with BED (60%) was significantly
higher than for those without BED (5%), providing preliminary evidence for the
familiality of the disorder.
In summary, family study data reveal an elevation in the lifetime prevalence
of eating disorders among the relatives of people with eating disorders. In
addition, the coaggregation in families of persons with AN, BN, or milder
170 BULIK
eating disturbances suggests shared etiological factors across these conditions.

Large controlled family studies of BED have yet to be conducted, although
preliminary evidence suggests that BED is familial as well. Such studies are
important as the demonstration of familiality and the determination of the
causes of that familiality have the potential to play a critical role in the ongoing
debate of whether BED should emerge as a freestanding eating disorders
diagnosis or is best subsumed under another existing condition—the current
psychiatric nosology (14–16).
Twin Studies of Eating Disorders.

Given that there have been no adoption studies of eating disorders, we have had
to rely on twin studies to separate out the effects of genes and environ ment on
the observed familial aggregation of these traits.
The Equal Environment Assumption

One of the key assumptions underlying twin studies is the equal environment
assumption (EEA), which posits that MZ and DZ twins are equally correlated for
their exposure to environmental influences that are of etiological relevance to
the trait under study (1). In other words, this means that MZ twins are no more
likely to have received similar exposure to an environmental factor that may
play a causal role in eating disorders than DZ twins. It is well known (and
commonly observed by the casual observer) that the environments shared by
MZ twins are often more similar than the environments shared by DZ twins. This
could include common environments such as sharing a bedroom or dressing
alike. Although this suggests a more correlated environment in MZ than DZ
twins, the relevant point is that this dimension is not one that is assumed to be of
etiological relevance to eating disorders. There are no extant data to suggest
that being dressed like one’s twin or sharing a bedroom increases one’s risk of
developing an eating disorder. Violations of the EEA are critical only when the
violation occurs in domains that are relevant to the etiology of the trait. Such an
example might be if mothers of MZ twins were more likely to put both twins
on a diet than mothers of DZ twins.
If the EEA is violated, then the greater resemblance of MZ twins in
comparison to DZ twins could actually be due to environmental factors. A
violation of the EEA does not necessarily invalidate the results of a twin study
but may influence the magnitude of the estimated genetic and environmental
components. Studies of the EEA with regard to eating disorders suggest that this
assumption has not been violated in twin studies (17–20).
There is more work to be done in validating the EEA in eating disorders

research. Many of the existing evaluations of the EEA focus on very
global environments. No tests of the EEA have yet been performed that probe
into specific environmental factors that may be of etiological relevance to AN
and BN (e.g., codieting with your twin, joining activities that promote dieting).
Twin Studles of Anorexia Nervosa

Isolated case reports of MZ twins concordant for AN have appeared sporadically
in the literature (21–24). The first systematic study of clinically ascertained
twins with AN found that concordance for MZ twins was substantially greater
than for DZ twins (25–27). Reanalysis of these data (assuming a population
prevalence of AN of 0.75%) revealed evidence of familial aggregation with
parameter estimates of 88% for a2, 0 for c2, and 12% for e2. This reanalysis
suggested the role of additive genetic effects and some unique environmental
effects with no contribution from common environment.
Twin studies of AN have been difficult to conduct due to the low prevalence
of the disorder. Because of the relative rarity, each of these studies has had to
adopt some strategy to boost statistical power either by broadening the criteria
for anorexia nervosa (28), basing the study on a large sample but simply relying
on self-report diagnosis of anorexia (29), or exploring the heritability of
anorexia in the context of a bivariate twin study paired with a higher prevalence
condition such as major depression (30). Although scientifically defensible,
having to adopt these strategies underscores the difficulties of conducting
sufficiently powered twin studies with rare complex traits.
Reviewing these studies briefly, Wade et al. (30) derived heritability
estimates for AN in the context of studying the nature of the comorbid
relationship between AN and major depression. The heritability of AN was
estimated to be 58%, although the authors could not rule out a contribution of
shared environment to the liability to AN due to limited statistical power.
Kortegaard et al. (29) conducted a twin study on 34,142 Danish twins based on
self-reported diagnosis of AN. They reported heritability estimates of 0.48 and
0.52 for narrow and broad definitions of AN, with no influence of common
environment. Finally, Klump et al. (28) estimated the heritability of broadly
defined AN to be 0.74 in 17-year-old female twins with the remaining variance
accounted for by individual-specific environmental effects.
On balance, we can conclude from family studies (see 31 for a review) that
AN is familial and at least preliminarily from twin studies that the familiality
appears to be due primarily to additive genetic effects. Before that can be stated
with confidence, we will require larger population-based studies or pooling of
data across existing twin samples.
172 BULIK
Twin Studies of Bulimia Nervosa

Twin studies of BN have been more successful than those of AN given the
higher population prevalence of the disorder. Similar to AN, hints of genetic
effects on bulimia first begin to emerge with reports of concordant MZ twins in
the literature (32). Case series of twins with BN revealed consistently greater
concordance for BN in MZ than DZ twin pairs (27,33,34). Pooling data from
these case series for twin modeling and assuming a population prevalence of BN
of 2.5% revealed evidence of familial aggregation with 47% of the variance
accounted for by additive genetic effects, 30% by common environmental
effects, and 23% by unique environmental effects. However, the sample sizes
were small and the estimates imprecise.
Population-based studies of BN have been conducted in the United
States (35,36), Australia (37,38), and via self-report diagnoses in
Denmark (29). The studies that have estimated the heritability of BN based on a
single occasion of measurement suggest a moderate contribution of additive
genetic effects (point estimates ranging from 0.31 to 0.54 with confidence
intervals ranging from 0 to 0.86), a negligible contribution of shared
environmental effects, and a more substantial contribution of unique
environmental effects to liability to BN (29,35,36,38). Despite the higher
prevalence of bulimia these studies still had limited statistical power as reflected
in the broad confidence intervals (36).
One approach to enhancing statistical power is to incorporate more than one
wave of measurement into the twin model. Two studies have successfully
employed this approach (36,38). This model (called the measurement twin
model) controls for unreliability of diagnosis, increases power to detect both a2
and c2, and provides the most reliable information regarding the nature and
magnitude of genetic and environmental contributions to latent liability to BN.
The results of these two measurement model studies reveal a markedly greater
contribution of additive genetic effects to the liability to BN (59% and 83%,
respectively), a negligible contribution of shared environment (0 in both
studies), and a moderate contribution of unique environmental effects (41% and
17%). It is critical to note that these two studies had greater power to detect
both c2 and a2. In these studies, the point estimates of a2 were higher and the
confidence intervals suggested a substantial contribution of a2. The point
estimates of c2 were quite low, but the confidence intervals allowed for the
possibility of some contribution of shared environment. In both studies the
confidence intervals for the a2 estimates and the c2 estimates were
nonoverlapping. Thus, by controlling for measurement error and increasing
statistical power, it appears that the contribution of additive genetic effects to
liability to BN is more substantive than the contribution of common

environment.
In summary, from twin and family studies, we can conclude that BN is
familial and that there appears to be a moderate to substantial contribution
made by genetic factors and unique environmental factors to liability to the
disorder. The contribution of shared environment is less certain but appears to
be less prominent than the effect of genes and of unique environment.
A reasonable next step for twin studies is to determine the precise nature of the
unique environmental effects that increase risk for developing BN.
Twin Studies of Binge Eating Disorder

As many of the twin registries were established prior to the emergence of BED
as a disorder requiring further investigation in the literature. Much of the
existing interview and self-report screenings did not include questions that
enable a diagnosis of BED. Two more recent waves of data collection are
currently underway which will allow us to address the heritability of
BED (39,40). In the meantime, some information regarding the heritability of
the symptom of binge eating is available.
Sullivan et al. (20) explored genetic and environmental contributions to the
symptoms of objective binge eating and self-induced vomiting by incorporating
both behaviors into a bivariate twin model. The prevalences of each of these
individual symptoms were higher than prevalences for the full syndrome of BN.
Therefore, the power to detect a2 and c2 was greater as reflected in the narrower
confidence intervals. Two findings from this study are relevant to BED. The
association between having ever binged (23.6%) and having ever induced
vomiting (4.8%) was very strong [odds ratio (OR) = 8.78, p 0.00001]. The
best-fitting model indicated that lifetime bingeing and vomiting were both
heritable (46% and 72%) and influenced by individual-specific environmental
factors (54% and 28%). The overlap between the genetic (ra = 0.74) and
individual-specific environmental factors (re = 0.48) for the two traits was
substantial but less than unity. So, what this study tells us that is relevant to BED
is first that binge eating as a symptoms is itself heritable, and although the
genetic and environmental correlations between binge eating and vomiting are
substantial, they are not complete. That is, although there appear to be genetic
and environmental factors that influence both binge eating and vomiting, there
are also genetic and environmental factors that uniquely influence these
symptoms (i.e., they are not perfectly correlated).
174 BULIK
USING TWIN STUDIES TO EXPLORE THE

RELATION BETWEEN ANOREXIA AND BULIMIA
NERVOSA
AN and BN are probably neither completely independent nor completely
overlapping conditions. Despite the frequency with which these disorders
occur both concurrently and sequentially, we know little about the
pattern and predictors of the observed comorbidity. A number studies of the
outcome of anorexia have been conducted that specifically address the
percentage of patients who have developed BN at the time of follow-up
assessment (41–48). The duration of follow-up intervals ranged from 4 to
22 years, and the percentage of individuals at follow-up who met diagnostic
criteria for DSMIII bulimia or DSM-IIIR BN ranged from 8% to 41%. Across
studies, the percentage of individuals with BN tended to be greater the longer
the followup interval. In a pilot study with a very long observation period of
22 years, Hsu et al. (44) reported that 19% of women with AN met criteria for
BN at follow-up. In addition to frank diagnoses of BN, 14–36% of women in
these studies met criteria for ED-NOS (42,43,45,46) which includes women
with bulimic symptoms of insufficient frequency or duration to qualify for a
diagnosis of BN. Further evidence for the relation between AN and BN is
reflected in the fact that a significant minority (22–37%) of women with BN in
clinical samples report a history of AN [(49–54), for example]. Finally,
crosssectional investigations identify a range of symptom combinations ranging
from the restricting subtype of anorexia, the mixed clinical picture of anorexia
and bulimia, current bulimia with a history of anorexia, and bulimia with no
history of anorexia (49,55).
The above clinical studies are all subject to referral bias, which could
conceivably inflate the observed frequency with which anorexia and bulimia
co-occur either concurrently or sequentially. Arguing against this, however,
epidemiological studies of nonclinical populations indicate an elevated odds ratio
between anorexia and bulimia (OR = 8.2) (35,56). To date, no twin study has
had sufficient power to explore the extent to which genetic and environmental
factors that contribute to liability to anorexia and bulimia are shared.
Bivariate twin analysis is able to decompose the covariance between two
disorders. A bivariate twin model (Fig. 1) decomposes the observed correlation
betwen two disorders into the part due to additive genes, familial environment,
and unique environmental influences. These are potentially powerful tools
because the degree to which two disorders share genetic or environmental risk
factors can have important implications for nosology. This key analysis has not
been accomplished in existing population-based studies of twins due to lack of
statistical power. By addressing this question in a larger sample, we would be able
FIGURE 1 Bivariate twin model exploring the nature of the genetic and environmental
relation between anorexia (AN) and bulimia nervosa (BN).
to address a critical question in the field, namely, what is the nature of the
relation between anorexia and bulimia nervosa?
WHY DO WE NOT DETECT COMMON

ENVIRONMENTAL EFFECTS IN TWIN STUDIES OF
EATING DISORDERS?
One of the most frequently asked questions is to explain the relative
infrequency with which common environment has been found to contribute to
etiology of AN and BN (57). To understand this it is important to approach the
question both methodologically (why is it hard to detect shared environment in
twin studies?) and definitionally (what is shared environment and how does it
differ from unique environment or from what is commonly termed family
environment?).
If we approach the methodological issues first, a quick glance at the point
estimates suggests that there is minimal contribution of shared environment in
most of the twin studies of eating disorders. As noted above, without the
confidence intervals the point estimates can be misleading. In all of the studies
reviewed above, the confidence intervals do not eliminate the possibility of a
contribution of c2 to the etiology of eating disorders. However, they do suggest
that the contribution of shared environment is substantially less than the
contribution of genes.
Second, is possible that “common” or “shared” environment truly does not
influence risk of eating disorders and that our results simply reflect reality.
176 BULIK
Saying that common environment does not influence a trait is not equivalent to
saying that “family environment” has no influence on risk for the trait. What we
casually refer to as “family environment” is composed of factors that qualify both
as common environment and as unique environmental features (58), and
research strongly suggests that children raised in the same families experience
surprisingly different environments (59).
For example, let’s assume for a moment that maternal dieting is an
environmental factor (although we will illustrate later that this may be
a fallacious assumption). Since both members of a twin pair are exposed to
maternal dieting, we could conceptualize this as a shared environmental factor.
Although the shared component is that both twins were exposed to this
behavior, the exposure may be uniquely experienced by each twin. Whereas
one twin might climb on board and adopt the behavior herself to try to lose
weight and boost her own self-esteem, the other twin might recognize the
futility and vow never to become so ruled by her weight and appearance. Thus,
what appeared to be a common environmental experience actually exerted
more influence through its unique environmental effects.
In addition, the relative absence of shared environmental effects does not
mean that the environment does not impact on liability to eating disorders at
all. Indeed, all studies have found that a substantial portion of the variance
resulted from nonshared environmental factors (57). These unique
environmental factors simply appear to be more etiologically relevant than the
shared ones.
Also, several observational and self-report studies have found differences
between the family environment of women with eating disorders and
controls (60–65). The problem with these family environment designs is that
they cannot take into account the presence of gene-environment correlations.
Many of the variables that are assumed to be “environmental” may be influenced
or elicited by features of either the child or the parent that are genetically
mediated.
Genotype-environment correlation arises when the exposure to positive or
negative environmental influences is not randomly distributed with respect to
genetic differences. For example, girls who are genetically more prone to drive
for thinness may evoke more intrusive comments about eating and dieting from
their parents (evocative gene-environment correlation) and actively seek peers
or activities that reinforce their drive to lose weight, such as, ballet or
gymnastics (active gene-environment correlation). Another type of gene-
environment correlation (“passive” correlation) reflects the fact that children
receive genotypes that are correlated with their family environment. This type
of gene-environment correlation reflects that one receives one’s genes from the
same individuals who create one’s environment. Moreover, the environment
that they create for you is in part determined by their genotype. Concretely,
one can imagine a mother with a history of an eating disorder who still has
lingering concerns about her own weight who has passed on this genetic
predisposition to her daughter. In addition, she contributes to an environment
in the household that is highly weight focused by only buying low-fat foods,
watching her daughter’s caloric intake, and often commenting about the size
and shape of her own body. Thus, the daughter is not only dealing with the
impact of her inherited genotype; she is also being raised in an environment that
may facilitate the expression of that genetic predisposition.
Last, shared environmental factors may be important to the etiology of eating
disorders but may only exert their influence in concert with an individual’s
genetic makeup. Thus, the independent main effect of shared environmental
factors may be small, but their effects via one of several types of gene
x-environment interactions might be profound (66).
IMPLICATIONS OF FAMILY AND TWIN FOR

PREVENTION AND CLINICAL PRACTICE
The upsurge in family, twin, and genetic research has generated enormous
interest in patient advocacy and family groups. In many ways these findings have
assisted in legitimizing the disorders and have helped to erase an existing
prejudice that somehow individuals with eating disorders “choose” to have AN or
BN. Results from twin studies underscore that these disorders are no less
influenced by genes than major psychiatric disorders such as bipolar disorder
and schizophrenia, which are comfortably viewed as biologically mediated in
most spheres.
One potential pitfall to be avoided in clinical practice is interpreting these
results as genetic determinism. When describing the genetic research to patients
and families it is imperative to underscore that the presence of a genetic
predisposition in no way guarantees expression of the trait. One way of
empowering families over their genotypes is to reframe the concept of
geneenvironment correlation in a positive light. Some parents develop guilty
and powerless feelings about having passed on genes that might increase risk for
eating disorders in their children. By emphasizing the fact that they have the
ability to influence environment and that environment can serve a protective
function as well as an evocative function, parents can begin to see that although
there is nothing they can do to alter the passing down of DNA, they can alter
environments that influence the likelihood of genes being expressed.
Another important ramification of family and twin studies concerns
prevention. Our global efforts at preventing eating disorders have been
unimpressive and in some cases have caused more problems than they
178 BULIK
prevented. Combined, the results of family and twin studies confirm that
offspring of women with anorexia and bulimia constitute a high-risk group for
eating disorders. Targeted prevention programs could be tested that intervene at
several different points along the continuum. The earliest intervention point
could focus on enhanced prenatal and postnatal care for women with current or
even historical eating disorders. These efforts would focus on healthy and
adequate nutrition during pregnancy, dealing with issues that arise as body size
increases with pregnancy, developing healthy feeding and parenting styles after
birth, and having realistic postdelivery expectations about the resumption of
normal body weight (67).
A second preventive option would be to develop programs for mothers with
eating disorders or food and weight issues to help them develop tools to model
healthy eating behaviors and healthy body esteem and self-esteem to their
daughters (and sons). As noted above, although at this stage little can be done to
alter the transmission of susceptibility genes to offspring of women and men
with eating disorders, we can assist in modifying the environments that these
individuals create that may contribute to the expression of those genetic
predispositions.
A third preventive option could focus on the offspring themselves. In much
the same manner that children of alcoholics are encouraged to delay the onset of
alcohol consumption, offspring of individuals with eating disorders could be
provided with assistance to avoid the pitfalls of dieting, for developing non-
body-centered self-esteem, and other ways to prevent the development of
eating disorders.
CONCLUSIONS
We have been aware for decades that family and society influence risk for eating
disorders. We have been less clear on how they exert their influence. Family
and twin studies have helped us understand the nontrivial role of genes in
influencing risk. Somewhat paradoxically, given recent advances in technology,
the study of genes has become more tractable and practical than the study of
environments. All of the twin studies to date have shown that nonshared
environment significantly influences liability to eating disorders (57). We should
not abandon research on environment simply because our tools to explore the
human genome have become more practical. The environment is amazingly
complex and no doubt exerts its influence in myriad often idiosyncratic ways.
Nonetheless, a complete understanding of pathways of risk to eating disorders will
necessarily require further research designed to identify both the specific genes
that influence risk and the specific environmental factors that evoke and inhibit
expression of those genes.
REFERENCES
1. Plomin R, DeFries JC, McClearn GE, Rutter M. Behavioral Genetics. 3d ed.
New York: W.H. Freeman & Co, 1994.
2. Klump KL, McGue M, Iacono WG. Genetic relationships between personality and
eating attitudes and behaviors. J Abnorm Psychol 2002; 111:380–389.
3. Allison D, Faith M. Genetic and environmental influences on human body weight:
implications for the behavior therapist. Nutrition Today 2000; 35:18–21.
4. Kendler KS, Thornton LM, Pedersen NL. Tobacco consumption in Swedish twins
reared apart and reared together. Arch Gen Psychiatry 2000; 57(9):886–892.
5. Gershon E, Schreiber J, Hamovit J, Dibble E, Kaye W, Nurnberger J, Anersen A,
Ebert M. Anorexia nervosa and major affective disorders associated in families: a
preliminary report. In: Guze SB, Earls FJ, Barrett JE, eds. Childhood
Psychopathology and Development. New York: Raven Press, 1983:279–284.
6. Hudson JI, Pope HG, Jonas JM, Yurgelun-Todd D, Frankenburg FR. A controlled
family history study of bulimia. Psychol Med 1987; 17:883–890.
7. Kassett J, Gershon E, Maxwell M, Guroff J, Kazuba D, Smith A, Brandt H,
Jimerson D. Psychiatric disorders in the first-degree relatives of probands with
8. Strober M, Lampert C, Morrell W, Burroughs J, Jacobs C. A controlled family
study of anorexia nervosa: evidence of familial aggregation and lack of shared
transmission with affective disorders. Int J Eating Disord 1990; 9:239–253.
9. Lilenfeld L, Kaye W, Greeno C, Merikangas K, Plotnikov K, Pollice C, Rao R,
Strober M, Bulik C, Nagy L. A controlled family study of restricting anorexia and
bulimia nervosa: comorbidity in probands and disorders in first-degree relatives.
10. Strober M, Freeman R, Lampert C, Diamond J, Kaye W. Controlled family study
of anorexia nervosa and bulimia nervosa: evidence of shared liability and
transmission of partial syndromes. Am J Psychiatry 2000; 157:393–401.
11. Woodside D, Field LL, Garfinkel P, Heinmaa M. Specificity of eating disorders
diagnoses in families of probands with anorexia nervosa and bulimia nervosa. Comp
Psychiatry 1998; 39:261–264.
12. Brody ML, Walsh BT, Devlin MJ. Binge eating disorder: reliability and validity of a
new diagnostic category. J Consult Clin Psychol 1994; 62(2):381–386.
13. Fowler S, Bulik C. Family environment and psychiatric history in women with
binge eating disorder and obese controls. Behav Change 1997; 14:106–112.
14. Kendell RE. Clinical validity. Br J Psychiatry 1989; 19:45–55.
15. Kendler KS. Toward a scientific psychiatric nosology. Arch Gen Psychiatry 1990;
47:969–973.
16. Robins E, Guze SB. Establishment of diagnostic validity in psychiatric illness: its
application to schizophrenia. Am J Psychiatry 1970; 126:107–111.
17. Klump KL, Holly A, Iacono WG, McGue M, Willson LE. Physical similarity and
twin resemblance for eating attitudes and behaviors: a test of the equal
environments assumption. Behav Genet 2000; 30(l):51–58.
180 BULIK
18. Bulik C, Sullivan P, Wade T, Kendler K. Twin studies of eating disorders: a review.
Int J Eat Disord 2000; 27:1–20.
19. Kendler KS, Neale MC, Kessler RC, Heath AC, Eaves LJ. A test of the equal
environment assumption in twin studies of psychiatric illness. Behav Genet 1993;
23:21–27.
20. Sullivan PF, Bulik CM, Kendler KS. The genetic epidemiology of binging and
vomiting. Br J Psychiatry 1998; 173:75–79.
21. Askevold F, Heiberg A. Anorexia nervosa: two cases in discordant MZ twins.
22. Suematsu H, Kuboki T, Ogata E. Anorexia nervosa in monozygotic twins.
23. Vandereycken W, Pierloot R. Anorexia nervosa in twins. Psychother Psychosom
1981; 35:55–63.
24. Nowlin N. Anorexia nervosa in twins: case report and review. J Clin Psychiatry
1983; 44:101–105.
25. Holland AJ, Hall A, Murray R, Russell GFM, Crisp AH. Anorexia nervosa: a study
of 34 twin pairs and one set of triplets. Br J Psychiatry 1984; 145:414–419.
26. Holland AJ, Sicotte N, Treasure J. Anorexia nervosa: evidence for a genetic basis.
J Psychosom Res 1988; 32(6):561–571.
27. Treasure J, Holland A. Genetic vulnerability to eating disorders: evidence from
twin and family studies. In: Remschmidt H, Schmidt M, eds. Child and Youth
Psychiatry: European Perspectives. New York: Hogrefe & Huber, 1989:59–68.
28. Klump KL, Miller KB, Keel PK, McGue M, Iacono WG. Genetic and
environmental influences on anorexia nervosa syndromes in a population-based
twin sample. Psychol Med 2001; 81(4):737–740.
29. Kortegaard LS, Hoerder K, Joergensen J, Gillberg C, Kyvik KO. A preliminary
population-based twin study of self-reported eating disorder. Psychol Med 2001;
31:361–365.
30. Wade TD, Bulik CM, Neale M, Kendler KS. Anorexia nervosa and major
depression: shared genetic and environmental risk factors. Am J Psychiatry 2000;
157:469–471.
31. Lilenfeld L, Kaye W, Strober M. Genetics and family studies of anorexia nervosa
and bulimia nervosa. In: Balliere’s Clinical Psychiatry, 1997:177–197.
32. Kaminer Y, Feingold M, Lyons K. Bulimia in a pair of monozygotic twins. J Nerv
Ment Dis 1988; 176:246–248.
33. Fichter MM, Noegel R. Concordance for bulimia nervosa in twins. Int J Eat Disord
1990; 9:255–263.
34. Hsu GLK, Chesler BE, Santhouse R. Bulimia Nervosa in eleven sets of twins: a
clinical report. Int J Eat Disord 1990; 9:275–282.
35. Kendler KS, MacLean C, Neale MC, Kessler RC, Heath AC, Eaves LJ. The genetic
epidemiology of bulimia nervosa. Am J Psychiatry 1991; 148:1627–1637.
36. Bulik CM, Sullivan PF, Kendler KS. Heritability of binge-eating and broadly
defined bulimia nervosa. Biol Psychiatry 1998; 44(12):1210–1218.
37. Wade T, Neale MC, Lake RIE, Martin NG. A genetic analysis of the eating and
attitudes associated with bulimia nervosa: dealing with the problem of
ascertainment. Behav Genet 1999; 29:1–10.
38. Wade TD, Martin N, Neale M, Tiggemann M, Trealor S, Heath A, Bucholz K,
Madden P. The structure of genetic and environmental risk factors for three
measures of disordered eating characteristic of bulimia nervosa. Psychol Med 1999;
29:925–934.
39. Anderson C, Bulik CM. A twin study of binge-eating disorder (in preparation).
40. Reichborn-Kjennerud T, Tambs K, Harris JR, Bulik CM. Gender differences in
binge-eating in the absence of compensatory behaviors.
41. Eckert ED, Halmi KA, Marchi P, Grove W, Crosby R. Ten-year follow-up
of anorexia nervosa: clinical course and outcome. Psychol Med 1995; 25:143–156.
42. Gillberg IC, Rastam M, Gillberg C. Anorexia nervosa outcome: six-year controlled
longitudinal study of 51 cases including a population cohort. J Am Acad Child Adol
Psychiatry 1994; 33:729–739.
43. Herpertz-Dahlman B, Wewetzer C, Schulz E, Remschmidt H. Course and
outcome in adolescent anorexia nervosa. Int J Eat Disord 1996; 19:335–345.
44. Hsu G, Crisp A, Callender J. Psychiatric diagnoses in recovered and unrecovered
anorectics 22 years after onset of illness: a pilot study. Comp Psychiatry 1992; 33:
123–127.
45. Schork EJ, Eckert ED, Halmi KA. The relationship between psychopathology,
eating disorder diagnosis, and clinical outcome at 10-year follow-up in anorexia
nervosa. Comp Psychiatry 1994; 35:113–123.
46. Smith C, Feldman S, Nasserbakht A, Steiner H. Psychological characteristics and
DSM-III-R diagnoses at 6-year follow-up of adolescent anorexia nervosa. J Am
Acad Child Adol Psychiatry 1993; 32:1237–1245.
47. van der Ham T, van Strien D, van Engeland H. A four-year prospective followup
study of 49 eating-disordered adolescents: differences in course of illness. Acta
Psych Scand 1994; 90:229–235.
48. Zipfel S, Lowe B, Reas DL, Deter HC, Herzog W. Long-term prognosis in
anorexia nervosa: lessons from a 21-year follow-up study. Lancet 2000; 355(9205):
721–722.
49. Braun DL, Sunday SR, Halmi KA. Psychiatric comorbidity in patients with eating
disorders. Psychol Med 1994; 24:859–867.
50. Fairburn CG, Cooper PJ. The clinical features of bulimia nervosa. Br J Psychiatry
1984; 144:238–246.
51. Pyle RL, Mitchell JE, Eckert ED. Bulimia: a report of 34 cases. J Clin Psychiatry
1981; 42:60–64.
52. Russell GFM. Bulimia nervosa: an ominous variant of anorexia nervosa. Psychol
Med 1979; 9:429–448.
53. Schmidt U, Keilen M, Tiller J, Treasure J. Clinical symptomatology and etiological
factors in obese and normal weight bulimic patients: a retrospective case-control
study. J Nerv Ment Dis 1993; 181:200–202.
182 BULIK
54. Sullivan PF, Bulik CM, Carter FA, Gendall KA, Joyce PR. The significance of a
prior history of anorexia in bulimia nervosa. Int J Eat Disord 1996; 20:253–261.
55. Herzog DB, Keller MM, Sacks NR, Yeh CJ, Lavori PW. Psychiatric comorbidity in
treatment-seeking anorexics and bulimics. J Am Acad Child Adol Psychiatry 1992;
31:810–818.
57. Klump KL, Wonderlich S, Lehoux P, Lilenfeld LR, Bulik CM. Does environment
matter? A review of nonshared environment and eating disorders. Int J Eat Disord
2002; 31(2):118–135.
58. Silberg JL, Erickson MT, Meyer JM, Eaves LJ, Rutter ML, Hewitt JK.
The application of structural equation modeling to maternal ratings of twins’
behavioral and emotional problems [published erratum appears in J Consult Clin
Psychol 1994 Dec; 62(6): 1234]. J Consult Clin Psychol 1994; 62(3):510–521.
59. Dunn J, Plomin R. Separate Lives: Why Siblings Are So Different. New York:
Basic Books, 1990.
60. Fallon BA, Sadik MSW, Saoud JB, Garfinkel RS. Childhood abuse, family
environment, and outcome in bulimia nervosa. J Clin Psychiatry 1994; 55(10):
424–428.
61. Humphrey LL, Apple RF, Kirschenbaum DN. Differentiating bulimic-anorexic
from normal families using interpersonal and behavioral observational systems.
J Consult Clin Psychol 1986; 54(2): 190–195.
62. Humphrey L. Family relations in bulimic-anorexia and nondistressed families.
Int J Eat Disord 1986; 5:223–232.
63. Wonderlich S. Relationship of family and personality factors in bulimia. In:
Crowther J, Tennenbaum D, Hobfall S, Parvis S, eds. The Etiology of Bulimia
Nervosa: The Individual and Familial Context. Washington, D.C.: Hemisphere
Publishers, 1992:103–126.
64. Woodside D, Shekter-Wolfson L, Garfinkel P, Olmsted M, Kaplan A, Maddocks
S. Family interactions in bulimia nervosa: 1. Study design, comparisons to
established population norms, and changes over the course of an intensive day
hospital treatment program. Int J Eat Disord 1995; 17:105–115.
65. Woodside D, Shekter-Wolfson L, Garfinkel P, Olmsted M. Family interactions in
bulimia nervosa II: complex intrafamily comparisons and clinical significance. Int J
Eat Disord 1995; 17:117–126.
66. Kendler KS, Eaves LJ. Models for the joint effect of genotype and environment on
liability to psychiatric illness. Am J Psychiatry 1986; 143:279–289.
67. Mitchell-Gieleghem A, Mittelstaedt ME, Bulik CM. Eating disorders and
childbearing: concealment and consequences. Birth 2002; 29(3): 182–191.
8
Psychiatric Comorbidity Associated with
Anorexia Nervosa, Bulimia Nervosa, and
Lisa Rachelle Riso Lilenfeld
Georgia State University
Atlanta, Georgia, U.S.A.
OVERVIEW
The term “comorbidity” was introduced in the medical literature by Feinstein
(1) to refer to patients with two co-occurring diseases. In recent years, this
concept has become well known in the field of psychiatry. Extensive
comorbidity has been documented among a number of psychiatric disorders,
not the least of which are eating disorders. Such comorbidity has several
important clinical and research implications, as reviewed by Klein and Riso (2).
First, the presence of comorbid disorders may affect the course and treatment
response of the primary disorder. Second, comorbidity can make it unclear as to
whether associations between the primary disorder and other variables are true
associations or instead correlations with a comorbid condition. Third, a great
degree of comorbidity may lead to reevaluation of the validity of a diagnostic
category. That is, it may suggest an alternative diagnostic grouping. Fourth, high
rates of comorbidity may provide important information regarding the etiology
of the primary disorder, as well as the comorbid disorders.
Klein and Riso (2) provide an in-depth review of the many reasons for which
comorbidity may exist. A brief overview of those most relevant to the field of
eating disorders will be provided here. The first, possibly least interesting, but
important reason for comorbidity that is often overlooked is simply chance.
Two disorders with high prevalences will co-occur within individuals simply on
the basis of chance due to base rates. While anorexia nervosa is a relatively
uncommon disorder with a lifetime prevalence rate of approximately 0.5–1.0%
among young females, bulimia nervosa and binge eating disorder are more
common with lifetime prevalence estimates of approximately 1–3% and 3%,
respectively, among this population (3). A second reason comorbidity may be
observed is sampling bias, often referred to as “Berkson’s bias” (4). This is
particularly relevant in the field of psychiatry, where studies using clinical
populations can produce high estimates of comorbidity. This occurs because
184 LILENFELD
individuals with multiple disorders are more likely to seek treatment than
individuals with one disorder. This bias has often led to higher comorbidity
estimates in studies where eating disorder subjects are recruited from treatment
centers as opposed to studies in which subjects are recruited from the
community. A third reason for comorbidity is population stratification. That is,
each of two disorders may be associated with distinct risk factors. If both sets of
risk factors happen to have an increased prevalence in the same subgroup of the
population (e.g., as has been proposed for bulimia nervosa and substance
abuse), the rate of comorbidity is higher than that expected by chance for the
population as a whole. A fourth reason is overlapping diagnostic criteria. For
example, one of the criteria for borderline personality disorder is potentially
dangerous impulsive behavior, among which binge eating is listed. Thus,
someone with this symptom would simultaneously partially meet one criterion
for borderline personality disorder, as well as for bulimia nervosa or binge
eating disorder, obviously increasing the likelihood of comorbidity of the
personality disorder and eating disorder. A fifth reason for comorbidity is
known as “multiformity,” referring to the fact that disorders may assume
multiple heterogeneous forms. For instance, the comorbid disorder might be an
alternative manifestation of the target disorder. For instance, some have
suggested that anorexia nervosa and obsessive-compulsive disorder (OCD) may
be alternate forms of the same underlying pathology (5).
The last two reasons for comorbidity, which the majority of eating disorder
comorbidity studies have focused on, are among the more interesting. First, one
disorder may be a risk factor for another. For instance, it has been suggested that
anxiety disorders may be a risk factor for eating disorders (6). The final reason
for comorbidity to be reviewed here is overlapping etiological processes. That
is, there may be overlapping risk factors that contribute to “pure” forms of each
disorder. Thus, the two disorders can appear in their pure forms as well as
comorbidly. For instance, low self-esteem and perfectionism may be risk factors
for both depression and eating disorders, which would contribute to their
co-occurrence. In addition, depression and eating disorders each have their own
set of unique risk factors, leading to these disorders often appearing without the
other.
As mentioned at the outset of this chapter, the different reasons for
comorbidity among eating disorders and other psychiatric disorders are
important from both a basic science and clinical perspective. Patterns of
comorbidity may help us further refine our diagnostic categories, which have
continued to evolve; however, there is general agreement that such categories
are in need of further refinement (7). Comorbidity patterns may help to identify
taxons, i.e., naturally occurring categories that exist in the world (8), such that
PSYCHIATRIC COMORBIDITY 185
our diagnostic system would reflect true relationships among different forms of
psychopathology.
Clinically, as much as comorbidity patterns may lead to a better
understanding of underlying etiological processes, this may help in our
conceptualization model of eating disorders, which may in turn lead to more
effective treatment aimed at these etiological factors. Alternatively, or
additionally, understanding comorbidity patterns may lead to better
identification of risk factors, which may ultimately aid in prevention. At the
very least, clinicians who treat eating-disordered patients must be well versed in
the most commonly observed comorbid psychiatric disorders in this population,
as these disorders should often be a focus of treatment, either separately (as
suggested by some in the case of comorbid substance abuse) or in conjunction with
the eating disorder treatment (as suggested in the case of comorbid depression).
Thus, clinicians must be sure to conduct a thorough psychiatric assessment
focused on eating disorder symptoms, as well as comorbid conditions,
particularly major depressive disorder, all anxiety disorders, substance use
disorders, and cluster B and C personality disorders. Onset patterns, i.e.,
whether the eating disorder preceded the comorbid disorder(s), may also be
relevant to treatment.
The remainder of the chapter is devoted to reviewing comorbidity among
anorexia nervosa, bulimia nervosa, and binge eating disorder, beginning with an
emphasis on general differences between subtypes in a diagnostic category.
EATING DISORDER SUBTYPES
Anorexia Nervosa
The two DSM-IV diagnostic subtypes of anorexia nervosa, restricting type and
bingeing-purging type, display somewhat different patterns of comor bidity.
Both subtypes share the same core symptoms (refusal to maintain a minimally
normal body weight, intense fear of weight gain, body image disturbance, and
amenorrhea in postmenarcheal females). The key difference between the two
subtypes is that weight loss is accomplished primarily through dieting, fasting,
or excessive exercise in restricting-type anorexia nervosa, without any regular
binge eating or purging. In the bingeing-purging type, the individual regularly
engages in bingeing and/or purging, so that weight loss may be accomplished
through misuse of laxative, diuretics, or enemas, in addition to the methods
used in the restricting type.
While some comorbid disorders are common among both types of anorexia
nervosa, such as depression and anxiety disorders, other comorbid disorders are
186 LILENFELD
found to be associated primarily with the bingeing-purging form of anorexia

nervosa, such as substance use disorders. In fact, in many ways, patterns of
comorbidity for bingeing-purging anorexia nervosa more closely resemble
comorbid patterns for bulimia nervosa than they do for restricting-type anorexia
nervosa. Specific comorbidity findings are reviewed in the following section.
Bulimia Nervosa
The two DSM-IV diagnostic subtypes of bulimia nervosa are purging type and
nonpurging type. Both share the same core symptoms of binge eating,
inappropriate compensatory methods to prevent weight gain, and selfevaluation
being unduly influenced by body shape and weight. What distinguishes the
subtypes is the compensatory methods used. Purging-type bulimia nervosa
involves regular use of self-induced vomiting or misuse of laxatives, diuretics,
or enemas. Those with nonpurging-type bulimia nervosa use fasting or exercise,
but not vomiting, laxatives, diuretics, or enemas, as compensatory methods to
prevent weight gain. Bulimic patients may also use diet pills (e.g., stimulants) to
prevent weight gain, although the DSM-IV does not specify this type of
compensatory strategy in either subtype definition. If none of the other purging
methods are used, then a patient who abuses diet pills as their compensatory
strategy of choice is usually classified as having nonpurgingtype bulimia nervosa.
Nearly all comorbidity studies of bulimia nervosa have been conducted with
individuals who have the purging subtype. Patterns of comorbidity among this
subtype of bulimia nervosa often resemble patterns seen with the bingeing-
purging subtype of anorexia nervosa. Both disorders share the symptoms of
binge eating and purging (though it may be one or the other rather than both in
bingeing-purging anorexia nervosa). Thus, it may be this (these) particular
symptom(s) that are critical in accounting for the observed comorbidity.

As binge eating disorder was only recently defined and included as an example of
Eating Disorder-Not Otherwise Specified (ED-NOS) in the DSM-IV, many
fewer studies of comorbidity have been conducted than such studies of anorexia
nervosa and the purging type of bulimia nervosa. However, comorbidity
patterns have recently begun to emerge and may help us to better understand
how this disorder may share important features with, as well as be distinguished
from, the more well-known eating disorders. Understanding comorbidity of
binge eating disorder may not only help our treatment of the condition but may
help further refine this diagnostic category, which has been a subject of great
debate since its inclusion in the appendix of the DSM-IV as “warranting further
study.” Thus, refinement of this disorder, specifically regarding the decision as

to whether it warrants its own diagnostic category, may be greatly aided by
comorbidity data.
PSYCHIATRIC COMORBIDITY
It is well recognized that both anorexia nervosa and bulimia nervosa are often
accompanied by other psychiatric symptoms and syndromes, in particular
depression, anxiety disorders and substance use disorders. Of great importance
in recognizing and studying comorbidity and eating disorders is that these
comorbidities are likely substantially exaggerated by malnutrition and
pathological eating behaviors (9). Thus, one must always question whether the
apparent comorbid presentation of mood disturbance or high anxiety is a
function of the physiological consequences of the eating disorder itself. Even if
the comorbid disorder predates the eating disorder, it is possible that the mood
or anxiety pathology is exacerbated by, if not caused by, the consequences of
starvation in the case of anorexia nervosa or erratic consummatory behaviors in
the case of bulimia nervosa. In some patients, it is clear that the comorbidity
antedates weight loss or disordered eating, or persists after weight recovery or
abstinence from bingeing and purging (10–12), suggesting that the comorbid
problems may not simply be sequalae of malnutrition or pathological eating
behavior. Whether such comorbid pathology enhances vulnerability to the
development of eating disorders remains uncertain.
In the absence of high-risk paradigms, unraveling the precise nature of the
mechanisms underlying eating disorder comorbidities is a formidable task.
Family studies, in which patterns of comorbidity in relatives are assessed, are
probably the single most useful means of testing alternative models of
comorbidity (2). Several recent family studies (13–15) have shed light on
comorbid disorders that may share a common familial vulnerability with
eating disorders and will be briefly reviewed. The three areas of
comorbid psychopathology of greatest relevance to eating disorders are mood
disorders, anxiety disorders, and substance use disorders, all to be reviewed
below with regard to what we know about comorbidities of each with anorexia
nervosa, bulimia nervosa, and binge eating disorder. In addition, impulse
control disorders, schizophrenia and other psychotic disorders, dissociative
disorders, somatoform disorders, as well as attention deficit-hyperactivity
disorder (ADHD) and disruptive behavior disorders will also be briefly covered.
188 LILENFELD
Mood Disorders
It is well known that individuals with eating disorders often have symptoms of
depression. While the reasons for this frequent comorbidity has been a subject of
debate, the frequent presence of comorbid depression among individuals with
eating disorders has not.
As previously reviewed, there are many potential reasons for comorbidity.
First, one must ask whether the rate of depression among people with eating
disorders exceeds that expected by a chance association of these potentially
independent disorders. The nature of the population under study will have a
great effect on the answer to this question. That is, rates of depression among
those with eating disorders recruited from a clinic or other treatment sample
will nearly always be significantly higher than rates of depression among a
community sample of individuals with eating disorders. Most comorbidity
studies have been conducted with samples of convenience (i.e., treatment
samples) and are thus subject to referral bias, which results in an increased rate
of comorbid depression observed. This is likely to occur because people with
more than one disorder are more likely to seek treatment (i.e., Berkson’s bias).
In general, a clinical sample is likely to be more disturbed than an
epidemiological sample. Nonetheless, depression comorbidity rates among a
clinical sample may be of greatest interest to clinicians, as these are the patients
they are most likely to encounter! Thus, comorbidity studies using clinical
samples may be of greatest relevance to clinicians treating patients with eating
disorders, though it is important to keep in mind that this is just a subset of the
larger population of individuals with eating disorders. So, the reasons for such
comorbidity remain uncertain and are likely related to sampling.
It is agreed that the presence of depressive comorbidity is very common, but
just how common? With the aforementioned in mind, rates of comorbid
depression among individuals with eating disorders have been found to vary
markedly, again likely due to referral biases. Rates among clinical samples have
nearly always been significantly higher than rates observed in nonpsychiatric
control subjects. In addition, elevated rates of depressive dis orders, particularly
major depressive disorder, have been observed among all types of eating
disorders. Lifetime rates of major depression appear to be relatively equally
distributed across eating disorder subtypes, though there is some suggestion that
bingeing-purging anorexic and purging bulimic patients have higher rates of
depression than restricting-type anorexic patients (16). It is generally accepted
that at least half of treatment-seeking individuals with anorexia or bulimia nervosa
will meet lifetime criteria for a depressive disorder, with rates ranging from
24% to 88% (17–25).
Rates of dysthymic disorder have not been consistently gathered in the same
way that those of major depressive disorder have, though existing data suggest
that this comorbid condition may also be elevated among those with bulimia
nervosa (26). In contrast to unipolar mood disorders, bipolar mood disorders
appear to be relatively uncommon among this population.
Seasonal affective disorder (SAD) has received quite a bit of attention in the
eating disorders field in the past decade. This is because a growing body of research
suggests seasonal variations in mood patterns and eating behaviors among
patients with eating disorders, particularly those with bulimia nervosa. Patients
diagnosed with SAD often demonstrate dysfunctional eating behaviors similar to
those found in bulimic patients. In fact, both disorders are characterized by
disturbances in eating, appetite, weight, mood, sleep, and energy. Research
conducted primarily in North America has estimated the comorbidity rates of
winter SAD to range between 27% and 42%, with lower percentages in other
seasons (27–30), based on the Seasonal Pattern Assessment Questionnaire.
While higher global seasonality scores have been reported among those with
various types of eating disorders compared to control subjects (31), most
studies have found that individuals with bulimia nervosa specifically are most
heavily influenced by seasonality (27,30,32,33). In general, patients with
anorexia nervosa show less seasonal mood and weight variation (32). Although
most of the research has been concentrated in North America, a Japanese study
also cited the greatest seasonal changes in bulimic patients specifically, albeit
less seasonal variation in mood and eating behavior than that found in North
American patients (33). Neurochemical dysregulation, such as serotonin
dysfunction, has been proposed as a possible link between these two disorders
(e.g., 34). Phototherapy has been demonstrated to be effective in decreasing
depressive and bulimic symptoms for some patients with SAD and bulimia
nervosa (35–37).
Some have proposed that depression and eating disorders share a similar
underlying diathesis, which may account for their frequent comorbidity (e.g., 38).
However, other family and twin study data refute the shared diathesis
hypothesis (14,39–41) and suggest that eating disorders are not simply an
alternate form of depressive disorder.
Indeed, there has been considerable controversy regarding the direction of
the association between depression and eating disorders. In an attempt to
unravel the nature of the association, several retrospective studies have found
that a significant proportion of individuals have depressive symptoms before the
onset of their eating disorder (19,21,25,42). However, a more recent review of
this literature suggests that it is more common for the eating disorder to
precede the onset of the depressive disorder (43). If this is the case, we must
remember the potential role of starvation and erratic consummatory patterns in
190 LILENFELD
producing secondary depressive symptoms (9). Thus, in some cases, it is likely

that the primary eating disorder may in fact produce, or at least exaggerate, the
observed depressive symptomatology.
The role of starvation is of such great clinical importance when treating an
eating-disordered patient that the patient should be informed of the likelihood
that her mood may improve with weight restoration and resumption of
normal eating patterns. It is a good idea to review the details of the landmark
Keys et al. (9) semistarvation study with the patient, so that she may appreciate
the nature of her mood symptoms as at least partly due to the consequences of
her disordered eating patterns.
The treating clinician might summarize the following for the patient: In the
1940s, Ancel Keys and colleagues at the University of Minnesota conducted a
landmark study of the effects of semistarvation among healthy young men who
were “conscientious objectors” to World War II. These men were carefully
screened at the beginning of the study to ensure that they were extremely
physically and psychologically healthy. It was discovered that after an imposed
period of semistarvation and a loss of body weight comparable to that observed
among individuals with anorexia nervosa, these previously healthy young men
began to experience numerous disturbing physical and psychological symptoms
that had not been present before their weight loss. These psychological and
behavioral symptoms included depressed mood, anxiety, ruminative thoughts
about food, hoarding, and other unusual food rituals. Thus, the point to
emphasize to one’s depressed anorexic patient is that the restriction of food and
resulting weight loss itself caused such problems in these previously healthy
individuals. Weight restoration was subsequently found to reverse these
symptoms in most subjects. This is an important point to make to the depressed
anorexic patient because, although there is no guarantee, particularly if the
patient evidenced premorbid depression, it is still likely that her mood will
improve with weight restoration. This same point may be made to the patient with
bulimia nervosa, as it is quite likely that the erratic eating patterns inherent in this
disorder may cause, or at least exacerbate, similar mood problems. In addition,
the bulimic patient, though usually at normal body weight, may in fact be below
her expected weight due to food restriction that is commonly observed between
bulimic episodes (44). Thus, the same process may apply and should be
explained to these patients as well.
All of the previously reviewed mood disorder comorbidity research has been
conducted with anorexic or bulimic individuals. Recent studies of individuals
with binge eating disorder similarly suggest elevated rates of depressive
disorders, but not bipolar mood disorders. Comparable to that observed among
treatment-seeking patients with anorexia nervosa and bulimia nervosa,
approximately 50% of binge eating disorder patients have been found to have a
lifetime history of major depressive disorder and high levels of depressive

symptomatology (45–47). In addition, women with binge eating disorder from
a non-treatment-seeking sample evidenced similarly elevated rates of major
depression compared to control subjects (48).
Although it is more common for people to lose weight during depressive
episodes, there is evidence that obese individuals may instead be especially likely
to gain weight while depressed. Results from a large epidemiological study of
female twins suggests the existence of three clinically significant depressive
syndromes, one of which was labeled “atypical depression” (49). This type of
depression was characterized by increased eating, hypersomnia, and a proclivity
to obesity, which describes the usual presentation seen in binge eating disorder
patients. Recent evidence has suggested that this pattern of depression may be
familial (49).
As is the case for anorexia and bulimia nervosa, the causal relationship
between binge eating disorder and depression is unclear. The Kendler
et al. (49) twin study suggests the possibility of a familial tendency toward
depression, which is accompanied by overeating and obesity. However, it is
important to recognize that both obesity and binge eating may contribute to
depressive symptomatology. Onset of binge eating disorder more often appears
to predate that of major depressive disorder (50), similar to the onset pattern
found with other eating disorders (43). Because obesity is associated with
stigmatization in our society, the effects of coping with such prejudice may be
depressogenic. Furthermore, the sense of feeling out of control over eating is
extremely aversive, and may cause or exacerbate depressive symptoms. In fact,
severity of depression among individuals with binge eating disorder has been
found to be strongly associated with the frequency of binge eating (51).
Anxiety Disorders
As with mood disorders, most comorbidity studies of anxiety disorders and
eating disorders have been conducted with anorexia and bulimia nervosa rather
than binge eating disorder, so these findings will be reviewed first. Also similar
to that found with mood disorders, lifetime prevalence rates of
anxiety disorders vary markedly across studies. Rates of any anxiety disorder
have ranged from 23% to 66% among anorexic and 25% to 75% among bulimic
individuals (22,23,25,38,42,52–62).
When subtypes of anorexic and bulimic subjects are examined, similar to
findings with depression, bulimic symptomatology appears to be more
predictive of anxiety disorder comorbidity. Specifically, bulimic individuals
have been found to have significantly higher lifetime rates of comorbid anxiety
disorders than restricting-type anorexic subjects, whereas anorexic subjects
192 LILENFELD
with bulimic symptomatology (i.e., bingeing-purging anorexia nervosa) appear

to fall in between the two groups (16).
Two of the specific anxiety disorders that are among the most common in
their co-occurrence with both anorexia and bulimia nervosa are social phobia
and OCD. Lifetime prevalence rates among clinical samples of patients with
anorexia nervosa range from 24% to 55% for social phobia and from 10% to
66% for OCD (14,16,52,63,64). Lifetime prevalence rates among clinical
samples of patients with bulimia nervosa range from 17% to 59% for social
phobia and from 3% to 43% for OCD (14,16,25,38,52,55,63–65). Two
studies of bulimic individuals from the community found a lifetime prevalence
rate of 42% to 46% for social phobia and one found a rate of 9% for
OCD (24,61).
In making a comorbid diagnosis of social phobia in an eating-disordered
patient, it is important to assess whether the patient’s social fears are limited to
eating behavior alone, as this is not at all uncommon among anorexic individuals
in particular. If the fear of humiliation or embarrassment and avoidance is not
specific to food and food-related cues, then an additional diagnosis of social
phobia may be considered. There has not been much focus on the potential
reasons for this comorbidity in the eating disorders literature. The fact that
social phobia has among the highest anxiety disorder comorbidity rates in an
eating-disordered population may simply be a function of the fact that it is the
most common anxiety disorder in the general population as well. Thus far, no
twin or family study data have suggested a meaningful etiological link between
anorexia or bulimia nervosa and social phobia.
In contrast, much has been written about a potentially meaningful
relationship between eating disorders, particularly anorexia nervosa, and OCD.
Similar to determining whether a comorbid diagnosis of social phobia is
warranted, in this case it is important for the clinician to evaluate whether the
content of the obsessions and/or compulsions is restricted to food, eating, or
weight concerns. It is very common for anorexic patients to engage in eating
rituals, such as chewing one’s food a specific number of times or cutting food
into a specific number of pieces. It is also common for anorexic and bulimic
patients to be preoccupied with thoughts and images about eating and weight
loss. Before making a comorbid diagnosis of OCD, it is important to ensure that
the “obsessions” and “compulsions” are not restricted to these areas of focus.
Indeed, many anorexic and bulimic patients have additional preoccupations
and rituals that warrant a comorbid diagnosis of OCD. Some have suggested
that anorexia nervosa is an etiologically similar variant of OCD (e.g., 5) and
may best be considered on the obsessive-compulsive spectrum of illness. While
this fits with our clinical experience of frequently obsessive and ritualistic
anorexic patients, it is not supported by empirical data. Indeed, several family
studies of anorexic and bulimic individuals have found elevated rates of OCD
among their relatives (14,15,22,66), which initially seemed suggestive of a
potentially shared etiology. However, closer examination of the patterns of
cosegregation of OCD and eating disorders suggests that these two disorders are
actually independently transmitted in families (14,15). That is, OCD has been
found to be elevated primarily among the relatives of those eating-disordered
individuals who themselves have OCD. Thus, although OCD and eating
disorders frequently co-occur within individuals and within families, there is no
evidence of a shared etiological factor from family study data. However, recent
research suggests the possibility that OCD and anorexia nervosa may both be
linked to a polymorphism of the 5-HT2a receptor (67–72), although some
studies have not supported this association for either or both disorders (73,74).
In addition to OCD and social phobia, post-traumatic stress disorder (PTSD)
is an anxiety disorder of great importance to consider in individuals with eating
disorders because it coexists specifically with binge eating problems so
frequently. Lifetime prevalence rates of PTSD among bulimic individuals have
been found to range from 11% to 52% (14,75–77). Most notably, the National
Women’s Study, a very large and representative epidemiological sample,
yielded lifetime prevalence rates of 37% among women with bulimia nervosa
and 22% among women with binge eating disorder (77). Elevated rates of both
sexual and aggravated assault among women with bulimia nervosa suggests that
victimization may contribute to the development of and/or maintenance of
bulimia nervosa. Others have concluded that while PTSD is a common and
important clinical variable among women with eating disorders, it may not be
directly related to the eating disorder per se but rather to the comorbid
depressive, anxiety, and dissociative symptoms that often coexist with an eating
disorder (76).
The suggestion that there may be a meaningful etiological relationship
between anxiety disorders and eating disorders remains. Because anxiety
disorders often precede the development of the eating disorder (25,78), it
has been suggested that anxiety disorders may serve as a risk factor for the
development of anorexia nervosa or bulimia nervosa. We are currently awaiting
more data to confirm or refute this hypothesis.
Anxiety disorder comorbidity appears to be more clearly linked with
anorexia nervosa and bulimia nervosa than binge eating disorder. Lifetime
prevalence rates of any anxiety disorder in obese binge eaters have ranged from
9% to 46% (45,47,48,79–81). Only one of these studies (45) has found
significantly increased rates of anxiety disorders among obese binge eaters
compared to obese nonbinge eaters. One issue to consider in reviewing these
findings is that the diagnostic criteria used for binge eating in some earlier
studies was not DSM-IV binge eating disorder. However, again, all but one
194 LILENFELD
study (45) that did use the current definition of binge eating disorder found no
significant group differences in lifetime rates of anxiety disorders. All of these
studies utilized treatment-seeking samples. One recent comorbidity study of a
non-treatment-seeking community sample of binge eating disorder individuals
likewise suggests no significantly elevated rates of any anxiety disorders
compared with non-eating-disordered overweight women from the
community (48). The one important exception to these findings is the National
Women’s Study, which found PTSD to occur at a rate of 22% among women with
binge eating disorder in the community (77).
At present, we can conclude that anxiety disorders are common among
anorexic and bulimic patients, though we know less about those individuals with
anorexia nervosa and bulimia nervosa in the community. Thus far, the common
co-occurrence does not seem to be due to a shared etiologic factor between
eating disorders and anxiety disorders, though more family and twin study
research would aid in making a firmer conclusion. Individuals with binge eating
disorder, including both treatment-seeking as well as community samples, do
not appear to have similarly elevated rates of anxiety disorders, with the
possible exception of PTSD. However, many fewer studies of binge eating
disorder have been conducted compared to anorexia nervosa and bulimia
nervosa.
Substance Use Disorders

Substance use disorders are the type of comorbidity where distinguishing
between anorexia nervosa and bulimia nervosa is most critical. Individuals
with bulimic symptomatology in the form of bulimia nervosa or the
bingeingpurging type of anorexia nervosa, have elevated rates of substance use
disorders (23,58,82), whereas those with restricting-type anorexia nervosa do
not (23,58). Since 1977, more than 50 studies have examined the relationship
between bulimic symptomatology and substance use disorders (83–85).
In a review of clinical samples of bulimic individuals, Holderness
et al. (83) estimated that 23% had alcohol abuse or dependence. Interestingly,
one recent prospective longitudinal study (86) found that anorexic women with
bulimic symptomatology were nearly seven times more likely to develop
substance use disorders than those with restricting-type anorexia nervosa.
Questions of a potentially shared vulnerability between bulimia nervosa and
substance use disorders have been raised, similar to the question regarding the
nature of the relationship between anorexia nervosa and OCD, for similar
reasons. First, rates of comorbidity are high. That is, a higher proportion of
bulimic individuals have a comorbid substance use disorder than would be
expected by chance. Second, there is overlap in the clinical presentation, at
least among a subset of eating-disordered individuals. Some have written about

“multi-impulsive bulimia” (87,88), characterized by multiple impulsive
behaviors, such as binge eating, substance abuse, shoplifting, promiscuity, and
reckless driving (reviewed below). Many have theorized that this shared
impulsivity between bulimia nervosa and substance use disorders may account
for the frequently observed comorbidity. Third, there are elevated rates of
substance use disorders among the relatives of bulimic individuals (13,89–91).
Again, this may be suggestive of a potentially shared etiology. However, closer
examination of the patterns of cosegregation of bulimia nervosa and substance
use disorders suggests that these two disorders are actually independently
transmitted in families (13,15), similar to what has been concluded about the
relationship between eating disorders and OCD. This conclusion comes from the
finding that substance use disorders are elevated primarily among the relatives
of those bulimic individuals who themselves have a substance use disorder.
Additional findings supporting this conclusion come from examining rates and
patterns of eating disorders among family members of alcohol-dependent
individuals (92). Finally, these family study findings converge with multivariate
genetic modeling of the large population-based twin study database investigated
by Kendler and colleagues (41), in which it was found that bulimia nervosa and
alcoholism were attributable to distinct genetic factors.
Thus, although substance use disorders and bulimia nervosa frequently
co-occur within individuals and within families, there is no evidence of a shared
etiological factor from these family study data. Interestingly, however, data
from a large national, representative, epidemiological sample of women suggest
that comorbid PTSD may be one explanation for why alcohol abuse so
commonly coexists with bulimia nervosa (93).
As has been true for mood and anxiety disorders, while there is a substantial
literature demonstrating elevated rates of substance use disorders in women
with bulimia nervosa (83) and the relatives of substance-abusing bulimics,
much less is known about substance use disorders in binge eating disorder
patients and their families. Lifetime rates of any substance use dis order have
ranged from 8% to 33% in studies of individuals with DSM-IV binge eating
disorder (45,47,48,81,94,95). All but one (81) of these studies found no
increased rates among the binge eating disorder group compared to a control
group. The first of these studies (45) found no increased rate of substance abuse
among obese binge eating-disordered patients themselves but, interestingly, did
find an elevated rate of substance abuse in the family members of obese binge
eating-disordered subjects compared to those of obese non-binge eating-
disordered subjects. Likewise, more recent studies of a non-treatment-seeking
sample (48), as well as a treatment sample compared to a general psychiatric
control group (95), found no evidence of increased rates of substance use
196 LILENFELD
disorders among binge eating-disordered individuals. Thus, most of the modest

existing literature suggests no elevated rates of substance use disorders among
binge eating-disordered individuals compared to control groups.
In summary, many bulimic and bingeing-purging anorexic patients
experience alcohol and/or drug problems, sometimes simultaneously. As a
result, eating disorders have often been viewed as a form of addiction. The
addiction model of eating disorders is illustrated by the 12-step approach of
Overeaters Anonymous and similar treatment programs. One assumption of
this treatment approach is that some people are biologically vulnerable to
certain foods (e.g., sugar) that can cause chemical dependence; thus, abstinence
from these foods is required. A second assumption is that an eating disorder is a
lifelong illness that can be managed but never truly eliminated. However, most
empirical research does not support these assumptions as they apply to eating
disorders and, thus, does not support the notion of eating disorders as a form of
addiction like alcoholism (96). More research is needed to understand the
reasons behind the co-occurrence of bulimic disorders and substance use
disorders, as well as the reasons for the lack of apparent relationship between
binge eating disorder and substance use disorders. A final interesting area of
future research is the possibility of a protective function against substance use
disorders served by restricting-type anorexia nervosa.
Impulse Control Disorders

As previously mentioned, some researchers have described a condition called
“multi-impulsive bulimia” (87,88), characterized by multiple impulsive
behaviors, such as binge eating, substance abuse, shoplifting, promiscuity, and
reckless driving. Indeed, much has been written about impulsivity in women
with eating disorders, particularly bulimia nervosa (e.g., 97). It has been
suggested by numerous investigators that decreased serotonin functioning may
account for such impulsive behavior. In addition to bingeing and
purging behavior, a subset of these patients also engage in intentional self-injury
(e.g., 98,99). In contrast, there are very few reports of comorbid DSM-IV
impulse control disorders, such as intermittent explosive disorder,
kleptomania, pyromania, pathological gambling, and trichotillomania, with a
few exceptions where shoplifting and other theft were reported in clinical
samples (100, 101).
Thus, impulse control problems most typically take the form of substance
abuse and self-injury among individuals with eating disorders, specifically those
with bulimic symptomatology. Of greatest clinical relevance, such impulsive
behaviors are associated with poor long-term prognosis among those with bulimic
symptomatology (i.e., bulimia nervosa or bingeing-purging anorexia nervosa),

even following treatment (102,103).
Schizophrenia and Other Psychotic Disorders

Investigations of comorbidity between psychotic disorders and eating disorders
has primarily focused on anorexia nervosa. Indeed, the delusional nature of the
body distortion and fear of fatness among clearly underweight and sometimes
emaciated individuals is noteworthy. The general consensus is that in most cases
where psychotic symptoms are present, they are likely to be secondary to the
eating disorder (e.g., as a result of malnutrition; 9) or secondary to another
comorbid disorder rather than a primary psychotic illness. A systematic study of
130 consecutive eating disorder inpatients found that 17 displayed psychotic
symptoms (104). Nearly all of these patients’ psychotic symptoms occurred in
the context of a primary mood disorder. No cases of schizophrenia or organic
psychosis were identified. Thus, the psychotic symptoms in each case were
attributable to a psychiatric disorder other than the eating disorder, nearly
always a mood disorder.
A few early reports describe several patients with both anorexia nervosa and
schizophrenia (105,106). However, further examination of these cases suggests
that many of these patients actually had major depression with psychotic
features rather than schizophrenia (104). Thus, a diagnosis of major mood
disorder should be considered likely in any eating disorder patient who displays
psychotic symptoms.
It has been suggested that there may be a link between psychosis and eating
disorders via the 5-HT1c receptor (107). Likewise, antipsychotic medications,
such as olanzapine and risperidone, are now often used in the management of
anorexia nervosa, even for patients without clear psychosis, in order to target
the delusional nature of body image distortions (108,109). Substantial weight
gain is an obvious side effect with interesting implications for the management of
a disorder whose hallmark symptom is an intense fear of fatness.
Dissociative Disorders
Nearly all studies of dissociative pathology and eating disorders have examined
dissociation as a continuous variable (110–112), as opposed to diagnosable
dissociative disorders. Specifically, the Dissociative Experiences Scale (113) is
the most commonly used measure for this purpose. Interestingly, however,
Waller and colleagues (114) successfully created a more effective categorical
measure of dissociation from this dimensional assessment instrument and used it
to classify eating disorder patients as “high” or “low” dissociators. Dissociation
198 LILENFELD
has been demonstrated particularly among patients with a history of trauma

(e.g., 115), self-mutilation (e.g., 116,117), and those who have a psychiatric
disorder that involves impulsive behaviors, such as substance abuse and
bulimia nervosa (118). It is thought that these behaviors serve the function of
allowing the individual to escape from an awareness of intolerable emotional
states (119,120). Among eating disorder populations, anorexic patients with
bingeing and purging symptomatology and bulimic patients demonstrate the most
pathological level of dissociation (118,121).
Somatoform Disorders
Very little empirical research on somatoform and eating disorder comorbidity
has been conducted. However, “somatoform dissociation” has been assessed
with the Somatoform Dissociation Questionnaire and has been found to be
characteristic of patients with dissociative disorders and a core feature in many
patients with somatoform disorders, as well as in some patients with eating
disorders (122,123). Examples of somatoform dissociation include motor
inhibitions and analgesia. It is not surprising that these symptoms are strongly
associated with a history of trauma (122).
ADHD and Disruptive Behavior Disorders of

Childhood
Little research on these childhood disorders has been conducted in the field of
eating disorders. ADHD and disruptive behavior disorders of childhood are
considered to be virtually nonexistent among individuals with restricting-type
anorexia nervosa, as these individuals instead report having been well behaved,
perfectionists, and having excelled in school, all of which are corroborated by
family member reports. Some individuals with bulimia nervosa likewise have
similar childhood histories. However, some report disruptive disorders of
childhood to predate their eating disorder. These are typically individuals who
would be classified as having multi-impulsive bulimia. There has been some
suggestion that eating might help fulfill the need for high stimulation among
individuals with ADHD, decrease agitation, or satisfy a need for control (124).
While some recommend screening for ADHD in bulimic and binge eating
disorder populations because of the presence of poor self-control and
impulsivity in this population, there is little empirical evidence to support the
suggested comorbid link.
Interestingly, there is one report of oppositional defiant disorder occurring at
elevated rates in adolescents with eating disorders where bingeing and purging
symptoms are present (125). In addition, there is one report of an association
between eating-disordered behavior and aggressive conduct in adolescent

girls (126). Conduct disorder per se was not assessed, but these authors
examined a continuous measure of conduct disorder-like symptoms, including
robbery and aggravated battery. They found that the odds of involvement in
aggressive behavior was significantly higher among those adolescent girls
endorsing eating disturbances, including not only bingeing and purging
behavior, but also food restriction. This finding contrasts with the typical clinical
description of the average eating-disordered adolescent who is usually described
as passive, compliant, and rarely aggressive (e.g., 105). Thus, there is little
empirical research on disruptive behavior disorders in this population, but there
are several reports suggestive of an association between eating disturbance and
aggressive behavior during adolescence.
SUMMARY
There are many reasons for the coexistence of two disorders. Reasons for the
comorbidity of eating disorders and other psychiatric disorders is under study
and much remains to be learned. We know that among clinical samples,
elevated rates of depressive disorders, particularly major depressive disorder,
have been observed among individuals with all types of eating disorders. In
addition, elevated rates of anxiety disorders, particularly social phobia and OCD,
have been found among individuals with anorexia nervosa and bulimia nervosa,
but not binge eating disorder. Substance use disorders are common among
individuals with bulimic symptomatology, in the form of bulimia nervosa or
bingeing-purging anorexia nervosa, but not binge eating disorder.
It is important to remember that most comorbidity studies have been
conducted with samples of convenience (i.e., treatment samples) and thus are
subject to referral bias, which will usually result in higher comorbidity rates
than would be found in a community sample. This is likely to occur because
people with more than one disorder are more likely to seek treatment (i.e.,
Berkson’s bias). Thus, in general, a clinical sample is likely to be more disturbed
than an epidemiological sample.
Finally, it is also important to remember that the comorbidity observed with
eating disorders (especially depressive and anxiety disorders) may be created, or
at least exacerbated, by malnutrition and pathological eating behaviors. The
extent to which such comorbid pathology actually increases the risk of
developing an eating disorder remains the focus of ongoing research.
200 LILENFELD
REFERENCES
1. Feinstein AR. The pre-therapeutic classification of co-morbidity in chronic disease.
J Chron Dis 1970; 23:455–468.
2. Klein DN, Riso LP. Psychiatric disorders: problems of boundaries and
comorbidity. In: Costello CG, ed. Basic Issues in Psychopathology. New York:
Guilford, 1993:19–66.
3. American Psychiatric Association. Diagnostic and statistical manual of mental
disorders. 4th ed. Washington, D.C., 1994.
4. Berkson J. Limitations of the application of fourfold table analysis to hospital data.
Biometrics 1946; 2:47–53.
5. Rothenberg A. Eating disorder as a modern obsessive-compulsive syndrome.
Psychiatry 1986; 49:45–52.
6. Fahy TA, Osacar A, Marks I. History of eating disorders in female patients with
obsessive-compulsive disorder. Int J Eat Disord 1993; 14:439–443.
7. Westen D. Unresolved issues in the classification, diagnosis, and comorbidity of
eating disorders. In: Maj M, Halmi K, Lopez-Ibor JJ, Sartorius N. eds. World
Psychiatric Association: Evidence and Experience in Psychiatry. Vol. 6. Eating
Disorders, New York: John Wiley and Sons, 2003:34–37.
8. Gleaves DH, Lowe MR, Snow AC. Continuity and discontinuity models of bulimia
nervosa: a taxometric investigation. J Abnorm Psychol 2000; 109:56–68.
10. Strober M. Personality and symptomatological features in young, nonchronic
anorexia nervosa patients. J Psychosom Res 1980; 24:353–359.
11. Srinivasagan NM, Plotnicov KH, Greeno C, Weltzin TE, Rao R, Kaye WH.
Persistent perfectionism, symmetry, and exactness in anorexia nervosa after long-
term recovery. Am J Psychiatry 1995; 152:1630–1634.
12. Pollice CP, Kaye WH, Greeno CG, Weltzin TE. Relationship of depression,
anxiety, and obsessionality to state of illness in anorexia nervosa. Int J Eat Disord
1997; 21:367–376.
13. Kaye WH, Lilenfeld LR, Plotnicov K, Merikangas KR, Nagy L, Strober M, Bulik
CM, Moss H, Greeno CG. Bulimia nervosa and substance dependence. Alcohol
Clin Exp Res 1996; 20:878–881.
14. Lilenfeld LR, Kaye WH, Greeno CG, Merikangas KR, Plotnicov K, Pollice C, Rao
R, Strober M, Bulik CM, Nagy L. A controlled family study of anorexia nervosa
and bulimia nervosa: psychiatric disorders in first-degree relatives and effects of
proband comorbidity. Arch Gen Psychiatry 1998; 55:603–610.
15. M. Strober. Unpublished data, 2000.
16. Laessle RG, Wittchen HU, Fichter MM, Pirke KM. The significance of subgroups
of bulimia and anorexia nervosa: lifetime frequency of psychiatric disorders. Int J
Eat Disord 1989; 8:569–574.
17. Hatsukami D, Eckert E, Marchi P, Sampugnaro V, Apple R, Cohen J. Affective
disorders and substance abuse in women with bulimia. Psychol Med 1984; 14:
701–704.
18. Herzog DB. Are anorexic and bulimic patients depressed? Am J Psychiatry 1984;
141:1594–1597.
19. Hudson JI, Pope HG, Jonas JM, Yurgelun-Todd D. Phenomenologic relationship
of eating disorders to major affective disorder. Psychiatry Res 1983; 9:345–354.
20. Herzog DB. Bulimia in the adolescent. Am J Dis Child 1982; 136:985–989.
21. Walsh BT, Roose SP, Glassman AH, Glades MA, Sadik C. Depression and bulimia.
Psychosom Med 1985; 47:123–131.
22. Halmi KA, Eckert E, Marchi P, Sampugnaro V, Apple R, Cohen J. Comorbidity of
psychiatric diagnoses in anorexia nervosa. Arch Gen Psychiatry 1991; 48:712–718.
23. Herzog D, Keller M, Sacks N, Yeh C, Lavori P. Psychiatric comorbidity in
treatment seeking anorexics and bulimics. J Am Acad Child Adol Psychiatry 1992;
31:810–818.
24. Bushnell JA, Wells JE, McKenzie JM, Hornblow AR, Oakley-Browne MA, Joyce
PR. Bulimia comorbidity in the general population and in the clinic. Psychol Med
1994; 24:605–611.
25. Brewerton TD, Lydiard RB, Herzog DB, Brotman AW, O’Neil PM, Ballenger JC.
Comorbidity of axis I psychiatric disorders in bulimia nervosa. J Clin Psychiatry
1995; 56:77–80.
26. Gwirtsman HE, Roy-Byrne P, Yager J, Gerner RH. Neuroendocrine abnormalities
in bulimia. Am J Psychiatry 1983; 140:559–563.
27. Blouin A, Blouin J, Aubin P, Carter J, Goldstein C, Boyer H, Perez E. Seasonal
patterns of bulimia nervosa. Am J Psychiatry 1992; 149:73–81.
28. Ghadirian AM, Marini N, Jabalpurwala S, Steiger H. Seasonal mood patterns in
eating disorders. Gen Hosp Psychiatry 1999; 21:354–359.
29. Lam RW, Solymon L, Tompkins A. Seasonal mood symptoms in bulimia nervosa
and seasonal affective disorder. Compr Psychiatry 1991; 32:552–558.
30. Lam RW, Golner EM, Grewal A. Seasonality of symptoms in anorexia and bulimia
31. Brewerton TD, Krahn DD, Hardin TA, Wehr TA, Rosenthal NE. Findings from
the Seasonal Pattern Assessment Questionnaire in patients with eating disorders and
control subjects: effects of diagnosis and location. Psychiatry Res 1994; 52:71–84.
32. Fornari VM, Braun DL, Sunday SR, et al. Seasonal patterns in eating disorder
subgroups. Compr Psychiatry 1994; 35:450–456.
33. Yamatsuji M, Yamashita T, Arii I, Taga C, Tatara N, Fukui K. Seasonal variations in
eating disorder subtypes in Japan. Int J Eat Disord 2003; 33:71–77.
34. Brewerton TD, Brandt HA, Lesem MD, Murphy DL, Jimerson DC. Serotonin in
eating disorders. In: Coccaro EF, Murphy DL, eds. Serotonin in Major Psychiatric
Disorders. Spiefel D, ed. Progress in Psychiatry Monograph Series. Washington,
DC: American Psychiatric Press, 1990:153–184.
35. Lam RW, Lee SK, Tam EM, Grewal A, Yatham LN. An open trial of light therapy
for women with seasonal affective disorder and comorbid bulimia nervosa. J Clin
Psychiatry 2001; 62:164–168.
202 LILENFELD
36. Braun DL, Sunday SR, Fornari VM, Halmi KA. Bright light therapy decreases
winter binge frequency in women with bulimia nervosa: a double-blind, placebo-
controlled study. Compr Psychiatry 1999; 40:442–448.
37. Blouin AG, Blouin JH, Iversen H, Carter J, Goldstein C, Goldfield G, Perez E.
Light therapy in bulimia nervosa: a double-blind placebo-controlled study.
38. Hudson JI, Pope HG, Yurgelun-Todd D, Jonas JM, Frankenburg FR. A controlled
study of lifetime prevalence of affective and other psychiatric disorders in bulimic
outpatients. Am J Psychiatry 1987; 144:1283–1287.
39. Strober M, Lampert C, Morrell W, Burroughs J, Jacobs C. A controlled family
study of anorexia nervosa: evidence of familial aggregation and lack of shared
transmission with affective disorders. Int J Eat Disord 1990; 9:239–253.
40. Biederman J, Rivinus T, Kemper K, Hamilton D, MacFayden J, Harmatz J.
Depressive disorders in relatives of anorexia nervosa patients with and without a
current episode of nonbipolar major depression. Am J Psychiatry 1985; 142:
1495–1496.
41. Kendler KS, Walters EE, Neale MC, Kessler RC, Heath AC, Eaves LJ. The structure
of the genetic and environmental risk factors for six major psychiatric disorders in
women. Arch Gen Psychiatry 1995; 52:374–383.
42. Piran N, Kennedy S, Garfinkel PE, Owens M. Affective disturbance in eating
disorders. J Nerv Ment Dis 1985; 173:395–400.
43. Cooper PJ. Eating disorders and their relationship to mood and anxiety disorders.
In: Brownell KD, Fairburn CG, eds. Eating Disorders and Obesity: A
Comprehensive Handbook. New York: Guilford Press, 1995:159–164.
44. Vitousek K, Manke F. Personality variables and disorders in anorexia nervosa and
bulimia nervosa. J Abnorm Psychol 1994; 103:137–147.
45. Yanovski SZ, Nelson JE, Dubbert BK, Spitzer RL. Association of binge eating
disorder and psychiatric comorbidity in the obese. Am J Psychiatry 1993; 150:
1472–1479.
46. Marcus MD, Wing RR, Ewing L, Kern E, Gooding W, McDermott M. Psychiatric
disorders among obese binge eaters. Int J Eat Disord 1990; 9:69–77.
47. Specker S, de Zwaan M, Raymond N, Mitchell J. Psychopathology in subgroups of
obese women with and without binge eating disorder. Compr Psychiatry 1994; 35:
185–190.
48. Telch CF, Stice E. Psychiatric comorbidity in women with binge eating disorder:
prevalence rates from a non-treatment-seeking sample. J Consult Clin Psychol
1998; 66:768–776.
49. Kendler KS, Eaves LJ, Walters EE, Neale MC, Heath AC, Kessler RC.
The identification and validation of distinct depressive syndromes in a
populationbased sample of female twins. Arch Gen Psychiatry 1996; 53:391–399.
50. Mussell MP, Mitchell JE, Weller CL, Raymond NC, Crow SJ, Crosby RD. Onset
of binge eating, dieting, obesity, and mood disorders among subjects seeking
treatment for binge eating disorder. Int J Eat Disord 1995; 17:395–401.
51. Smith DE, Marcus MD, Lewis C, Fitzgibbon M, Schreiner P. Prevalence of binge
eating disorder, obesity and depression in a biracial cohort of young adults. Ann
Behav Med 1998; 20:227–232.
52. Fornari V, Kaplan M, Sandberg DE, Mathews M, Skolnick N, Katz JL. Depressive
and anxiety disorders in anorexia nervosa and bulimia nervosa. Int J Eat Disord
1992; 12:21–29.
53. Laessle RG, Kittl S, Fichter MM, Wittchen H, Pirke KM. Major affective disorder
in anorexia nervosa and bulimia: a descriptive diagnostic study. Br J Psychiatry
1987; 151:785–789.
54. Toner BB, Garfinkel PE, Garner DM. Affective and anxiety disorders in the long-
term follow-up of anorexia nervosa. Int J Psychiatry Med 1988; 18:357–364.
55. Schwalberg MD, Barlow DH, Alger SA, Howard LJ. Comparison of bulimics,
obese binge eaters, social phobics, and individuals with panic disorder on
comorbidity across DSM-III-R anxiety disorders. J Abnorm Psychol 1992; 101:
675–681.
56. Smith C, Feldman SS, Nasserbakht A, Steiner H. Psychological characteristics and
DSM-III-R diagnoses at 6-year follow-up of adolescent anorexia nervosa. J Am
Acad Child Adol Psychiatry 1993; 32:1237–1245.
57. Bossert-Zaudig S, Zaudig M, Junker M, Wiegand M, Krieg JC. Psychiatric
comorbidity of bulimia nervosa inpatients: relationship to clinical variables and
treatment outcome. Eur Psychiatry 1993; 8:15–23.
58. Braun DL, Sunday S, Halmi KA. Psychiatric comorbidity in patients with eating
disorders. Psychol Med 1994; 24:859–867.
59. Deep AL, Nagy LM, Weltzin TE, Rao R, Kaye WH. Premorbid onset of
psychopathology in long-term recovered anorexia nervosa. Int J Eat Disord 1995;
17:291–297.
60. Thiel A, Broocks A, Ohlmeier M, Jacoby GE, Schussler G. Obsessive-compulsive
disorder among patients with anorexia nervosa and bulimia nervosa. Am J
Psychiatry 1995; 152:72–75.
61. Garfinkel PE, Lin E, Goering P, Speff C, Goldbloom DS, Kennedy S, Kaplan AS,
62. Keck PE Jr, Pope HG Jr, Hudson JI, McElroy SL, Yurgelun-Todd D, Hundert EM.
A controlled study of phenomenology and family history in outpatients with
bulimia nervosa. Compr Psychiatry 1990; 31:275–283.
63. Thornton C, Russell J. Obsessive compulsive comorbidity in the dieting disorders.
64. Godart NT, Flament MF, Lecrubier Y, Jeammet P. Anxiety disorders in ano rexia
nervosa and bulimia nervosa: comorbidity and chronology of appearance. Eur
Psychiatry 2000; 15:38–45.
65. Powers PS, Coovert DL, Brightwell DR, Stevens BA. Other psychiatric disorders
among bulimic patients. Compr Psychiatry 1988; 29:503–508.
66. Pasquale L, Sciuto G, Cocchi S, Ronshi P, Billodi L. A family study of obsessive
compulsive, eating, and mood disorders. Eur Psychiatry 1994; 9:33–38.
204 LILENFELD
67. Collier DA, Arranz MJ, Li T, Mupita D, Brown N, Treasure J. Association

between 5-HT2A gene promoter polymorphism and anorexia nervosa. Lancet
1994; 350:412.
68. Hinney A, Ziegler A, Nothen MM, Remschmidt H, Hebebrand J. 5-HT2A receptor
gene polymorphisms, anorexia nervosa, and obesity. Lancet 1997; 350:
1324–1325.
69. Enoch MA, Kaye WH, Rotondo A, Greenberg BD, Murphy DL, Goldman D.
5-HT2A promoter polymorphism -1438G/A, anorexia nervosa, and
obsessivecompulsive disorder. Lancet 1998; 351:1785–1786.
70. Sorbi S, Nacmias B, Tedde A, Ricca V, Mezzani B, Rotella CM. 5-HT2A promoter
polymorphism in anorexia nervosa. Lancet 1998; 351:1785.
71. Nacmias B, Ricca V, Tedde A, Mezzani B, Rotella CM, Sorbi S. 5-HT2A receptor
gene polymorphisms in anorexia nervosa and bulimia nervosa. Neurosci Lett 1999;
277:134–136.
72. Nishiguchi N, Matsushita S, Suzuki K, Muryama M, Shirakawa O, Higuchi S.
Association between 5HT2A receptor gene promoter region polymorphism and
eating disorders in Japanese patients. Biol Psychiatry 2001; 50:123–128.
73. Ziegler A, Hebebrand J, Gorg T, Rosenkranz K, Fichter MM, HerpertzDahlmann
B, Remschmidt H, Hinney A. Further lack of association between the 5-HT2A gene
promoter polymorphism and susceptibility to eating disorders and a meta-analysis
pertaining to anorexia nervosa. Mol Psychiatry 1999; 4: 410–412.
74. Karwautz A, Rabe-Hesketh S, Hu X, Zhao J, Sham P, Collier DA, Treasure JL.
Individual-specific risk factors for anorexia nervosa: a pilot study using a discordant
sister-pair design. Psychol Med 2001; 31:317–329.
75. Turnbull SJ, Troop NA, Treasure JL. The prevalence of post-traumatic stress
disorder and its relation to childhood adversity in subjects with eating disorders.
Eur Eating Dis Rev 1997; 5:270–277.
76. Gleaves DH, Eberenz KP, May MC. Scope and significance of posttraumatic
symptomatology among women hospitalized for an eating disorder. Int J Eat Disord
1998; 24:147–156.
77. Dansky BS, Brewerton TD, Kirkpatrick DG, O’Neil PM. The National Women’s
Study: relationship victimization and posttraumatic stress disorder to bulimia
78. Bulik CM, Sullivan PF, Fear JL, Joyce PR. Eating disorders and antecedent anxiety
disorders: a controlled study. Acta Psychiatr Scand 1997; 96:101–107.
79. Hudson JI, Pope HG Jr, Yurgelun-Todd D. Bulimia and major affective disorder:
experience with 105 patients. Psychiatr Psychobiol 1988; 3:37–47.
80. de Zwaan M, Nutzinger DO, Schoenbeck G. Binge eating in overweight women.
Compr Psychiatry 1992; 33:256–261.
81. Mussell MP, Mitchell JE, de Zwaan M, Crosby RD, Seim HC, Crow SJ. Clinical
characteristics associated with binge eating in obese females: a descriptive study.
Int J Obes Relat Metab Disorders 1996; 20:324–331.
82. Bulik CM, Sullivan PF, Epstein LH, McKee M, Kaye WH, Dahl RE, Weltzin TE.
Drug use in women with anorexia and bulimia nervosa. Int J Eat Disord 1992; 11:
213–225.
83. Holderness CC, Brooks-Gunn J, Warren WP. Co-morbidity of eating disorders
and substance abuse. Int J Eat Disord 1994; 16:1–34.
84. Krahn DD. The relationship of eating disorders and substance abuse. J Subst Abuse
1991; 3:239–253.
85. Wilson GT. The addiction model of eating disorders: a critical analysis. Adv Behav
Res Ther 1991; 13:27–72.
86. Strober M, Freeman R, Bower S, Rigali J. Binge eating in anorexia nervosa predicts
later onset of substance use disorder: a ten-year prospective, longitudinal follow-up
of 96 adolescents. J Youth Adol 1996; 25:519–532.
87. Lacey J, Evans C. The impulsivist: a multi-impulsive personality disorder. Br J Addict
1986; 81:641–649.
88. Fichter MM, Quadflieg N, Rief W. Course of multi-impulsive bulimia. Psychol
Med 1994; 24:591–604.
89. Lilenfeld LR, Kaye WH, Greeno CG, Merikangas KR, Plotnicov K, Pollice C, Rao
R, Strober M, Bulik CM, Nagy L. Psychiatric disorders in women with bulimia
nervosa and their first-degree relatives. Int J Eat Disord 1997; 22:253–264.
90. Bulik CM. Family histories of bulimic women with and without comorbid alcohol
abuse or dependence. Am J Psychiatry 1991; 148:1267–1268.
91. Mitchell JE, Hatsukami D, Pyle R, Eckert E. Bulimia with and without a family
history of drug abuse. Addict Behav 1988; 12:245–251.
92. Schuckit MA, Tipp JE, Anthenelli RM, Bucholz KK, Hesselbrock VM, Nurnberger
JI. Anorexia nervosa and bulimia nervosa in alcohol-dependent men and women
and their relatives. Am J Psychiatry 1996; 153:74–82.
93. Dansky BS, Brewerton TD, Kilpatrick DG. Comorbidity of bulimia nervosa and
alcohol use disorders: results from the National Women’s Study. Int J Eat Disord
2000; 27:180–190.
94. Brody MJ, Walsh BT, Devlin MJ. Binge eating disorder: reliability and validity of a
new diagnostic category. J Consult Clin Psychol 1994; 62:381–386.
95. Wilfley DE, Friedman MA, Dounchis JZ, Stein RI, Welch RR, Ball SA. Comorbid
psychopathology in binge eating disorder: relation to eating disorder severity at
baseline and following treatment. J Consult Clin Psychol 2000; 68: 641–649.
96. Wilson GT. Eating disorders and addictive disorders. In: Brownell KD, Fairburn
CG, eds. Eating Disorders and Obesity: A Comprehensive Handbook. New York:
Guilford Press, 1995:165–170.
97. Steiger H, Lehoux PM, Gauvin L. Impulsivity, dietary control and the urge to
binge in bulimic syndromes. Int J Eat Disord 1999; 26:261–274.
98. Favaro A, Santonastaso P. Impulsive and compulsive self-injurious behavior in
bulimia nervosa: prevalence and psychological correlates. J Nerv Ment Dis 1998;
186:157–165.
99. Lledo EP, Waller G. Bulimic psychopathology and impulsive behaviors among
nonclinical women. Int J Eat Disord 2001; 29:71–75.
206 LILENFELD
100. Bridgeman J, Slade PD. Shoplifting and eating disorders: a psychological medical-
legal perspective. Eur Eat Dis Rev 1996; 4:133–148.
101. Penas-Lledo E, Vaz FJ, Ramos MI, Waller G. Impulsive behaviors in bulimic patients:
relation to general psychopathology. Int J Eat Disord 2002; 32:98–102.
102. Agras WS, Crow SJ, Halmi KA, Mitchell JE, Wilson GT, Kraemer HC. Outcome
predictors for the cognitive behavior treatment of bulimia nervosa: data from a
multisite study. Am J Psychiatry 2000; 157:1302–1308.
103. Sohlberg S, Norring C, Holmgren S, Rosmark B. Impulsivity and long-term
prognosis of psychiatric patients with anorexia nervosa/bulimia nervosa. J Nerv
Ment Dis 1989; 177:249–258.
104. Hudson JI, Pope HG Jr, Jonas JM. Psychosis in anorexia nervosa and bulimia. Br J
Psychiatry 1984; 145:420–423.
105. Bruch H. Eating Disorders: Obesity, Anorexia Nervosa and the Person Within.
106. Hsu LKG, Meltzer ES, Crisp AH. Schizophrenia and anorexia nervosa. J Nerv
Ment Dis 1981; 169:273–276.
107. Salmon G. Atypical anorexia nervosa and schizophrenia: are they linked by the
5-HTlc receptor? Eur Eat Disord Rev 1996; 4:189–194.
108. Lavia MC, Gray N, Kaye WH. Case reports of olanzapine treatment of anorexia
109. Newman-Toker J. Risperidone in anorexia nervosa. J Am Acad Child Adol
Psychiatry 2000; 39:941–942.
110. Everill JT, Waller G, MacDonald W. Dissociation in bulimic and non-eating
disordered women. Int J Eat Disord 1995; 17:127–134.
111. Everill JT, Waller G. Dissociation and bulimia: research and theory. Eur Eat Dis
Rev 1995; 3:129–147.
112. Vanderlinden J, Vandereycken W, van Dyck R, Vertommen H. Dissociative
experiences and trauma in eating disorders. Int J Eat Disord 1993; 13:187–193.
113. Carlson EB, Putnam FW. An update on the dissociative experiences scale.
Dissociation 1993; 6:16–27.
114. Waller NG, Putnma FW, Carlson EB. Types of dissociation and dissociative types:
a taxometric analysis of dissociative experiences. Psychol Measures 1996; 1:
300–321.
115. van der Kolk BA, Pelcovitz D, Roth S, Mandel FS, McFarlane A, Herman JL.
Dissociation, somatization, and affect dysregulation: the complexity of adaptation
to trauma. Am J Psychiatry 1996; 153:83–93.
116. Shearer SL. Dissociative phenomena in women with borderline personality
disorder. Am J Psychiatry 1994; 151:1324–1328.
117. Brodsky BS, Cloitre M, Dulit RA. Relationship of dissociation to self mutilation
and childhood abuse in borderline personality disorder. Am J Psychiatry 1995; 152:
1788–1792.
118. Waller G, Ohanian V, Meyer C, Everill J, Rouse H. The utility of dimensional and
categorical approaches to understanding dissociation in the eating disorders. Br J
Clin Psychol 2001; 40:387–397.
119. Baumeister R, Heatherton T, Tice D. Losing Control: How and Why People Fail
at Self-Regulation. London: Academic Press, 1994.
120. Suyemoto KL. The functions of self-mutilation. Clin Psychol Rev 1998; 18:
531–554.
121. Nagata T, Kiriike N, Iketani T, Kawarada Y, Tanaka H. History of childhood
sexual or physical abuse in Japanese patients with eating disorders: relationship with
dissociation and impulsive behaviors. Psychol Med 1999; 29:935–942.
122. Nijenhuis ERS. Somatoform dissociation: major symptoms of dissociative
disorders. J Trauma Dissoc 2000; 1:7–32.
123. Zerbe KJ, Giorgio C. Whose body is it anyway? Understanding and treating
psychosomatic aspects of eating disorders. Rev Psychoanal Meth Res Clin Exp
1999; 6:13–32.
124. Fleming J, Levy L. Eating disorders in women with AD/HD. In: Quinn P, Nadeau
K, eds. Gender Issues and AD/HD: Research, Diagnosis, and Treatment. Silver
Spring, MD: Advantage Books, 2002.
125. Geist R, Davis R, Heinmaa M. Binge/purge symptoms and comorbidity in
adolescents with eating disorders. Can J Psychiatry 1998; 43:507–512.
126. Thompson KM, Wonderlich SA, Crosby RD, Mitchell JE. The neglected link
between eating disturbances and aggressive behavior in girls. J Am Acad Child
Adol Psychiatry 1999; 38:1277–1284.
208 LILENFELD
9
Personality Traits and Disorders Associated
with Anorexia Nervosa, Bulimia Nervosa, and
Howard Steiger and Kenneth R.Bruce
Douglas Hospital and McGill University
Montreal, Quebec, Canada
In 1689, Thomas Morton reported two cases of anorexia nervosa (AN): one
involving a “sad and anxious” girl who “poured over books,” another involving a
young boy prone to “studying too hard” (1). With these reports, Morton not
only introduced AN to the medical literature; he introduced the concept that
AN frequently co-occurs with perfectionistic or compulsive personality traits.
Likewise, 19th century observations on bulimia nervosa (BN) documented
instability of mood and behavior (2). All of this says that there is a history
behind the association between eating disorders (EDs) and problematic
personality tendencies.
The connection may not be new, but many questions remain: Is there an
ED-prone personality type? Do AN and BN coincide with different personality
characteristics? Do EDs always involve personality problems? In this chapter, we
study personality traits and disorders in AN, BN, and binge eating disorder
(BED). We examine the coaggregation of eating symptoms with personality
trait variations and evidence that such co-occurrence involves shared
developmental, familial, neurobiological, and hereditary pathways. We also
discuss the bearing of personality pathology, in the EDs, for clinical typology,
treatment needs, and prognosis.
EATING AND PERSONALITY PATHOLOGIES

Personality “traits” refer to enduring patterns of perception, thought, and
feeling. Traits are presumed to exist on a continuum (from absent to strongly
present) and are believed to structure feelings, actions, and reactions in
predictable ways across situations (3). For example, someone who regularly
worries about things going wrong might be said to display “trait anxiety.”
Personality “disorders” (PDs) also represent stable (or frequently recurring)
patterns of inner experience and behavior. However, they imply more pervasive
anomalies that adversely and profoundly affect thought, mood, behavior, and
210 STEIGER AND BRUCE
social functioning (3). PDs are conceived to be categorical (i.e., present or

absent) and to involve a systematic convergence of tendencies (e.g., mood
lability, impulsivity, and interpersonal instability).
Personality Traits in the EDs

The first formal psychometric studies on the EDs were published in the
mid1970s. These reports (on AN, at the time) led to emphasis on various
traits—obsessionality, social anxiety, introversion, “neuroticism,” and
depression. These same reports introduced a distinction between an ED subtype
characterized solely by restriction of food intake (the “restricters”) and variants
implicating binge eating and purging (the “bingers-purgers”). AN, restricting
subtype (AN-R) was associated with a “conforming, obsessional, and
emotionally and socially reserved” personality stereotype, whereas AN/
bingepurge subtype (AN-BP) was linked with a more dramatic
psychopathology, with an “impulsive, antisocial, or externalizing” flavor (4).
Later studies on BN broadened the boundaries of a hypothetical association
between binge eating and impulsive/erratic characteristics (5)—and the stage was
set for belief in a systematic distinction, as to associated personality
characteristics, between “restrictive” ED variants (i.e., AN-R) and “binger-
purger” variants (exemplified by BN or AN-BP).
Although fit proves to be imperfect, contemporary studies generally
corroborate the “restricter/binger difference” concept: With the
Multidimensional Personality Questionnaire, Casper and colleagues (6) found
greater self-control, conscientiousness, and emotional inhibition in AN-R than
in BN or AN-BP. Norman and colleagues (7), using the Millon Clinical
Multiaxial Inventory, found Avoidant personality traits to predominate in AN-R,
and Histrionic traits in BN or AN-BP. Partially “echoing” such tendencies,
studies applying the Tridimensional Personality Questionnaire (TPQ) have
reported normal-weight bulimics to display elevated novelty seeking relative to
restricter anorexics (8,9). However, results on TPQ dimensions of Harm
Avoidance and Reward Dependence have not lent themselves to simple
interpretation within a restricter/binger difference framework. Using a
comparable scale, the Temperament and Character Inventory (TCI), Diaz-
Marsa and colleagues (10) found anorexic patients to have stronger loadings on
a measure of persistence (largely “compulsive” personality) than bulimics,
whereas bulimics showed more impulsivity. Finally, bingers have been reported
to display more ideational disturbances and more anger than restricters on the
Minnesota Multiphasic Personality Inventory 2 (MMPI2) (11). The preceding
supports belief in a systematic coaggregation (at a group level) between AN-R
and perfectionistic, approval seeking, or compulsive traits, on the one hand, and
PERSONALITY TRAITS AND DISORDERS 211
between binger-purger ED variants (ANBP and BN) and impulsivity,

emotionality, or risk taking on the other. However, a caveat is needed:
“Average” traits in diagnostic subgroups risk obscuring meaningful within-
subtype heterogeneities. We address the issue of within-subtype heterogeneity
in a later section.
Specific Traits
A. Perfectionism. A hypothetical link between AN and perfectionism seems to be
well-supported empirically, with restricter and binger-purger anorexics both
producing higher self-rated perfectionism scores than healthy controls (12,13)
Recent findings suggest that a connection with perfectionism may extend to
BN (14) and to BED (15). However, any such link may be stronger for BN than
for BED (16). Perfectionism appears to exist premorbidly in ED sufferers (17)
and in AN sufferers after long-term weight recovery (13), and furthermore to
be common in relatives of those who develop an ED (14,18). While such
findings all support the speculation that perfectionism may be a risk factor for
ED development, specificity of the association with EDs (versus other forms of
maladjustment) still needs to be established.
B. Impulsivity. Findings suggest remarkable capacities for nonreflectiveness or
behavioral impulsivity in “bingers-purgers.” Self-report scales show that bulimic
individuals often fail to consider consequences of their actions and often act on
impulse (19). Furthermore, various studies report bulimics to show prominent
parasuicidality or self-mutilation: Mitchell and colleagues (20) estimated that
about 20% of their bulimic cohort showed repeated suicidal gestures, self-
mutilation, or other forms of self-destructive impulsivity. Likewise, Newton
and colleagues (21) found almost half of a cohort of normal-weight bulimics to
engage in combinations of substance abuse, drug overdoses, suicidal gestures, self-
mutilation, sexual disinhibition, or shoplifting. A similar report, based on
semistructured interviewing techniques, showed nearly 40% of bulimic
individuals to display multiple forms of destructive impulsivity (22). Combined
characteristics of suicidality, selfaggression, shoplifting, substance abuse, and
sexual promiscuity compose the features of what Hubert Lacey and his
colleagues (23) some years ago dubbed “multi-impulsive” bulimia—a subtype of
BN in which binge eating represented only one instance of pervasive
dysregulation of impulse controls. The general propensity toward impulsivity
may extend, it appears, to “binger” populations in general. For example, studies
have supported the idea that binge-purge AN has a more complicated clinical
course, with greater risk for suicide than restricting AN (24). Similarly, a study
of 495 eating-disordered outpatients by Favaro and Santonastaso (25) indicated
that 13% reported a past suicide attempt and 29% reported current suicidal
ideation, with suicidality being more prevalent among binge-purge anorexics

and purging bulimics than other ED subtypes. This study also linked suicidal
acts to a history of sexual abuse (a theme to which we return below). Extending
the putative link between binge eating and self-destructiveness even further,
Paul and his group (26) assessed self-injuriousness in women with AN, BN, or
BED, and found lifetime rates to be elevated in both BED and BN (at 35.8% and
34.3%, respectively).
The idea that binge eating syndromes are linked to behavioral impulsivity or
response inhibition may hold as a generalization. However, potentially
important heterogeneities emerge among “binger” patients (5,27–31)—with a
sizable proportion of cases found to display the perfectionistic or compulsive
tendencies that have been typically associated with restrictingtype AN. Such
variations have led some theorists to postulate the existence of distinct “binger”
subtypes, with distinct etiological paths (5,23,30,31). The idea here is that in
only some bulimics is binge eating driven by a pervasive “dysregulatory
pathology” (affecting mood and impulse regulation). In others, binge eating is
thought to have a more circumscribed etiology. We elaborate on this point in
“Within-Subtype Heterogeneity as to Personality Features” (to follow).
Personaiity Disorders in the EDs

Details on the link between eating and personality disorders are provided in
several excellent, comprehensive reviews (4,5,27–29). Consequently, in the
present context, we enumerate broad, empirically supported generalizations:
1. PDs are undoubtedly more common in anorexic and bulimic women than
in healthy comparison cases. Estimated PD prevalences in eating-
disordered samples vary from roughly 20% to nearly 100%, with most
commonly cited figures occupying the 50–75% range. Corroborating this
point, a recent meta-analytical review of findings from 28 available studies
finds 58% of women with an ED to have a PD, compared to 28% of
comparison women (29).
2. Restrictive AN is associated mainly with DSM Cluster C (“Anxious-
Fearful”) PD subtypes, typified by anxiousness, orderliness, and preference
for control. More rarely, AN-R coincides with Cluster A (“Odd-
Eccentric”) PDs (e.g., Schizoid PD)—having a particularly stilted quality,
but still characterized by inhibition and restricted emotionality. The PD
subtype reported to occur most frequently in AN-R is Obsessive-
Compulsive PD (with Avoidant and Dependent PDs also sometimes named
as modal diagnoses).
3. Binger-purger variants show greater heterogeneity as to comorbid PD

variations. If present, PDs will often occupy the DSM Cluster B
(“Dramatic-Erratic”) spectrum (i.e., Borderline, Histrionic, and
Narcissistic PDs), in which attention and sensation seeking, and mood
lability, excitability, and impulsivity are prominent. However, suggesting
that there is no simple dichotomy as to personality traits across restricter
and binger ED subtypes, and no overly direct connection between binge
eating potentials and “dysregulation” (as a trait), bingers-purgers are noted
to display PDs of the Cluster C (“Anxious-Fearful”) type about as
frequently as those of the Cluster B (“Dramatic-Erratic”) type.
Heterogeneous personality profiles associated with the binge-purge
syndromes have, again, been taken to imply that these syndromes may have
heterogeneous etiological paths (see “Within-Subtype Heterogeneity as to
Personality Features”).
4. Borderline personality disorder (BPD) is a particularly malignant Dramatic-
Erratic variant, characterized by massive dysregulation of affects and
impulse controls, chaotic interpersonal functioning, and repeated
parasuicidality. Modal figures suggest that BPD, or an apparent “borderline-
spectrum” personality (in which self-destructive impulsivity and
parasuicidality are prominent), may be present in up to a one-third of
bingers-purgers. Although rare in AN-R, BPD is apparently common in
AN-BP.
5. Patterns of PD comorbidity in BED resemble those obtained in BN:
Accumulating evidence suggests a general propensity towards personality
pathology, possibility of special affinity with Cluster B (“Dramatic-Erratic”)
PD subtypes (32), but nonnegligible presence of Cluster C (“Anxious-
Fearful”) PDs (33,34).
State-Trait Debate
Since malnutrition can adversely affect personality functioning (35), apparent
personality problems seen in ED sufferers could be attributed to “state” sequelae
of an active ED. Consistent with this, various findings show EDs to exacerbate
features with a “characterological coloring” (36,37). However, in counterpoint,
retrospective findings often suggest that personality problems predate ED onset
in ED sufferers. Råstam (38) estimated that 67% of a group of 51 adolescent
anorexics had a PD prior to ED onset, the most common of which (in 35% of
cases) was obsessive-compulsive PD. Similarly, Nagata and colleagues estimated
that parasuicidality predated onset of eating symptoms in 80% of ED patients
showing such self-destructiveness (39). Suggesting a similar independence of
personality pathology from ED sequelae, many findings indicate personality
problems to be relatively persistent in recovered (or recovering) ED sufferers.

One study reports 26% of recovered anorexics or bulimics to show an ongoing
PD more often a Dramatic-Erratic Cluster PD in the case of BN (40). Others
report persistence of obsessions, compulsions, social interaction problems (41),
and perfectionism (42) in AN, even years after weight recovery.
PUTATIVE SUBSTRATES FOR AN EATING- AND

PERSONALITY-PATHOLOGY “INTERFACE”
In the following sections, we explore psychosocial and psychobiological factors
that may contribute to convergence of eating and personality disturbances.
Family/Developmental Characteristics
Given systematic coaggregation of personality traits with ED subtypes, data
associating family interaction patterns with eating and personality problems
present various explicative potentials. On the personality side of this question,
some principled connections emerge: There have been inconsistent indications
of a link between obsessive-compulsive personality traits and familial
overinvolvement or overcontrol (43), and between impulsive (or borderline)
personality traits and parental neglect, hostility, or inconsistency (44).
In light of the preceding, data on family interaction patterns in ED subtypes
become suggestive. Studies on anorexic restricters and their relatives have often
corroborated concepts of enmeshment, overprotection, and emotional
constraint. In contrast, studies have associated binge-purge syndromes with
relatively more overt familial discord, disengagement, and hostility (31, 45).
Data on family interaction patterns in BED are inconclusive but suggest
comparable family interactions to those found in BN. For example, Hodges and
her colleagues (46) found BED women to describe their families as less cohesive,
expressive, and independent, and more conflictual and controlling, than did
normal-control women. Reiterating this notion, Tachi (47), working in Japan,
found women with BN and BED to perceive their families as being relatively
disengaged. Similarly, obese women with BED have been found to report more
paternal rejection than do obese women without BED (48).
Several research groups have noted severity of family dysfunction to
correspond much more closely to severity of personality pathology in
eatingdisordered individuals than to severity of eating symptoms (49,50). This
raises doubts about the extent to which family interaction patterns may be
specific to ED development—and the possibility that systematic convergences
between family interaction patterns and ED subtypes might reflect indirect
(personality-mediated) effects of family-environment variables acting on
eating-related-symptom development. For example, familial disturbances might

heighten explosiveness or impulsivity, and indirectly amplify impulse-driven ED
behaviors (like frequent vomiting).
Childhood Abuse
EDs, personality pathology, and especially BPD have all been associated with
childhood abuse or neglect. Studies suggest, for example, that more than 50% of
individuals suffering from BPD (51,52) and roughly 30% of those suffering from
an ED (53,54) have experienced childhood sexual abuse. Childhood abuse
experiences therefore represent one possible area of commonality underlying
convergence of eating and personality disturbances. However, assessment of
this possibility asks that various findings be properly considered:
1. Clinical studies have associated childhood abuse preferentially with bulimic

(rather than restrictive) forms of EDs (55,56). Corroborating this view,
survivors of childhood abuse report higher frequencies of BN than of
AN (57). In addition, studies in nonclinical, community samples suggest a
selective association between victimization experiences and bulimic
symptoms (58).
2. Studies in samples of clinical EDs generally show childhood abuse to covary
more directly with personality variables (impulsivity, dissociation,
interpersonal problems, etc.) than with eating-related symptom variables.
We cite a few illustrative examples: One study of 133 eating-disordered
women showed those reporting childhood abuse to display elevated
personality pathology but not more severe eating symptoms (58). Another
showed bulimics with a history of sexual or physical abuse to have more
posttraumatic symptoms and more substance dependence, but not more
eating symptoms (40). Several studies by our group are pertinent. One
showed a systematic association between childhood sexual or physical
abuse and likelihood of comorbid BPD, with no corresponding elevation in
eating-related symptom severity (59). A second showed childhood abuse
(sexual or physical) to predict more self-injuriousness, posttraumatic stress
disorders, and (nonsignificantly) more BPD in bulimic women (60). A
third indicated an association between severity of sexual and physical trauma
and severity of specific pathological personality characteristics, such as
submissiveness (61). Bearing closely on this question, Wonderlich and
colleagues documented findings suggesting that abuse may moderate eating-
related symptom severity indirectly, through exaggeration of tendencies
such as impulsivity (62).
Although it is premature to draw firm conclusions, preliminary observations

have suggested that sexual abuse may be less prevalent in BED than in BN. For
example, Yanovski and colleagues (32) did not find sexual abuse to be more
common in obese BED participants than in obese participants without BED.
Providing an apparent parallel, a telephone survey of 3006 women linked
history of forcible assault or rape to BN more than to BED or noneating-
disordered status (57).
Neurobiology of Eating and Personality

Disturbances
As a general rule, studies on the neurobiology of eating behavior indicate that
reducing brain serotonin (5-hydroxytryptamine, 5-HT) activity disinhibits
eating, whereas increasing 5-HT neurotransmission results in reduced
eating (63). In this light, it is intriguing to note that studies of the 5-HT
system in ED patients have associated BN and AN-BP with reduced activity of
the 5-HT system, while AN-restrictive subtype has been associated with the
converse (63,64). An obvious conjecture is that binge eating symptoms
correspond to reductions in 5-HT neurotransmission, whereas restrictive
symptoms implicate increases.
In light of the preceding, it is intriguing that findings indicate 5-HT status to
covary along with ED-relevant personality trait variations as follows: The
literature indicates “impulsive” populations in general (suicide completers, fire
setters, violent offenders, individuals with BPD) to often display decreased
5-HT activity (65). In contrast, anxious and compulsive populations, at least
inconsistently, display increased 5-HT tone. Such data have supported the
proposal that “impulsive” and “compulsive” traits may occupy opposite poles of a
continuum of 5-HT hypo- to hyperactivity (66). If so, neurobiological
correlates of personality traits—like impulsivity and compulsivity—could
underlie the association of bulimic ED variants with decreased 5-HT tone and of
restrictive ED variants with increased 5-HT. In turn, 5-HT mechanisms might
act as neurobiological “bridge” between impulsive personality traits and binge-
eating tendencies, and between compulsive traits and dietary overcontrol.
Several studies in ED sufferers are consistent with just such
personalitymediated variations of 5-HT activity. One series, in BN, links
impulsivity (or disinhibition) with alterations in 5-HT activity, showing
(a) reduced neuroendocrine response after administration of a 5-HT agonist in
self-reportedly hostile (67) or self-mutilative (68) bulimics; (b) lower platelet
monoamine oxiydase (MAO) concentrations (a proxy for level of 5-HT activity)
in bulimics reporting impulsivity or borderline traits (69); (c) lower platelet
paroxetine binding in bulimics reporting greater “nonplanning”
impulsivity (70). In addition, we have recently documented unusually high

baseline prolactin and elevated density of platelet 5-HT reuptake sites in highly
compulsive bulimics (71). Together, such findings suggest that 5-HT alterations
in the EDs might reflect different personality tendencies in different segments
of the eating-disordered population. In other words, there may be serotonergic
mediation of affinities observed between ED subtypes and personality trait
variations.
We add that we have recently found childhood abuse (in bulimics and, to a
lesser extent, in nonbulimics) to predict (a) blunted prolactin reponses to a
5-HT agonist and (b) abnormally low basal plasma cortisol concentrations (60).
Such findings associate childhood abuse with alterations in adult brain function of
a type that could (in theory) impact adversely on impulse controls, stress
responses, and satiety. In other words, abuse-linked neurobiological alterations
could contribute to ED vulnerability and to various aspects of ED
phenomenology.
Genetics
Findings on the molecular genetics of eating and personality disorders are in a
nascent state. Nonetheless, preliminary findings may help inform our study of
the convergence of ED subtypes and personality trait variations. In BN, data
have pointed to gene variations that may be simultaneously linked to binge
eating and impulsive potentials. For example, Nishiguchi and colleagues (72)
reported a link, in a heterogeneous anorexic-bulimic sample, between a
polymorphism in the 5-HT2A receptor gene and the presence of (a) disinhibited
eating and (b) traits (largely impulsivity). In a related vein, Devlin and
colleagues linked drive for thinness and obsessionality, in 196 patients with
AN-restrictive subtype, to specific gene loci (73). Such findings raise the
intriguing specter that genetic factors may be substrates of a link between
bulimic eating tendencies and an “impulsive/behaviorally dysregulated”
phenotype—and between restrictive eating patterns and compulsive personality
characteristics.
CLINICAL IMPLICATIONS OF PERSONALITY

PATHOLOGY
Symptoms
Intuitively, personality pathology might be though to confer risk for more
severe eating symptomatology. However, empirical treatments provide limited
support for this idea. (Here we see the first of several instances of apparent
independence between eating and personality pathologies.) Johnson and his
colleagues (74) compared patients with and without self-reported BPD features.
Borderlines differed from nonborderlines in showing greater depression, self-
mutilation, drug abuse, and history of sexual abuse, but not greater binge/purge
frequencies or other eating-specific symptoms. In a similar study, Wonderlich
and Swift (75) examined clinical profiles in 75 eating-disordered women,
divided (according to structured interviews) into those with and without BPD.
Borderline patients displayed more suicidality, depression, anxiety, and self-
mutilation than did their nonborderline counterparts. However, except for
more frequent vomiting, borderline patients displayed negligibly more severe
eating symptoms. Our group conducted a similar multidimensional evaluation,
rating eating and general psychopathological symptoms in a sample of 91
bulimics, in whom axis II diagnoses were established with the aid of structured
interviews (76). As in the study of Wonderlich and Swift, borderlines evinced
more pronounced anomalies on diverse measures of general psychopathology
without showing more severe eating symptoms. Thus, while it is undeniably
associated with greater generalized psychopathology in BN sufferers, PD may
not have a marked connotation for severity of ED symptoms. In apparent
contrast, personality pathology has been associated with more severe binge
eating in BED (33,77).
Outcome
Comorbid personality pathology has been associated with poorer global
outcome from AN, BN, and BED.
Anorexia Nervosa
Following an extensive review (encompassing 119 studies and 5590 patients),
Steinhausen (78) concluded that obsessive-compulsive PD is associated with an
unfavorable prognosis in AN. Intriguingly, he noted hysterical personality
features (somewhat the antithesis of obsessive-compulsive features) to coincide
with favorable prognosis. Several results suggest the prognostic utility in AN of
other personality measures. A recent study reported response in 42 restricter
anorexics after 180 days of multimodal therapy (79). Categorical PD diagnoses
did not predict improvement, but dimensional measure (low novelty seeking,
and high asceticism and maturity fears) were systematically linked to poorer
response. Suggesting similar effects at long-term follow-up, Bulik and her
colleagues (36) found anorexics who remained ill after 12 years to be
characterized partly by more severe eating disturbances and partly by
greater harm avoidance (i.e., anxious avoidance and overcontrol) and lower
self-directedness. In a similar 4-year prospective study for AN, lower selfesteem
and stronger maturity fears were associated with poorer outcome (80).
Bulimia Nervosa
Results for BN also seem to associate personality pathology with poorer global
treatment response: Herzog et al. (81) linked comorbid PD (diagnosed using
structured interview in 179 bulimics) with lower probability of abstinence from
BN after 9 or more months. Similarly, Rossiter et al. (82) reported that Dramatic-
Erratic personality traits predicted greater chance of persistent vomiting in BN
after one year. Other studies corroborate the same trend: One shows BN patients
with high initial Borderline Syndrome Index scores to be more likely to remain
bulimic after one year than patients with initially lower scores (83). Another
reports that bulimic patients who showed persistent “borderline features” over
the first 3 months of therapy showed poorest response of bulimic symptoms
after 6 and 12 months (84).
When finer grained outcome dimensions (rather than “global ED outcome”)
are assayed, however, subtler effects sometimes emerge. Wonderlich et al. (85)
reported that PD comorbidity predicted poorer global response, but only weak
differences on indices of bulimic symptoms, in 30 bulimics followed for
4–5 years. Our group studied progress in treatment, in function of interview-
based PD diagnoses, of bulimics after 3, 6, and then 12 and more months.
Borderline bulimics showed greater psychiatric symptoms than did
nonborderline bulimics at all points in time but limited differences on measures
reflecting the course of eating symptoms (76,86). Norring (87) reported similar
findings in heterogeneous ED cases. Patients showing a “borderline
organization” showed poorest response after 2 and 3 years—but differences
were more pronounced on comorbid symptoms than on eatingspecific ones.
Such tendencies led Grilo (27), after an excellent and comprehensive review, to
conclude that PD may be more closely associated with the longtudinal course of
general psychiatric symptoms or psychosocial functioning in ED patients than
with fluctuations in the course of eating symptoms.
It is possible that “impulsivity” measures may yield more reliable prognostic
effects. After an extensive literature review, Keel and Mitchell (88) concluded
that impulsivity was foremost among prognostic indices for BN, with PD having
inconsistent effects. In addition, isolated personality dimensions have emerged as
having prognostic value for BN: Joiner et al. (89) found “maturity fears” and
“perfectionism” to be weak long-term (10-year) predictors of bulimic
symptoms. Bulik and her colleagues (90) examined predictors of outcome one
year after completion of a randomized psychotherapy trial. PD symptoms were
not predictive, but a measure reflecting selfdirectedness was associated with

lower postreatment frequency of bingeing and purging.

Available studies have suggested that PD may also predict poorer outcome for
BED. For example, Wilfley and colleagues (33) found that presence of a
Dramatic-Erratic PD predicted significantly more binge eating in BED patients
at one year after treatment. Likewise, Stice and colleagues documented poorer
treatment response in BED patients with a concurrent PD (77).
Treatment Usage
Personality pathology apparently has an additional clinical relevance as a
predictor of treatment usage: Several studies document more frequent
hospitalizations and medication usage in bulimics with Dramatic-Erratic or
borderline PDs than in those with lesser PD comorbidity (74,75,86). Suggesting
that this trend cuts across ED subtypes, Keel and colleagues (91) reported a link,
in 294 women treated for AN or BN, between presence of a PD and use of
psychotherapy and medication after 5 years.
WITHIN-SUBTYPE HETEROGENEITY AS TO
PERSONALITY FEATURES
The patterning of association between personality and eating symptom
variations has inspired various speculations on the “shaping” role, in ED
expression, of personality factors. Relatively circumscribed comorbidity of a
compulsive type in restrictive AN makes an obvious appeal to the idea that
obstinacy, constriction, and hypersensitivity to social approval may “power”
rigid dieting and driven pursuit of thinness. Likewise, affinities between bulimic
behaviors and tendencies toward behavioral dyscontrol encourages the view that
bingeing and purging may only be isolated instances, in affected individuals, of a
more generalized dysregulation. Although intuitively appealing, the latter view
is challenged by the reality that individuals who binge-eat display heterogeneous
personality characteristics—sometimes impulsive, sometimes compulsive.
Diversity of personality characteristics seen in bulimic syndromes has (as
discussed earlier) been thought to correspond to different etiological
paths to binge eating-sometimes reflecting a primary and rather pervasive
disruption of controls over mood, impulses, and appetite, and sometimes a
more circumscribed erosion of appetitive controls, following prolonged
dieting (5, 31,92,93). The idea that binge eating results from prolonged dietary
restraint is established clinically and empirically, and fundamental to Polivy and

Herman’s (94) well-known restraint theory. However, recent studies reveal a
subgroup of BN and BED sufferers in whom binge eating seems not to follow
from antecedent dieting (95)—and to whom restraint theory seems to be less
applicable. This “binge-first” subgroup is furthermore reported to show more
marked personality pathology than do more traditional “diet-first” bulimics—
which supports the conjecture that personality pathology might (sometimes)
directly enhance susceptibility to binge eating (via emotion- or impulse-linked
pathways). Recent data from our group resonate with this idea. We had
51 bulimic women record ongoing emotions and behaviors over a roughly
3-week span (96). Findings suggested that restraint may have been a weaker
hour-to hour binge antecedent in impulsive than in nonimpulsive bulimics—or
that impulsivity may have been intrinsically conducive to binge eating in the
relative absence of antecedent dietary restraint.
Other findings are similarly relevant to the notion that personality trait
variations might help define meaningful ED subtype distinctions: Studies
applying cluster-analytical techniques to personality indices obtained in eating-
disordered samples indicate that there may be something quite repeatable about
a tripartite classification of ED patients into the following subgroups:
(a) “psychologically relatively intact”, (b) “overregulated (compul-sive)” and
(c) “dysregulated (impulsive)” (93). Furthermore, these findings link AN-
restrictive subtype consistently with the overcontrolled personality profile, but
associate bulimic ED variants about equiprobably with any of the three
personality profiles. Assuming that this is so, we find it intriguing to
contemplate the ways in which such distinctions may correspond to data on
neurobiological variations and developmental typologies. For example, the
spectrum from over- to underregulated personality characteristics could, in
theory, span a continuum from hyperactivity to hypoactivity of serotonin
mechanisms. Similarly, the overregulated subtype could correspond to familial
overinvolvement (with low risk of childhood trauma), whereas the dysregulated
subtype might be linked to familial incohesion (and greater risk of childhood
abuse).
INTEGRATION AND CLINICAL APPLICATIONS

Findings in the EDs converge on the idea that personality variables correspond
to meaningful variations as to developmental, familial, and neurobiological
typologies. Furthermore, findings suggest that personality factors may be
relevant to global functioning, treatment outcome, and prognosis in ED
sufferers. These trends suggest that personality-linked considerations may have
an important contribution to make in clinical formulation and clinical decision
making in this area. To illustrate how this may be so, we offer a few general
conceptual and clinical guidelines:
(1) Eating and personality disturbances are separate but interdependent entities.
There are reasons to assume that eating and personality pathologies are
independently determined. For example, one can be severe, whereas the other
is mild; one can resolve, whereas the other persists. This implies, as suggested
by available outcome literature, that eating problems can to some extent be
treated independently of associated personality characteristics. However, that
eating and personality disturbances co-occur so frequently also suggests
nonnegligible measures of causal overlap. Consider the following:
Neurobiological vulnerabilities (e.g., altered serotonin neurotransmission) may,
along with predictable effects on personality manifestations (e.g., heightened
impulsivity), confer vulnerability to impaired satiety. If so, having a particular
personality trait (like impulsivity) might increase vulnerability to a particular
disturbance in eating behavior (like binge eating) because of neurobiological
propensities that are linked to both the trait and the appetitive behavior. Given
this notion, it becomes fascinating to contemplate all of the possible sources of
influence that may act upon the neurobiological status in a given individual at a
given moment in time (e.g., heredity, nutritional state, effects of
developmental and current stressors, etc.)—and, in turn, to contemplate the
complexity of processes that determine whether or not a given personality trait
(e.g., impulsivity) or eating symptom (e.g., binge eating) is likely to come to
expression. We address such convergences elsewhere (31,60,68,70). In a
related vein, common developmental factors (e.g., childhood sexual abuse) can
be thought to impact, not only upon interpersonal expectations and
functioning, but also upon the sense of bodily integrity, control, and esteem.
Such effects might, in turn, heighten susceptibility to concurrent interpersonal
difficulties and maladaptive weight control practices or food avoidances.
(2) Eating disturbances worsen personality disturbances. The view that ED sequelae
exacerbate obsessionality, impulsivity, or other apparent personality problems
in ED sufferers has a simple wisdom. Evidence indicates that EDs can
exaggerate personality problems, and this certainly explains recent practice
preferences leaning toward symptom-focused models of therapy for the EDs. A
rationale (aside from the obvious aim of reducing presenting symptoms) is that
improving nutritional status in an ED sufferer often improves her personality
and general functioning. Consequently, we urge that it is never appropriate to
ignore eating disturbances and focus on underlying personality issues alone. At
the same time, primacy of personality disturbances, and persistence following
recovery, suggests that it is untenable to regard personality factors solely as ED
sequelae and hence (swinging to the other extreme) to ignore them in
treatment. In a closely related vein:
(3) Eating and personality problems have mutual “pathoplastic effects”

(30)—pathology ofone sort “shaping” variations in the other. For example, impulse
control problems may exacerbate intensive vomiting, self-image disturbances
may amplify body dissatisfaction, and general compulsivity may support
compulsive dieting. This means that work aimed at generalized behavioral or
self-image problems can often have “spinoff” benefits, as far as resolution of
eating symptoms is concerned. As patients learn to manage anxiety and
impulsive responses, they become less reliant on binge episodes to sooth eating-
induced anxiety. As they learn to be less angry, they become less reliant on
binge episodes as a means of diffusing angry affects. As they learn to process
social experiences more adaptively, they become less prone to injury in
interpersonal experiences and to recourse to eating symptoms as a means of
regulating negative “self’ experiences (97).
We introduce a fourth point, not yet treated in our review but one that (we
believe) is essential to any adequate discussion of the clinical implications of
personality pathology in the EDs.
(4) Personality pathology has powerful interpersonal repercussions. PDs impact on
all relationship experiences, including therapeutic ones. For this reason, we
advocate close attention to relational aspects in therapy of ED patients. We
propose that you cannot manage eating disturbances without caring for the
relationship experience—not only because it is the vehicle of alliance, but also,
because (we believe) it is a direct ingredient of change. To illustrate, we bring
in a concept proposed by clinician-theorists who concern themselves with the
impact of therapeutic alliance upon therapy outcome (98). The concept holds
that when patients have dysfunctional beliefs and expectancies about
interpersonal experiences (a given with comorbid personality pathology), they
can invite dysfunctional behaviors from others—including their therapists. For
example, the hostile patient (depending on the therapist’s own proclivities) can
evoke complementary hostility (or defensive withdrawal); the regressive patient
can evoke excessive nurturance (or unrealistic demands for autonomous
functioning); the grandiose patient can evoke competitiveness (or adoration);
and the compulsive patient can stimulate struggles around power and control
(or excessive acquiescence and gentleness). It is an obvious point that, to be
therapeutic, therapists need not “complement” patients’ dysfunctional
interpersonal responses (in the ways described), but should evoke healthy
interpersonal experiences that disconfirm dysfunctional relationship
expectancies.
CONCLUSIONS
Our thinking in this chapter has been structured around the concept that eating
and personality pathologies have multiple and (often) shared biological,
psychological, and social determinants. This means that personality cannot be
conceptualized as a unidimensional psychological aspect of what occurs in
people with EDs—but rather, needs to be understood as an integral part of the
ED syndrome, tightly woven into its complex of causes and symptoms, and
relevant as both an expression of, and influence on, the ED’s biopsychosocial
“mix.”
This perspective asks that we revise our thinking about psychopathological
syndromes and traits, especially as far as the idea of separating “syndromic” and
“personality” components is concerned. It may be that we can never fully
segregate “having an eating disorder” (in either phenomenological or etiological
respects) from “concurrent personality tendencies.” Rather, “personality”
phenomena (like impulsivity, compulsivity, or perfectionism) may be at least as
relevant to defining ED syndrome “variants” as any one of a number of
superficially more relevant aspects of ED phenom enology (e.g., presence of
binge eating or laxative abuse). Indeed, available data argue that the personality
variations may be the stronger predictors of clinical phenomenology,
neurobiological substrates, sexual abuse history, treatment outcome, prognosis,
and other aspects. This argues for a more central attention to personality trait
variations in the development of ED diagnostic classifications—a view that has
been advocated by several others (30,31,93).
The same concept is relevant to treatment considerations. To date, ED
specialists have succeeded in standardizing and formalizing interventions aimed
at eating disturbances. A next step, we would urge, is to be equally rigorous in
our efforts to develop, integrate, and perfect interventions aimed at personality
components in the EDs (and their inextricable consequences for the therapeutic
alliance). We are thinking here not of an “adjunctive” use of personality
concepts—treating personality features as some sort of “grease on the wheels” of
eating-focused interventions—but of a central attention to the ways in which
personality may shape expression of eating symptoms, the relational aspects of
the therapeutic experience and, in turn, the process of change.
Various theorists have argued that eating symptoms need to be contextualized
in light of comorbid personality characteristics (5,30,31,93). For example, a
symptom like vomiting may have quite a different meaning when it arises in
(a) the emotionally constricted and weight gain-phobic patient, or (b) the highly
dysregulated or posttraumatic patient. In the former, vomiting may serve as a
phobic avoidance, driven by relatively circumscribed weight gain fears-in the
latter, as a desparate response to felt emptiness, posttraumatic memories, or
overwhelming rage. Until equipped to process such experiences adaptively,

encouraging the very highly dysregulated patients to defer vomiting seems
barely adequate and risks “missing the mark.” In other words, therapeutic
prescriptions need to be tailored to the personality context. In summary, if we
hope to understand the EDs’ phenomenological complexity, we believe that it is
heuristic to minimize the conceptual distance between eating-specific aspects of
phenomenology and more generalized aspects usually thought to define
“personality.” This, we believe, is because eating-specific and personality-linked
components of the ED may become quite inextricably intermingled once the
eating syndrome becomes instated.
REFERENCES
1. Gordon RA. Anatomy of a social epidemic. Oxford: Basil Blackwell, 1990:12.
2. Ziolko H-U. Bulimia: a historical outline. Int J Eat Disord 1996; 20:345–358.
Disorders 4th ed. Washington, DC: DSM-IV, 1994.
4. Sohlberg S, Strober M. Personality in anorexia nervosa: an update and a theoretical
integration. Acta Psychiatr Scand 1994; 89(Suppl 378): 1–15.
5. Vitousek K, Manke F. Personality variables and disorders in anorexia nervosa and
bulimia nervosa. J Abnorm Psychol 1994; 103:137–147.
6. Casper RC, Hedeker D, McClough JF. Personality dimensions in eating disorders
and their relevance for subtyping. J Am Acad Child Adol Psychiatry 1992; 31:
830–840.
7. Norman DK, Blais D, Herzog D. Personality characteristics of eating-disordered
patients as identified by the Millon Clinical Multiaxial inventory. J Pers Disord
1993; 7:1–9.
8. Bulik CM, Sullivan PF, Weltzin TF, Kaye WH. Temperament in eating disorders.
Int J Eat Disord 1995; 17:251–261.
9. Brewerton TD, Hand JD, Bishop EM. The Tridimensional Personality
Questionnaire in patients with eating disorders. Int J Eat Disord 1993; 14:213–218.
10. Diaz-Marsa M, Carrasco JL, Saiz J. A study of temperament and personality in
anorexia and bulimia nervosa. J Pers Disord 2000; 14:352–359.
11. Cumella EJ, Wall AD, Kerr-Almeida N. MMPI-2 in the inpatient assessment of
women with eating disorders. J Pers Assess 2000; 75:387–403.
12. Halmi KA, Sunday SR, Strober M, Kaplan A, Woodside DB, Fichter M, Treaure J,
Berrettini W, Kaye WH. Perfectionsim in anorexia nervosa: variation by clinical
subtype, obsessionality, and pathological eating behavior. Am J Psychiatry 2000;
157:1799–1805.
13. Srinivasagam NM, Kaye WH, Plotnicov KH, Greeno C, Weltzin TE, Rao R.
Persistent perfectionism, symmetry, and exactness after long-term recovery from
anorexia nervosa. Am J Psychiatry 1995; 152:1630–1634.
14. Lilenfeld LR, Stein D, Bulik CM, Strober M, Plotnicov K, Pollice C, Rao R,
Merikangas KR, Nagy L, Kaye WH. Personality traits among currently eating
disordered, recovered and never ill first-degree female relatives of bulimic and
control women. Psychol Med 2000; 30:1399–1410.
15. Pratt EM, Telch CF, Labouvie EW, Wilson GT, Agras WS. Perfectionism in
women with binge eating disorder. Int J Eat Disord 2001; 29:177–186.
for binge eating disorders: a community-based study. Arch Gen Psychiatry 1998;
55:425–432.
17. Fairburn CG, Cooper Z, Doll HA, Welch SL. Risk factors for anorexia nervosa:
three integrated case-controlled comparisons. Arch Gen Psychiatry 1999; 56:
468–476.
18. Woodside DB, Bulik CM, Halmi KA, Fichter MM, Kaplan A, Berrettini WH,
Strober M, Treasure J, Lilenfeld L, Klump K, Kaye WH. Personality,
perfectionism, and attitudes toward eating in parents of individuals with eating
19. Wolfe BE, Jimerson DC, Levine JM. Impulsivity ratings in bulimia nervosa:
relationship to binge eating behaviours. Int J Eat Disord 1994; 15:289–292.
20. Mitchell JE, Hatsukami D, Pyle RL, Eckert ED. The bulimia syndrome: course of
the illness and associated problems. Comp Psychiatry 1986; 27:165–170.
21. Newton JR, Freeman CP, Munro J. Impulsivity and dyscontrol in bulimia nervosa:
is impulsivity an independent phenomenon or a marker of severity? Acta Psychiatr
Scand 1993; 87:389–394.
22. Crosby RD, Wonderlich SA, Redlin J, Engel SG, Simonich H, Jones-Paxton M,
Myers T, Norton M, Thompson KM, Mitchell JE. Impulsive behavior patterns in a
sample of females with bulimia nervosa. Poster presented at the annual meeting of
the Eating Disorders Research Society, Bernalillo, New Mexico, 2001.
23. Lacey JH, Evans CD. The impulsivist: a multi-impulsive personality disorder. Br J
Addict 1986; 81:641–649.
24. Herzog DB, Dorer DJ, Keel PK, Selwyn SE, Ekeblad ER, Flores AT, Greenwood
DN, Burwell RA, Keller MB. Recovery and relapse in anorexia and bulimia
nervosa: a 7.5-year follow-up study. J Am Acad Child Adol Psychiatry 1999; 38:
829–837.
25. Favaro A, Santonastaso P. Suicidality in eating disorders: clinical and psychological
correlates. Acta Psychiatr Scand 1997; 95:508–514.
26. Paul T, Schroeter K, Dahme B, Nutzinger DO. Self-injurious behavior in women
27. Grilo CM. Recent research of relationships among eating disorders and personality
disorders. Curr Psychiatry Rep 2002; 4:18–24.
28. Johnson C, Wonderlich S. Personality characteristics as a risk factor in the
development of eating disorders. In: Crowther JH, Tennenbaum DL, Hobfell SE,
Stephens MAP, eds. The Etiology of Bulimia Nervosa: The Individual and Family
Context. New York: Hemisphere Publishing, 1992.
29. Rosenvinge JH, Martinussen M, Ostensen E. The comorbidity of eating disorders

and personality disorders: a meta-analytic review of studies published between
1983 and 1998. Eat Weight Disord 2000; 5:52–61.
30. Wonderlich SA, Mitchell JE. Eating disorders and comorbidity: empirical,
conceptual and clinical implications. Psychopharmacol Bull 1997; 33:381–390.
31. Steiger H, Bruce K, Israel M. Eating Disorders. In: Stricker G, Widiger TA,
Wiener IB, eds. Comprehensive Handbook of Psychology. New York: John Wiley
and Sons, 2003:173–194.
32. Yanovski SZ, Nelson JE, Dubbert BK, Spitzer RL. Association of binge eating
disorder and psychiatric comorbidity in obese subjects. Am J Psychiatry 1993; 150:
1472–1479.
33. Wilfley DE, Friedman MA, Dounchis JZ, Stein RI, Welch RR, Ball SA. Comorbid
psychopathology in binge eating disorder: relation to eating disorder severity at
baseline and following treatment. J Consult Clin Psychol 2000; 68:641–649.
34. Grilo CM, Masheb RM. DSM-IV psychiatric comorbidity in binge eating disorder:
relation to gender and eating disorder symptomatology. Paper presented at the
annual meeting of the Association for the Advancement of Behaviour Therapy.
Philadelphia, November 2001.
35. Keys A, Brozek J, Henschel A, Mickelson O, Taylor H. The Biology of Human
36. Bulik CM, Sullivan PF, Fear JL, Pickering A. Outcome of anorexia nervosa: eating
attitudes, personality, and parental bonding. Int J Eat Disord 2000; 28:139–147.
37. Garner DM, Olmsted R, Davis R, Rockect W, Goldbloom D, Eagle M. The
association between bulimic symptoms and reported psychopathology. Int J Eat
Disord 1990; 9:1–16.
38. Råstam M. Anorexia nervosa in 51 swedish adolescents: Premorbid problems and
39. Nagata T, Kawarada Y, Kiriike N, Iketani T. Multi-impulsivity of Japanese patients
with eating disorders: primary and secondary impulsivity. Psychiatry Res 2000; 94:
239–250.
40. Matsunaga H, Kaye WH, McConaha C, Plotnicov K, Pollice C, Rao R, Stein D.
Psychopathological characteristics of recovered bulimics who have a history of
physical or sexual abuse. J Nerv Ment Dis 1999; 187:472–477.
41. Nilsson EW, Gillberg C, Gillberg IC, Rastam M. Ten-year follow-up of adolescent
onset anorexia nervosa: Personality disorders. J Am Acad Child Adol Psychiatry
1999; 38:1389–1395.
42. Pla C, Toro J. Anorexia nervosa in a Spanish adolescent sample: an 8-year
longitudinal study. Acta Psychiatr Scand 1999; 100:441–446.
43. Nordahl HM, Stiles TC. Perceptions of parental bonding in patients with various
personality disorders, lifetime depressive disorders, and healthy controls. J Pers
Disord 1997; 11:391–402.
44. Nickell AD, Waudby CJ, Trull TJ. Attachment, parental bonding and borderline
personality disorder features in young adults. J Pers Disord 2002; 16:148–159.
45. Wonderlich SA. Relationship of family and personality factors in bulimia. In:
Crowther JH, Tennenbaum DL, Hobfell SE, Stephens MAP, eds. The Etiology of
Bulimia Nervosa: The Individual and Family Context. New York: Hemisphere
Publishing, 1992.
46. Hodges EL, Cochrane CE, Brewerton TD. Family characteristics of binge-eating
disorder patients. Int J Eat Disord 1998; 23:145–151.
47. Tachi T. Family environment in eating disorders: a study of the familiar factors
influencing the onset and course of eating disorders. Seishin Shinkeigaku Zasshi
1999; 101:427–445.
48. Dominy NL, Johnson W-B, Koch C. Perception of parental acceptance in women
with binge-eating disorder. J Psychol 2000; 134:23–36.
49. Schmidt U, Humfress H, Treasure J. The role of general family environment and
sexual and physical abuse in the origins of eating disorders. Eur Eat Disord Rev
1997; 5:184–207.
50. Steiger H, Liquornik K, Chapman J, Hussain N. Personality and family
disturbances in eating-disorder patients: comparison of “restricters” and “bingers”
to normal controls. Int J Eat Disord 1991; 10:501–512.
51. Herman JL, Perry C, van der Kolk BA. Childhood trauma in borderline personality
disorder. Am J Psychiatry 1989; 146:490–495.
52. Paris J, Zweig-Frank H, Guzder J. Psychological risk factors for borderline
personality disorder in female patients. Compr Psyhiatry 1994; 35:301–305.
53. Everill J, Waller G. Reported sexual abuse and eating psychopathology: a review of
evidence for a causal link. Int J Eat Disord 1995; 18:1–11.
54. Wonderlich SA, Brewerton TD, Jocic Z, Dansky B, Abbott DW. Relationship
of childhood sexual abuse and eating disorders. J Am Acad Child Adolesc
Psychiatry 1997; 36:1107–1115.
55. Steiger H, Zanko M. Sexual traumata in eating-disordered, psychiatric and normal
female groups: comparison of prevalences and defense styles. J Interpers Violence
1990; 5:74–86.
56. Bushnell JA, Wells JE, Oakley-Brown MA. Long term effects of intrafamilial
sexual abuse in childhood. Acta Psychiatr Scand 1992; 85:36–142.
57. Dansky BS, Brewerton TD, Kilpatrick DG, O’Neil PM. The National Women’s
Study: relationship of victimization and posttraumatic stress disorder to bulimia
58. Moreno JK, Selby MJ, Neal S. Psychopathology in sexually abused and nonsexually
abused eating-disordered women. Psychother Private Pract 1998; 17:1–9.
59. Steiger H, Jabalpurwala S, Champagne J. Axis-II comorbidity and developmental
adversity in bulimia nervosa. J Nerv Ment Dis 1996; 184:555–560.
60. Steiger H, Gauvin L, Israel M, Koerner N, Ng Ying Kin NMK, Paris J, Young SN.
61. Leonard S, Steiger H, Kao A, Childhood and adulthood abuse in bulimic and
nonbulimic women: prevalence and psychological correlates. Int J Eat Disord
2003; 33:397–405.
62. Wonderlich SA, Crosby R, Mitchell J, Thompson K, Redlin J, Demuth G, Smyth

J, Haseltine B. Eating disturbance and sexual trauma in childhood and adulthood.
Int J Eat Disord 2001; 30:401–412.
63. Brewerton TD. Toward a unified theory of serotonin dysregulation in eating and
related disorders. Psychoneuroendocrinology 1995; 20:561–590.
64. Kaye WH, Ebert MH, Gwirtsman HE, Weiss SR. Differences in brain serotonergic
metabolism between bulimic and nonbulimic patients with anorexia nervosa. Am J
Psychiatry 1984; 141:1598–1601.
65. Asberg M, Schalling D, Träskman-Bendz L, Wägner A. Psychobiology of suicide,
impulsivity and related phenomena. In: Meltzer HY, ed. Psychopharmacology:
Third Generation of Progress. New York: Raven Press, 1987:655–688.
66. Cloninger CR, Svrakic DM, Przybeck TR. A psychobiological model of
temperament and character. Arch Gen Psychiatry 1993; 50:975–990.
67. Waller DA, Sheinberg AL, Gullion C, Moeller FG, Cannon DS, Petty F, Hardy
BW, Orsulak P, Rush A. Impulsivity and neuroendocrine response to buspirone in
bulimia nervosa. Biol Psychiatry 1996; 39:371–374.
68. Steiger H, Koerner NM, Enghleberg M, Israel M, Ng Ying Kin NMK, Young SN.
Self-destructiveness and serotonin function in bulimia nervosa. Psychiatry
Res2001; 103:15–26.
69. Carrasco JL, Diaz-Marsa M, Hollander E, Cesar J, Saiz-Ruiz J. Decreased platelet
monoamine oxidase activity in female bulimia nervosa. Eur Neuropsychopharmacol
2000; 10:113–117.
70. Steiger H, Young SN, Kin NM, Koerner N, Israel M, Lageix P, Paris J.
Implications of impulsive and affective symptoms for serotonin function in bulimia
nervosa. Psychol Med 2001; 31:85–95.
71. Steiger H, Gauvin L, Israel M, Ng Ying Kin NMK, Young S, Walsh S, Hudon N.
Serotonin indices predict personality traits and likelihood of sexual abuse in bulimia
nervosa. Paper presented at the 2002 Annual Meeting of the Eating Disorders
Research Society, Charleston, South Carolina, Nov. 21–23, 2002.
72. Nishiguchi N, Matsushita S, Suzuki K, Murayama M, Shirakawa O, Higuchi S.
eating disorders in Japanese patients. Biol Psychiatry 2001 Jul 15; 50:123–128.
73. Devlin B, Bacanu SA, Klump KL, Bulik CM, Fichter MM, Halmi KA, Kaplan AS,
Strober M, Treasure J, Woodside DB, Berrettini WH, Kaye WH. Linkage analysis
of anorexia nervosa incorporating behavioral covariates. Hum Mol Genet2002; 11:
689–696.
74. Johnson C, Tobin D, Enright A. Prevalence and clinical characteristics of
borderline patients in an eating-disordered population. J Clin Psychiatry 1989; 50:
9–15.
75. Wonderlich SA, Swift WJ. Borderline versus other personality disorders in the
eating disorders clinical description. Int J Eat Disord 1990; 9:629–638.
76. Steiger H, Thibaudeau J, Leung F, Houle L, Ghadirian AM. Eating and psychiatric
symptoms as a function of axis II comorbidity in bulimic patients: threemonth and
six-month responses after therapy. Psychosomatics 1994; 35:41–49.
77. Stice E, Agras WS, Telch CF, Halmi KA, Mitchell JE, Wilson T. Subtyping binge
eating-disordered women along dieting and negative affect dimensions. Int J Eat
Disord 2001; 30:11–27.
78. Steinhausen HC. The outcome of anorexia nervosa in the 20th century. Am J
Psychiatry 2002; 159:1284–1293.
79. Fassino S, Abbate Dagga G, Amianto F, Leombruni P, Fornas B, Garzaro L,
D’Ambrosio G, Rovera GG. Outcome predictors in anorexic patients after 6
months of multimodal treatment. Psychother Psychosom 2001; 70:201–208.
80. Van der Ham T, van Strien DC, van England H. Personality characteristics predict
outcome of eating disorders in adolescents: a 4-year prospective study. Eur Child
81. Herzog DB, Keller MB, Sacks NR, Yeh CJ, Lavori PW. Psychiatric comorbidity in
treatment-seeking anorexics and bulimics. J Am Acad Child Adol Psychiatry 1992;
31:810–818.
82. Rossiter EM, Agras WS, Telch CF, Schneider JA. Cluster B personality disorder
characteristics predict outcome in the treatment of bulimia nervosa. Int J Eat
Disord 1993; 13:349–357.
83. Johnson C, Tobin DL, Dennis A. Differences in treatment outcome between
borderline and nonborderline bulimics at one-year follow-up. Int J Eat Disord
1990; 9:617–627.
84. Steiger H, Stotland S, Houle L. Prognostic implications of stable versus transient
“borderline features” in bulimic patients. J Clin Psychiatry 1994; 55:206–214.
85. Wonderlich SA, Fullerton D, Swift WJ, Klein MH. Five-year outcome from eating
disorders: Relevance of personality disorders. Int J Eat Disord 1994; 15:233–244.
86. Steiger H, Stotland S. Prospective study of outcome in bulimics as a function
of axis-II comorbidity: long-term responses on eating and psychiatric symptoms.
Int J Eat Disord 1996; 20:149–161.
87. Norring C. Borderline personality organization and prognosis in eating disorders.
Psychoanal Psychol 1993; 10:551–572.
88. Keel PK, Mitchell JE. Outcome in bulimia nervosa. Am J Psychiatry 1997; 154:
313–321.
89. Joiner TE Jr, Heatherton TF, Keel PK. Ten-year stability and predictive validity of
five bulimia-related indicators. Am J Psychiatry 1997; 154:1133–1138.
90. Bulik CM, Sullivan PF, Joyce PR, Carter FA, McIntosh VV. Predictors of 1-year
treatment outcome in bulimia nervosa. Compr Psychiatry 1998; 39:206–214.
91. Keel PK, Dorer DJ, Eddy KT, Delinsky SS, Franko DL, Blais MA, Keller MB,
Herzog DB. Predictors of treatment utilization among women with anorexia and
92. Meyers C, Waller G, Waters A. Emotional states and bulimic psychopathology. In:
Hoek H, Treasure J, Katzman M, eds. Neurobiology in the Treatment of Eating
Disorders. New York: John Wiley and Sons, 1998:263–279.
93. Westen D, Harnden-Fischer J. Personality profiles in eating disorders: rethinking
the distinction between axis I and axis II. Am J Psychiatry 2001; 158:547–562.
94. Polivy J, Herman CP. Dieting and binging: a causal analysis. Am Psychol 1985; 40:
193–201.
95. Borman Spurrell E, Wilfley DE, Tanofsky MB, Brownell KD. Age of onset for
binge eating: Are there different pathways to binge eating. Int J Eat Disord 1997;
21:55–65.
96. Steiger H, Lehoux P, Gauvin L. Impulsivity, dietary restraint, and the urge to binge
in bulimic eating syndromes. Int J Eat Disord 1999; 26:261–274.
97. Steiger H, Gauvin L, Jabalpurwala S, Séguin JR, Stotland S. Hypersensitivity to
social interactions in bulimic eating syndromes: relationship to binge-eating.
98. Safran J, Segal ZV. Interpersonal Process in Cognitive Therapy. Northvale, NJ:
Aronson, 1996.
232
10
Medical Comorbidity of Anorexia Nervosa,
Pauiine S.Powers and Yvonne Bannon
University of South Florida
Tampa, Florida, U.S.A.
Anorexia nervosa (AN), bulimia nervosa (BN), and binge eating disorder (BED)
are the three major eating disorders, affecting between 5 million and 10 million
people in the United States (1–3). More people have eating disorders than have
Alzheimer’s disease (4), and patients typically fall ill early in their life rather
than at the end of their life. Eating disorders are more common than
schizophrenia, which affects 2.2 million people in the United States (5). The
eating disorders are expensive to treat, in part because of the multiple medical
complications associated with them. The average yearly cost of treating eating
disorders in the United States is about $6 billion (6) compared to the global cost
of antipsychotic medication, which is about $7 billion (7). AN is the least common
eating disorder but the most expensive to treat: the average annual cost of
treating a patient with AN is $6045 compared to $4824 for a patient with
schizophrenia (6). BN and BED are somewhat less expensive to treat than AN,
similar to or more expensive than the treatment of obsessive-compulsive
disorder (8).
The eating disorders have high rates of medical comorbidity. AN has the
highest premature mortality rate of any psychiatric disorder. Untreated, after
20 years, the mortality rate is 18–20% (9). Since the typical patient falls ill in
midadolescence, by age 35 one-fifth of untreated patients have
died. Approximately two-thirds of the deaths are due to cardiac or renal
complications and about one-third due to suicide. Less is known about the
mortality for BN, but after 15–20 years 5% may have died (10). The premature
mortality rate for BED is poorly understood, but since most patients are obese,
the medical risks are probably those associated with obesity. These risks include
type 2 diabetes mellitus, hypertension, dyslipidemia, cardiovascular disease,
stroke, sleep apnea, gallbladder disease, hyperuricemia and gout, and
osteoarthritis. Although BED was described in 1959 (11), studies specific to the
complication of BED are only now being undertaken; the most work has been
done on the relationship between BED and diabetes mellitus. Although the
night eating syndrome was described by Stunkard nearly 50 years ago (12), it is
234 POWERS AND BANNON
only recently that the physiological concomitants of this condition have been
investigated.
DIAGNOSTIC ISSUES
The current diagnostic nomenclature for eating disorders includes AN, BN, and
eating disorders-not otherwise specified (ED-NOS). Under the current
system (13), BED is technically classified as an ED-NOS, along with
subsyndromal and atypical cases of AN and BN. It is likely that the next revision
of the Diagnostic and Statistical Manual (DSM) will include BED. The night eating
syndrome, often associated with obesity, is currently under study and may
eventually be classified as an eating disorder. The night eating syndrome is
usually defined as having three main features: morning anorexia, hyperphagia at
night, and insomnia (12,14). Nocturnal sleep-related eating disorder is a
condition in which night eating occurs during total or partial unconsciousness,
often during stage 3 or 4 sleep (14). One report (15) found that 60% of patients
with nocturnal sleep-related eating disorder had sleepwalking (somnambulism).
One important issue that has not been resolved is whether or not obesity
should be considered an eating disorder. Although obesity is often associated
with abnormalities in both food consumption and energy expenditure, early
studies found that there were no typical psychological symptoms associated with
obesity and therefore obesity was not considered a psychiatric disorder.
However, recent evidence documenting some of biological underpinnings of the
typical eating disorders and evidence documenting multiple psychological
problems in many individuals with obesity may result in obesity eventually
being considered an eating disorder.
Figure 1 illustrates some of the relationships between AN, BN, BED, and other
forms of obesity. As weight increases, the condition becomes more common
(e.g., AN is less common than BN, which is less common than BED, which is
less common than obesity unassociated with binge eating). With increasing
weight, more males are affected. For example, less than 10% of patients with
AN are male, about 20% of patients with BN are male, and about 30% of patients
with BED are male.
ASSESSMENT
Ideally, the initial evaluation of patients with eating disorders includes an
assessment of the common medical complications. Many of the physiological
complications are associated with signs and symptoms. For example, patients
with BN may have enlarged parotid glands on physical examination, and
laboratory testing may reveal elevated salivary amylase levels; patients with BED
MEDICAL COMORBIDITY 235
FIGURE 1 Eating Disorders Continuum.

may have headaches and elevated blood pressure. However, many of the serious
complications of eating disorders are not associated with symptoms or signs.
For example, osteopenia, which occurs in more than 60% of AN patients, may
be silent for years. Dyslipidemias that occur in BED may be unassociated with
signs or symptoms. Table 1 lists the laboratory tests that should be done for all
patients with eating disorders and specific tests for patients with AN, BN, or
BED.
During the assessment procedure, it is often helpful to engage the patient and
family in a discussion of the common medical complications that have been
identified. It is helpful to assist the patient in recognizing the relationship
between his or her behavior and the medical complication. For example, a
patient who is abusing laxatives may develop chest pain, have an arrhythmia on
physical examination, have an abnormal electrocardiogram with a prolonged QT
interval, and have hypokalemia. Helping the patient to see the relationship
between laxative abuse and the chest pain along with the risk of a fatal
arrhythmia may allow the patient to enter treatment and participate
meaningfully in the recovery process. This therapeutic use of signs and
symptoms is often best accomplished by the physician who initially assesses the
patient and makes the diagnosis.
The exact nature of the medical comorbidity depends on the patient’s age and
underlying physiological vulnerability, the intensity and type of purge behavior,
rapidity and extent of weight change, and duration of the eating disorder. For
example, a patient who loses 30% of her body weight in a 6month period is
likely to be at greater risk than a patient who loses 30% of her body weight over
a 3-year period. If other factors are equal, use of prescription diuretics and
certain types of laxatives appears to pose greater risk than similar intensity of
purging by vomiting.
TABLE 1 Laboratory Testing. SGPT (ALT)=Serum Glutamic Pyruvic Transaminase,

SGOT (AST)=Serum Glutamic Oxaloacetic Transaminase, DEXA=Dual Energy X-ray
Absorptiometry
MEDICAL COMPLICATIONS
Cardiac Complications
Cardiac complications are common in patients with eating disorders and are
associated with significant mortality and morbidity. About one-third of the
deaths in patients with AN are due to cardiac complications (16), and many of
the unexpected deaths in patients with BN are probably due to cardiac
arrhythmias. BED patients have higher psychophysiological arousal levels than
controls (17), and this may increase cardiac risk. Also, since most BED patients
are overweight or obese, they have an increased risk for cardiac comorbidity,
including hypertension and atherosclerosis.
Patients with AN or BN frequently have symptoms or signs of cardiac
malfunction, although patients with BED often do not. Common symptoms
suggesting cardiac abnormalities in AN or BN include light-headedness or
dizziness, coldness, palpitations, and chest pain. Associated symptoms include

orthostatic hypotension, hypothermia, acrocyanosis, arrhythmias, and
midsystolic click. An example of the relationship between symptoms, signs, and
laboratory data is as follows: A patient with chest pain may have a midsystolic
click and on echocardiogram have evidence of mitral valve prolapse.
Cardiac arrhythmias are a particularly important issue for patients with AN
and some patients with BN. QTc (corrected QT) interval prolongation and
dispersion are indicators of increased risk for cardiac arrhythmia and sudden
death, and is present in many patients with AN and some patients with BN (18).
QTc is a measure of repolarization of the heart, and prolongation or dispersion
(differences in various leads on the electrocardiagram) predisposes to a number
of arrhythmias, including torsade de pointes (TDP), which is a ventricular
tachycardia that can result in sudden death. Females are at greater risk for QT
prolongation and for TDP (19), and patients with AN or BN have an even
greater risk. Hypokalemia appears to increase the risk even more, although
some eating disorder patients without electrolyte disturbances may have
prolonged QTc intervals and TDP. This may be because serum potassium levels
may be normal even when total body potassium stores are decreased (20).
Certain medications are associated with prolonged QTc, including ziprasadone
(Geodon) and thioridazine (Mellaril). These and other medications that prolong
QTc should be avoided in eating disorder patients.
Mitral valve prolapse (MVP) occurs in a large percentage of patients (21).
Although there may be no symptoms, the presence of chest pain or a
midsystolic click is an indication for an echocardiogram, that may reveal MVP with
or without regurgitation. Tricuspid valve prolapse also occurs but is less
common. The etiology of the prolapse appears to be related to a decrease in size
of the heart with semistarvation and development of a socalled valvular-
ventricular disproportion (22). When chest pain occurs with MVP it may be
particularly difficult to treat. For patients with MVP, antibiotics should be taken
prior to procedures that result in a bacterial shower (e.g., dental procedures).
Ipecac is used to induce vomiting after accidental poisonings and is available
over the counter. About 1.3% of patients with eating disorders utilize ipecac to
induce vomiting (23). Ipecac is toxic to muscle (including the muscle of the
heart) and may result in a cardiomyopathy that can be fatal. Emetine, one of the
alkaloid constituents of ipecac, is particular dangerous. Patients who present
with proximal muscle weakness and a waddling gait should be assessed for use
of ipecac.
The refeeding syndrome (24) may occur during the early treatment of
patients with AN. The refeeding syndrome is the cardiovascular collapse that
can occur with refeeding and is very serious. Several factors contribute to this
syndrome including reduced myocardial mass, which makes it hard for the heart
to handle the increased circulatory load during refeeding and changes in serum
levels of phosphate, potassium, and magnesium. The immediate etiologic factor
appears to be that glucose in the food causes phosphate to enter the intracellular
space leading to low phosphate, which can affect contractility of the heart and lead
to heart failure. The symptoms include swollen ankles, shortness of breath,
tiredness, and anxiety. Signs include pedal edema, rales in the lungs, elevated
jugular venous pressure, and a midsystolic click. The chest radiograph reveals an
enlarged heart and evidence of congestive heart failure. Treatment includes
correcting electrolytes, decreasing fluid and caloric intake, diuretics, and
sometimes digitalis. Patients with the refeeding syndrome usually should be in
the coronary care unit with telemetry.
The key to preventing the refeeding syndrome is conservative management
for seriously underweight patients. Caloric intake should be low initially
coupled with correction of dehydration and electrolytes. Phosphorus and
electrolytes should be obtained every other day for the first 7–10 days and then
weekly. Calories should be increased slowly by 100–200 calories once or twice
a week during the first 2 weeks.
Renal Complications
Among AN patients, kidney failure accounts for about one-third of deaths.
Herzog and colleagues (25) have shown that high serum creatinine levels and
elevated uric acid levels predict a chronic course in AN patients. Although less
well understood, a number of BN patients also develop chronic renal
abnormalities. Several factors have been implicated in the kidney abnormalities
that develop, particularly electrolyte abnormalities. Hypokalemia and
metabolic acidosis that can result from persistent vomiting or laxative abuse are
implicated in the catabolism that occurs in uremia (26). End-stage renal disease
has been reported in patients with longstanding eating disorders; hypokalemic
nephropathy has been implicated in these cases (27). Volume depletion with
repeated dehydration from either semistarvation or purging or both contributes
to elevated blood urea nitrogen and elevated creatinine levels. Many patients
with eating disorders, especially AN, have decreased creatinine clearance as
well. Elevated levels of uric acid may account for the occasional patient who
develops gout. AN associated with rhabdomyolysis has been associated with
hypophosphatemia and renal failure (28). Nephrolithiasis may be more common
in patients with AN because chronic dehydration is known to contribute to the
development of renal stones (29). One study has found a lower filtration factor
in AN as well as a decreased urinary concentration capacity following fluid
deprivation both before and after administration of vasopressin (30).
Electrolyte Abnormalities
Electrolyte abnormalities occur with semistarvation and various types of purging
behavior, including vomiting, diuretic and laxative abuse, and ipecac abuse.
About half of AN and BN patients have electrolyte abnormalities (31); the most
common abnormality is elevated serum bicarbonate (metabolic alkalosis).
Potassium and chloride abnormalities also occur. Potassium depletion can occur
during semistarvation if there is inadequate potassium intake. During vomiting,
hydrochloric acid is vomited and there is compensatory excretion of potassium
through the kidneys, resulting in a decrease in potassium. Total body potassium
is often decreased even when serum potassium is normal (20); thus, a random
test of serum potassium may not detect abnormally low total body potassium
and the patient may still be at risk for various cardiac complications related to
hypokalemia. Abnormalities of chloride and bicarbonate may be present prior to
the detection of hypokalemia. Among patients who are dehydrated, potassium
may be in the normal range, but when hydration occurs the hypokalemia may
become apparent.
Clinical findings suggestive of electrolyte abnormalities include weakness,
constipation, dizziness, and depression. Patients with hypokalemia may develop
leg cramps. Patients with either hypomagnesemia or hypocalcemia may develop
tetany (sharp flexion of the ankle or wrist joints).
Many patients with bulimia nervosa are very secretive about use of
selfinduced vomiting, laxatives, or diuretics. Sometimes it is helpful to assess both
serum and urinary electrolytes to determine the type of purge behavior that has
occurred. Self-induced vomiting or diuretic abuse typically results in metabolic
alkalosis with decreased serum potassium and chloride and ele vated
bicarbonate, whereas laxative abuse more typically results in metabolic acidosis.
Endocrine Abnormalities
Multiple endocrine abnormalities occur in the eating disorders (32). AN is the
best studied (see Table 2), but there are similar if less pronounced abnormalities
in BN. The endocrine abnormalities seen in BED are usually those seen in
patients with obesity, including type 2 diabetes mellitus, but other endocrine
abnormalities also occur.
The key symptom reflecting abnormalities in the hypothalamic-pituitary-
gonadotropin (HPG) axis among AN patients is amenorrhea or irregular
menses. Nearly one-third of patients with AN develop amenorrhea prior to the
onset of weight loss and this finding is unexplained. It may relate to stress, but
may also indicate a vulnerability of the endocrine system in patients with anorexia
nervosa. Patients with AN have decreased gonadotropinreleasing hormone,
decreased follicle stimulating hormone, and decreased estrogen levels which

contribute to the development of osteopenia.
Abnormalities in the hypothalamic-pituitary-adrenal (HPA) axis are also very
common. A key endocrine finding in patients with AN is hypercortisolism
(which probably promotes the development of osteoporosis); elevated
corticotropin-releasing hormone (CRH) is also characteristic, but typically there
are normal adrenocorticotrophic hormone (ACTH) levels. Some patients with
BN have normal cortisol levels, but some have findings similar to those seen in
AN. Abnormal dexamethasone suppression tests are common in patients with
BN. The HPA axis has also been shown to be disturbed among patients with the
night eating syndrome. In these patients, there is an increased diurnal secretion
of cortisol and an attenuated response of ACTH and cortisol in plasma after
injection of CRH (33).
Among patients with BED, the emergence of non-insulin-dependent diabetes
mellitus (type 2 diabetes mellitus) is probably common. One study (34) found
that among obese type 2 diabetic adult patients, one-fourth had BED, although
it is unclear if BED preceded or followed onset of the diabetes. One method of
assessing diabetic control is measurement of glycosylated hemoglobin (HbAlc),
which gives an estimate of blood sugar over the previous 3 months. The
principle behind this is that the amount of hemoglobin that has become
conjugated with glucose is proportional to the average level of blood glucose
during the life of the red blood cell (an average of 3 months). Surprisingly, one
study found that patients with type 2 diabetes mellitus who also had BED did not
have evidence of higher glycosylated hemoglobin levels than type 2 diabetes
mellitus patients without BED (33). However, several studies have shown that
body mass index is positively correlated with increased levels of HbAlc, and
since some studies have shown that obese BED patients consume more calories
than obese non-BED patients, indirectly BED may result in elevated levels of
HbAlc (35).
Among patients with type 1 diabetes mellitus, the presence of BN or AN is
associated with higher levels of HbAlc than type 1 diabetic patients without
these conditions (36). Furthermore, patients with type 1 diabetes who have either
AN or BN have an increased risk of diabetic complications (37).
Leptin levels have been shown to be low in AN and normal-weight BN
patients but elevated in BED patients (38). Brewerton and colleagues (39) have
shown that plasma leptin levels are low in BN when compared to normal controls;
the differences do not appear related to body weight but may be related to HPA
axis activation and serotonin dysregulation. Among patients with night eating
syndrome, there is an attenuation of the nocturnal rise in leptin and
melatonin (40).
TABLE 2 Neuroendocrine Abnormalities in Anorexia Nervosa
ACTH, adrenocorticotrophic hormone; CRH, corticotropin-releasing hormone; GH,

growth hormone; FSH, follicle-stimulating hormone; GHRH, GH-releasing
hormone; GIH, somatostatin; GnRH, gonadotropin-releasing hormone; 5-HIAA,
5-hydroxyindoleacetic acid, a serotonin metabolite; HVA, homovanillic acid; IGF,
insulin-like growth factor; LH, lutenizing hormone; TSH, thyroid-stimulating hormone;
GH, growth hormone; NPY, neuropeptide Y.
a CSF level rather than serum level.
Source: Chial HJ and McAlpine DE. Anorexia nervosa: Manifestations and Management
for the Gastroenterologist. Reprinted with permission from the American College of
Gastroenterology, (Am J Gastroenterol 2002; 97:225–269).
Gastrointestinal Complications
Gastrointestinal complaints occur in more than 50% of patients with eating
disorders and include bloating, flatulence, constipation, decreased appetite,
abdominal pain, borborygmi, and nausea (41,42). Multiple complications can
occur involving the entire gastrointestinal track. The most common
complications include dental caries, enlarged parotid glands, gastroesophageal
reflux disorder, delayed gastric emptying, and delayed colonic transit.
Semistarvation and various types of purge behavior affect metabolic and
hormonal processes that influence gastrointestinal functioning. The
complications depend in part on the type and intensity of the behavior. For
example, an underweight patient who utilizes large quantities of stimulant-type
laxatives (e.g., senna) on a daily basis is likely to be at greater risk than a
normal-weight patient who utilizes bulk agents (e.g., docusate) infrequently.
Oral and dental complications are common among patients who binge eat and
purge by vomiting. A common dental problem is enamel erosion (perimolysis)
due to chronic exposure of tooth structures to hydrochloric acid from vomiting;
this is particularly pronounced on the lingual side of the teeth. Dental caries
(cavities) occur because of consumption of high-calorie binge foods and
weakened tooth structure. Hypersensitivity of the teeth to hot, cold, and sweet
foods occurs because of exposed dentin and root surfaces. Fillings and other
restorations may fail because of exposure to hydrochloric acid. Periodontitis
(gum disease) with gingival recession occurs due to nutritional deficiencies and
trauma to the mucosa. Salivary abnormalities include reduced salivary flow
(zerostomia) experienced symptomatically as a dry mouth and enlarged parotid
and submandibular glands.
Management of dental and oral problems begins with a full assessment.
The dentist may be the first health care professional to recognize the
problem (43,44). Dental care includes emergency treatment and management of
pain, basic fillings, and other routine care. If possible, restorative cosmetic
dental care should be postponed until the eating disorder is under control;
however, if this is not possible, porcelain and ceramics may be less likely to fail
than resins in patients with active BN. Educating patients who are still binge-
eating and purging can often help prevent some dental problems (45). For
example, it may help decrease dental complications if the patient uses antiacids
after vomiting and delays brushing. Application of fluoride and use of zylitol
chewing gum may also prevent some problems.
Parotid gland enlargement (sialadenosis) is common among patients with BN
and also occasionally occurs in those with restricting AN. Sialadenosis is a
noninflammatory recurrent enlargement of the salivary glands, most commonly
the parotids, which is almost always associated with an underlying systemic
disorder, including diabetes, alcoholism, malnutrition, and eating disorders.

The cause is thought to be a peripheral autonomic neuropathy resulting in
disordered metabolism and secretion (46). Serum amylase concentration is
elevated in patients with parotid gland enlargement and is correlated with
frequency of bulimic symptoms (47).
Delayed gastric emptying is common among patients with AN and also occurs in
BN patients (48). Most AN patients have delayed emptying of both solid and
liquids associated with decreased frequency, intensity, and coordination of
gastric contractions. Symptoms of nausea, vomiting, and gastric fullness
correlate with slowed gastric emptying. With weight gain most patients have an
improvement in gastric emptying as well as a decrease in gastrointestinal
symptoms. During weight restoration, liquids may be better tolerated than
solids, and weight gain alone usually results in improved gastric emptying.
Metoclopramide has been shown to improve gastric emptying and reduce
gastrointestinal symptoms in anorexic patients (49). One problem with
metoclopramide is that it crosses the blood-brain barrier and can result in
extrapyramidal symptoms; underweight patients seem to be particularly
vulnerable to this side effect. Patients with BN also may have delayed gastric
emptying as well as larger mean stomach capacity and blunted postprandial
cholecystokinin (CCK) release (50). The delayed CCK release may contribute
to the impaired satiety seen in BN patients.
Few studies of gastrointestinal functioning have been undertaken among
either BED or night eating syndrome patients. However, one study has found
that obese BED patients have an increase in gastric capacity compared to non-
BED obese patients (51).
Multiple esophageal problems can occur ranging from mild esophagitis to
esophageal rupture (52). Exposure to stomach acid secondary to self induced
vomiting can lead to esophagitis, erosion, ulcerations, esophageal strictures,
gastroesophageal reflux disease (GERD), or Barrett’s esophagus (characterized
by columnar-lined cells at the distal end of the esophagus). Four case reports
have suggested that BN may lead to Barrett’s esophagus and the eventual
development of carcinoma (53).
GERD symptoms are common among eating disorder patients, especially
those who purge by vomiting, although the incidence has not been fully
established (54). Symptoms include heartburn, regurgitation, dysphagia, and
angina-like chest pain. Treatment includes elevation of the head of the bed and
use of antacids. The next step of treatment is H2 antagonists that inhibit the
action of histamine at the H2 receptors of the parietal cells of the stomach. The
H2 antagonists include cimetidine (Tagamet) given 400–800 mg twice daily and
rantidine (Zantac) given 150 mg twice daily. Proton pump inhibitors can be
utilized to suppress the H+ K+ -ATPase enzyme. These medications include
omeprazole (Prilosec) 20 mg/day and lansoprazole (Prevacid) 30 mg/day; these

medications should be continued for 2 weeks after symptoms subside. It may be
that chronic GERD can result in Barrett’s esophagus.
Constipation is a very common problem that can result from semistarvation or
the long-term use of stimulant laxatives. Whole-gut transit times as well as
colonic transit time is delayed in both AN and BN patients (55,56). Weight and
eating normalization improves constipation. For patients who abuse laxatives,
several strategies can be helpful. Preparing the patient for the consequences of
stopping laxatives can be helpful. Most patients develop constipation, fluid
retention, bloating, and temporary weight gain. The patient should be advised
to discontinue use of all laxatives (including herbal remedies), drink 8–10 cups
of water per day (and no caffeinated fluids since they may act as a diuretic), to
exercise regularly and moderately, and to eat normally. Dietary changes that
promote normal bowel function include (a) consumption of whole-grain
breads, cereals, and crackers and wheat bran or foods with wheat bran added;
(b) addition of fruits and vegetables; and (c) avoidance of prunes, which contain
an ingredient that is actually an irritant laxative. When laxatives are required for
patients with chronic constipation or who are withdrawing from laxatives,
psyllium or docusate, both of which are bulk agents, can be used.
Hunger and satiety functions are frequently disturbed in eating disorder
patients. Multiple factors contribute to these problems. For example,
increased growth hormone, increased cortisol-releasing hormone, decreased
(3-endorphin, and autonomic insufficiency all contribute to delayed gastric
emptying, which leads to an increase in satiety. Similarly, a decrease in 5-HT,
an increase in neuropeptide Y, and a decrease in leptin also increase satiety.
Ghrelin, a hormone produced by the endocrine cells of the stomach, that is a
potent appetite stimulant, and is present at higher levels in BN patients than
in normal controls (57), may partly account for the impaired satiety seen in
these patients.
Central Nervous System Complications

There is much interest in the effects of starvation and malnutrition on the brain
and the nervous system. Several abnormalities have been noted. Structural
abnormalities in the brain have been identified in AN patients. Specifically,
ventricular and sulcal enlargement have been identified, which may not be fully
reversible with weight gain (58). Enlarged cerebrospinal fluid spaces have been
associated with sleep disruption and sleep-onset insomnia; sleep studies of
patients with enlarged cerebrospinal fluid spaces have shown an increased
amount of time spent in slow-wave sleep, and the duration of rapid eye
movement sleep is reduced (59). It is unclear if this is a result of nutritional
deficiencies or a particular cerebral dysfunction. Underweight patients with AN

have decreases in both the gray and white matter of the brain; white matter
volume appears to return to normal with weight gain, but the gray matter has
not been shown return to normal with weight gain (60). Cortisol has been
postulated to have a role in these abnormalities; urinary free cortisol is
positively correlated with total cerebral spinal fluid volume and inversely
correlated with total gray matter volume (61). Various studies have found an
incidence of cerebral atrophy in AN patients ranging from 25% to 75% (62).
When these abnormalities are suspected or confirmation is needed, magnetic
resonance imaging, rather than computed tomography, provides the most
information at the least risk to the patient.
It has been recognized for decades that patients with eating disorders have
abnormalities in cognition that have typically been classified as cognitive
distortions such as dichotomization or personalization (63); these abnormalities
have been generally thought to be due to psychological developmental delay or
regression. However, several studies have found neurocognitive abnormalities in
AN patients, including attention deficits, difficulty with abstraction and
flexibility of thought, abnormalities in executive function, visual spatial
abnormalities, and impaired memory (64–67). The relationships between these
neurocognitive abnormalities and the structural abnormalities and cortisol
abnormalities are unclear but may be important. There may also be a
relationship between neurocognitive abnormalities and body image (68), but the
exact nature of this relationship is unknown. Although many of the
neurocognitive abnormalities resolve with weight gain (69), subtle
abnormalities remain and may predispose to relapse (70).
A few studies have looked at various types of abnormalities in functional
imaging. Positron emission tomography (PET) has revealed that BN patients do
not have the right cortical activation seen in normal women (71). Brain imaging
has found a hypometabolism of glucose in several cerebral regions in both AN
and BN patients (72,73). In a study by Troop and colleagues (74), 74 patients with
a variety of eating disorders were compared to normal patients using PET to
assess activation during presentation of items expected to elicit the reaction of
disgust; disgust was found to be positively related to eating disorder symptoms
in the patients but not in the controls.
Many AN patients drink large quantities of water to assuage hunger, and
some patients with AN surreptitiously drink large quantities of water to gain
weight prior to being weighed by their physician. Occasionally this results in
severe water intoxication and electrolyte imbalance (particularly hyponatremia)
that can lead to neurological deficits including ataxia, cerebral edema, seizures,
central pontine myelinolysis, coma, and even death (75). Even when patients do
not have water intoxication, seizures can occur in about 5% of patients with
AN (76), probably due to electrolyte disturbances or hypoglycemia.
Peripheral Nervous System Complications

Peripheral neuropathies occur in 8–13% of anorexia nervosa patients (76,77).
The likely cause is chronic malnutrition including repeated episodes of
hypoglycemia. Patients with AN are also at increased risk for developing
localized compression neuropathies due to loss of subcutaneous tissue.
In AN and BN there is a reduction in noradrenergic activity in the central and
peripheral nervous systems. The clinical complications of this reduction are
hypotension, bradycardia, hypothermia, and depression (78).
In BN and BED elevated pain thresholds have been found. This may be
another partial explanation of the abnormal satiety response seen in both groups
of patients since the feeling of fullness and subsequent termination of a meal are
activated by the vagus nerve (79). In addition, Brewerton and colleagues (80)
have shown that BN patients have lower cerebral spinal fluid concentrations of
(3-endorphin than normal controls; furthermore, β-endorphin concentrations
were inversely correlated with measures of depression.
Musculoskeletal Complications
Tendons and muscles are also affected by AN. Diminished or absent tendon
reflexes may be found on physical examination. An early report (76) found that
generalized muscle weakness was present in 43% of 47 patients with AN.
Proximal muscle weakness that is associated with a selective disturbance of the
skeletal muscle metabolism is often seen. This is due to a diminished
lactate response to ischemic exercise (81) and is associated with reduced
serum carnosinase activity. This myopathy is associated with selective type
2 fiber atrophy with abnormal accumulation of glycogen within the muscle
fibers (82).
Short stature is a common finding among AN patients. One explanation has
been that if AN develops in adolescence prior to the closure of the epiphyses
there may be a potential irreversible growth retardation resulting in short
stature. However, it has also been suggested (83) that the observed short stature
seen in many AN patients may have a different etiology. Among 85 patients,
80% developed AN after menarche, but, compared to a control group, 76% of
the patients were below the 50th height percentile and nearly 15% were below
the 5th percentile.
Scoliosis may also be more common in patients with AN. Forty-four young
women with idiopathic scoliosis were found to be similar in height to controls,
but they had significantly lower body mass indices than controls and 25% were
in the range considered to represent anorexia (84). The reasons for this finding
are unknown.
Osteoporosis or osteopenia affects up to 90% of patients with AN and many
patients with BN who have a past history of AN (85). Among obese BED
patients, osteoporosis may be less common because obesity seems to be a
relative protective factor against loss of bone mineral density. Furthermore,
osteopenia develops early in the course of AN, often within the first year. Men
are affected as well, and there is some evidence that the loss may be greater in
men than in women (86). Furthermore, the loss of bone mineral density has long-
term consequences. Among a group of 208 patients followed an average of
40 years after diagnosis, the cumulative incidence of any fracture was 57% (87).
Among a group of 19 women who had been fully recovered for a mean of 21
years, femur BMD was still significantly less than among controls (88).
Although the cause of this significant bone loss is poorly understood, it is clear
that it is different from postmenopausal osteoporosis, which affects primarily
bone resorption; the osteoporosis of AN is related to both bone formation and
bone resorption.
The World Health Organization criteria for osteoporosis (89) is based on a
system of T scores that indicate the standard deviation (SD) from the mean
derived from a young normal sex-matched reference population. Normal bone
mineral density is defined as less than 1 SD below the normal mean; osteopenia
(low bone mass) is defined as 1 SD below to 2.5 SD below normal; osteoporosis
is defined as more than 2.5 SD below the normal mean, and severe osteoporosis
is more than 2.5 SD below the mean with one or more fractures. Z scores
compare the bone density of a subject with that of an age- and sex-matched
control.
The majority of peak bone mass is achieved by age 20, although there is some
continuing accumulation until age 30. AN typically develops during adolescence
at a time when peak bone mass has not been achieved. The osteoporosis that
occurs in AN affects both trabecular bone (found mainly in the spine) and
cortical bone (found mainly in the long bones). The spine and hip seem to be
particularly vulnerable.
Bone is constantly in a state of remodeling, which involves balanced coupling
of bone formation and bone resorption. Osteoblasts are cells involved in bone
formation and osteoclasts are cells involved in boneresorption. Serum and
urinary markers of bone formation include osteocalcin, bone-specific alkaline
phosphatase, and procollagen-I carboxy terminal propeptide; bone resorption
markers include urine deoxypyridinoline and serum carboxy terminal type I
propeptide and 7V-telopeptide. In AN, there is an increase in resorption
accompanied by a decrease or no change in the rate of bone formation.
The cause of this uncoupling of bone formation and resorption is unknown

but a number of factors have been implicated. In AN, there are decreased levels
of insulin-like growth factor-1 (IGF-1), which is a bone tropic factor, and levels
of IGF-1 increase with weight gain (90). Although not proven, elevated cortisol
levels (a very common finding in AN) may facilitate trabecular bone loss by
suppressing osteoblast proliferation. The exact role of amenorrhea and
hypoestrogenemia in the pathogenesis of osteoporosis is unclear. Although bone
density has been shown to be related to the length of amenorrhea (91), and
mean serum estrogen and progesterone levels are consistently low in
underweight anorexics, estrogen administration has not been consistently shown
to improve bone density (92,93).
At present, a dual-energy X-ray absorptiometry (DEXA) is the gold standard
for detection of osteoporosis (94). Patients who have osteopenia should have
repeat tests every 18–20 months. Although ultrasound is less expensive and
avoids radiation exposure, it has not yet been shown to be effective in detecting
osteoporosis at the sites most vulnerable in AN patients.
Unfortunately, treatments have not been shown to be completely effective in
restoring normal bone mineral density (91,95). Weight gain prevents further
loss of BMD and after time may improve bone mineral density. Shortterm
nutrition rehabilitation has been shown to improve bone formation markers as
well as IGF-1 and leptin, but whether this translates into improved bone mineral
density is not yet known (96). Although it seems reasonable that estrogen
supplementation would improve bone mineral density, its efficacy has not been
established despite many attempts to do so. One study has shown that
recombinant human IGF-1 has some benefit in restoring BMD and that
administration with oral contraceptive may be of additional benefit (97).
However, this treatment is still experimental. At present, the most reasonable
approach is to restore normal weight and normal menses while ensuring
adequate calcium intake (1000–1500 mg/day) and vitamin D (600 IU
daily). Some patients may benefit from estrogen supplementation, and some
patients with osteoporosis may benefit from biphosphates which reduce bone
resorption. Alendronate or risendronate sodium might be considered, although
they both cause gastrointestinal side effects, including esophagitis, and must be
used carefully. Alendronate (Fosamax) is currently available in a once-a-week
preparation, which may make it easier for selected patients to use.
Gynecological and Obstetrical Complications

Eating disorder patients have multiple gynecological and obstetrical
complications. Amenorrhea is a key finding in most AN patients, and many
patients with BN also have amenorrhea or irregular menses. Multiple factors
contribute to the amenorrhea, including weight loss, decrease in body fat,

stress, excessive exercise, and decreased estrogen and follicle stimulating
hormone levels.
Among AN patients and some BN patients, the ovarian and uterine volume is
decreased, depth of endometrial thickness is decreased, and atrophic changes
occur in the vaginal mucosa (98). Multifollicular ovaries also occur in AN (99).
The change from microfollicular to multifollicular to dominant follicle and
finally to ovulatory follicle occurs during puberty. Many underweight AN
patients have multifollicular ovaries; with weight gain there is progression to the
ovulatory stage. Some of these findings can be helpful in establishing a goal
weight during treatment of AN patients (100). For example, increase to normal
thickness of the endometrium and progression to a single ovulatory follicle have
been used to signal achievement of normal body weight. Polycystic ovaries have
also been reported in patients with BN (101) and with BED (102).
Many women with AN are infertile, but most women with BN are fertile.
Complications for both the mother and the baby are common when the eating
disorder is active. Many patients with AN do not gain adequate weight during
pregnancy. Some patients are able to gain adequate weight or interrupt purge
behavior for the safety of the baby, but the eating disorder frequently recurs
after delivery (103); women with chronic AN or those with BN with a past
history of AN were more likely to relapse after delivery (104). Women with
active eating disorders are at greater risk for cesarean section and for postpartum
depression (105). Complications for the infant include preterm delivery, low
birthweight, intrauterine growth restriction, and low Apgar scores (106). If
possible, it is wise for the patient to delay pregnancy until her eating disorder is
in remission. For patients who become pregnant while their eating disorder is still
active, a multidisciplinary team is usually needed and should include an
obstetrician who specializes in high-risk pregnancies, a neonatologist, and a
psychiatrist (107).
There may be effects on the infants of women with eating disorders. One
study compared infants of mothers who had an eating disorder during the first
year postpartum to infants of mothers without such a history (108). Mothers
with a history of an eating disorder postpartum were more intrusive with their
infants during mealtimes and during play, and their infants had more negative
emotional tones, and the mealtimes were more conflicted than those of infants
of normal mothers. Furthermore, infants of these mothers tended to be lighter
in weight during their second year of life than infants of control mothers.
CONCLUSIONS
Multiple medical complications are associated with the three major eating
disorders. These complications frequently require treatment by several
specialists at various points in the recovery process. It is important for the many
physicians and other care providers involved to work collaboratively to address
the problems in an organized way utilizing an agreed-upon timeline. For
example, during the initial management of AN, the cardiologist and psychiatrist
may need to work together to design a program that prevents emergence of the
refeeding syndrome. Later in the course of treatment when the patient is near
ideal body weight, the activity therapist may have to be actively involved in
designing an appropriate exercise program to reduce the impact of
osteoporosis. Once the patient has achieved ideal body weight, the gynecologist
may be needed to assess possible need for hormonal intervention if menses have
not resumed. During the entire course of treatment from initial assessment to
eventual discharge from treatment (often several years later), there needs to be
a team leader organizing the delivery of care. When this occurs, outcome is
often improved.
REFERENCES
1. Yager J, Andersen A, Devlin M, Egger H, Herzog D, Mitchell J, Powers P, Yates
A, Zerbe K. American Psychiatric Association Work Group on Eating Disorders:
practice guideline for the treatment of patients with eating disorders (revision). Am
J Psychiatry 2000; 157(suppl):l-39.
2. Garfinkel PE, Lin E, Goering P, Spegg C, Goldbloom D, Kennedy S, Kaplan AS,
Woodside DB. Should amenorrhea be necessary for the diagnosis of anorexia
nervosa? Br J Psychiatry 1996; 168:500–506.
3. National Institute of Mental Health. The Numbers Count. 2002. Web Address
Http//www.nimh.nih.gov/publicat/numbers.cfm.
4. McDowell I. Alzheimer’s disease: insights from epidemiology. Aging (Milano)
2001; 13:143–162.
5. Narrow WE. One-year prevalence of mental disorders, excluding substance use
disorders, in the US: NIMH ECA prospective data. Population estimates based on
US Census estimated residential population age 18 and over on July 1, 1998.
Unpublished.
6. Striegel-Moore RH, Leslie D, Petrill SA, Garvin V, Rosenheck RA. One year use
and cost of inpatient and outpatient services among female and male patients with
an eating disorder: evidence from a national database of health insurance claims. Int
J Eat Disord 2000; 27:381–389.
7. Tamminga CA, Lieberman JA. Schizophrenia research series: from molecule to
public policy. Biol Psychiatry 1999; 46:3.
8. Agras WS. The consequences and costs of the eating disorders. Psychiatr Clin
North Am 2001; 24:371–379.
9. Theander S. Anorexia nervosa. A psychiatric investigatio of 94 female patients.
Arch Psychiatr Scand Suppl 1970; 214:1–194.
10. Nielsen S, Moller-Madsen S, Isager T, Jorgensen J, Pagsberg K, Theander S.
Standardized morality in eating disorders—a quantitative summary of previously
published and new evidence. J Psychosom Res 1998; 44:413–434.
11. Stunkard AJ. Eating patterns and obesity. Psychiatr Q 1959; 33:284–294.
12. Stunkard AJ, Grace WJ, Wolff HG. The night-eating syndrome: a pattern of food
intake among certain obese patients. Am J Med 1955; 19:78–86.
Disorders. 4th ed. Washington, DC: Text Revision (DSM-IVTR), 2000: 583–586,
785–787.
14. Stunkard AJ, Allison KC. Two forms of disordered eating in obesity: binge eating
and night eating. Obesity 2003; 27:1–12.
15. Schenck C, Mahowald MW. Review of nocturnal sleep-related eating disorders.
Int J Eat Disord 1994; 15:343–356.
16. Schocken DD, Holoway JD, Powers PS. Weight loss and the heart: effects of
anorexia nervosa and starvation. Arch Intern Med 1989; 149:877–881.
17. Vogele C, Florin I. Psychophysiological responses to food exposure: an
experimental study in binge eaters. Int J Eat Disord 1997; 21:147–157.
18. Swenne I, Larsson PT. Heart risk associated with weight loss in anorexia nervosa
and eating disorders: risk factors for QTc interval prolongation and dispersion.
Acta Paediatr 1999; 88:304–309.
19. Wolbrette D. Gender differences in the proarrhythmic potential of QT-prolonging
drugs. Curr Womens Health Rep 2002; 2:105–109.
20. Powers PS, Tyson IB, Stevens BA, Heal AV. Total body potassium and serum
potassium among eating disorder patients. Int J Eat Disord 1995; 18:269–276.
21. de Simone G, Scalfi L, Galderisi M, Celentano A, Di Biase G, Tammaro P,
Garofalo M, Mureddu GF, de Divitiis O, Contaldo F. Cardiac abnormalities in
young women with anorexia nervosa. Br Heart J 1994; 71:287–292.
22. Oka Y, Ito T, Matsumoto S, et al. Mitral valve prolapse in patients with anorexia
nervosa. Two-dimensional echocardiographic study. Jpn Heart J 1987; 28:
873–882.
23. Powers PS, Mitchell JE, Garner DM, Monson N. International eating
disorders database: preliminary findings. Presented at Academy for Eating
Disorders, 2001 International Conference on Eating Disorders, April 27, 2001,
Boston.
24. Mehler PS. Eating disorders: 1. Anorexia nervosa. Hosp Pract 1996; 31:109–113.
25. Herzog W, Deter HC, Fiehn W, Petzold E. Medical findings and predictors of
long-term physical outcome in anorexia nervosa: a prospective, 12-year followup
study. Psychol Med 1997; 27:269–279.
26. Franch HA, Mitch WE. Catabolism in uremia: the impact of metabolic acidosis.
J Am Soc Nephrol 1998; 9:878–81.
27. Abdel-Rahman EM, Moorthy AV. End-stage renal disease (ESD+RD in patients
with eating disorders. Clin Nephrol 1997; 47:106–111.
28. Wada S, Nagase T, Koike Y, Kugai N, Nagata N. A case of anorexia nervosa with
acute renal failure induced by rhabdomyolysis: possible involvement of
hypophosphatemia or phosphate depletion. Intern Med 1992; 31:478–482.
29. Silber TJ, Kass EJ. Anorexia nervosa and nephrolithiasis. J Adol Health Care 1984;
5:50–52.
30. Aperia A, Broberger O, Fohlin L. Renal function in anorexia nervosa. Acta
Paediatr Scand 1978; 67:219–224.
31. Mitchell JE, Pyle RL, Eckert ED, Hatsukami D, Lentz R. Electrolyte and other
physiological abnormalities in patients with bulimia. Psychol Med 1983; 14:
273–278.
32. Chial HJ, McAlpine DE, Camilleri M. Anorexia nervosa: manifestations and
management for the gastroenterologist. Am J Gastroenterol 2002; 97:255–269.
33. Birketvedt GS, Sundsfjord J, Florholmen JR. Hypothalamic-pituitary-adrenal axis
in night eating syndrome. Am J Physiol Endocrinol Metab 2002; 282(2): 366–369.
34. Crow S, Kendall D, Praus B, Thuras P. Binge eating and other psychopathology in
patients with type II diabetes mellitus. Int J Eat Disord 2001; 30:222–226.
35. Goebel-fabbri AE, Fikkan JL, Connell A, Vangsness L, Anerson B. J Binge eating,
BMI, emotional distress, and health status in women with type 2 diabetes.
Presented at the Academy for Eating Disorders 2002 International Conference on
Eating Disorders and Clinical Teaching Day, Boston.
36. Birk R, Spencer ML. The prevalence of anorexia nervosa, bulimia, and induced
glycosuria in IDDM females. Diabetes Educator, 1989, 336–341.
37. Rydall A, Rodin G, Olmsted M, Devenyi R, Daneman D. A four year follow-up
study of eating disorders and medical complications in young women with insulin-
dependent diabetes mellitus [Abstr]. Psychosom Med 1994; 56:179.
38. Monteleone P, DiLieto A, Tortorella A, Longobardi N, Maj M. Circulating leptin
in patients with anorexia nervosa, bulimia nervosa or binge-eating disorder:
relationship to body weight, eating patterns, psychopathology and endocrine
changes. Psychiatry Res 2000; 94(2): 121–129.
39. Brewerton TD, Lesem MD, Kennedy A, Garvey WT. Reduced plasma leptin
concentrations in bulimia nervosa. Psychoneuroendocrinology 2000; 25:548–649.
40. Birketvedt GS, Florholmen J, Sundsfjord J, Osterud B, Dinges D, Bilker
W, Stunkard A. Behavioral and neuroendocrine characteristics of the night-eating
syndrome. JAMA 1999; 282:657–663.
41. Chami TN, Anderson AE, Crowell MD, Schuster MM, Whitehead WE.
Gastrointestional symptoms in bulimia nervosa: effects of treatment. Am J
Gastroenterology 1995; 90:88–92.
42. McClain CJ, Humphries LL, Hill KK, Nicki NJ. Gastrointestinal and nutritional
aspects of eating disorders. J Am Coll Nutr 1993; 12:466–474.
43. Steel AW, Mehler PS. Oral and dental complications. In Eating Disorders: A Guide
to Medical Care and Complications. Baltimore: Johns Hopkins University Press,
1999:144–152.
44. Schwartz MI. Dentistry and eating disorders. In AABA Newsletter, Fall 1997.
45. Sundaram G, Bartlett D. Preventative measures for bulimic patients with dental
erosion. Eur J Prosthodont Restor Dent 2001; 9:25–29.
46. Coleman H, Altini M, Nayler S, Richards A. Sialadenosis: a presenting sign in
bulimia. Head Neck 1998; 20:758–760.
47. Metzger ED, Levine JM, McArdle CR, Wolfe BE, Jimerson DC. Salivary gland
enlargement and elevated serum amylase in bulimia nervosa. Biol Psychiatry 1999;
45:1520–1522.
48. Rigaud D, Bedig G, Merrouche M, Vulpillat M, Bonfils S, Apfelbaum M. Delayed
gastric emptying in anorexia nervosa is improved by completion of a renutrition
program. Dig Dis Sci 1988; 33:919–925.
49. Saleh JW, Lebwohl P. Metoclopramide-induced gastric emptying in patients with
anorexia nervosa. Am J Gastroenterol 1980; 74:27–32.
50. Devlin MJ, Walsh BT, Guss JL, Kissileff HR, Liddle RA, Petkova E. Postprandial
cholecystokinin release and gastric emptying in patients with bulimia nervosa. Am J
Clin Nutr 1997; 65:114–120.
51. Geliebter A, Hashim SA. Gastric capacity in normal, obese and bulimic women.
Physiol Behav 2001; 74:743–746.
52. Kiss A, Wiesnagrotski S, Abatzi TA, Meryn S, Haubenstock A, Base W. Upper
gastrointestinal endoscopy findings in patients with long-standing bulimia nervosa.
Gastrointest Endos 1989; 35:516–518.
53. Dessureault S, Coppola D, Weitzner M, Powers P, Karl RC. Barrett’s oesophagus
and squamous cell carcinoma with psychogenic vomiting. Int J Gastrointest Cancer
2002; 32:57–61.
54. Bartlett DW, Evans DF, Smith BG. The relationship between gastroesophageal
reflux disease and dental erosion. J Oral Rehabil 1996; 23:289–297.
55. Kamal N, Chami T, Andersen A, Rosell FA, Schuster MM, Whitehead WE.
Delayed gastrointestinal transit times in anorexia nervosa and bulimia nervosa.
Gastroenterology 1991; 101:1320–1324, 1991.
56. Chun AB, Sokol MS, Kaye WH, Hutson WR, Wald A. Colonic and anorectal
function in constipated patients with anorexia nervosa. Am J Gastroenterol 1997;
92:879–883.
57. Tanaka M, Naruo T, Muranaga T, Yasuhara D, Shiiya T, Nakazato M, Matsukura S,
Nozoe S. Increased fasting plasma ghrelin levels in patients with bulimia nervosa.
Eur J Endocrin 2002; 146:R1-R3.
58. Katzman DK, Zipursky RB, Lambe EK, Mikulis DJ. A longitudinal
magnetic resonance imaging study of brain changes in adolescents with anorexia
nervosa. Arch Pediatr Adolesc Med 1997; 151:793–797.
59. Eiber R, Friedman S. Correlation between eating disorders and sleep disturbances.
Encephale 2001; 27:429–434.
60. Lambe EK, Katzman DK, Mikulis DJ, Kennedy SH, Zipursky RB. Cerebral gray
matter volume deficits after weight recovery from anorexia nervosa. Arch Gen
Psychiatry 1997; 54:537–542.
61. Katzman DK, Lambe EK, Mikulis DJ, Ridgley JN, Goldbloom DS, Zipursky RB.
Cerebral gray matter and white matter volume deficits in adolescent girls with
anorexia nervosa. J Pediatr 1996; 129:794–803.
62. di Pietralata GM. Imaging techniques in the management of anorexia and bulimia
nervosa. Eat Weight Disord 2002; 7:146–151.
63. Garner DM, Fairburn CG, Davis R. Cognitive-behavioral treatment of bulimia
nervosa. A critical appraisal. Behav Modif 1987; 11:398–431.
64. Fassino S, Piero A, Daga GA, Leombruni P, Mortara P, Rovera GG. Attentional
biases and frontal functioning in anorexia nervosa. Int J Eat Disord 2002; 31:
274–283.
65. Neumarker KJ, Bzufka WM, Dudeck U, Hein J, Neumarker U. Are there specific
disabilities of mnumber processing in adolescent patients with anorexia nervosa?
Evidence from clinical and neuropsychological data when compared to
morphometric measures from magnetic resonance imaging. Eur Child Adol
Psychiatry 2000; 9:111–121.
66. Seed JA, Dixon RA, McCluskey SE, Young AH. Basal activity of the hypothalamic-
pituitary-adrenal axis and cognitive function in anorexia nervosa. Eur Arch
Psychiatry Clin Neurosci 2000; 250:11–15.
67. Mathias JL, Kent PS. Neuropsychological consequences of extreme weight loss and
dietary restriction in patients with anorexia nervosa. J Clin Exp Neuropsychol
1998; 20:548–564.
68. Epstein J, Wiseman CV, Sunday SR, Klapper F, Alkalay L, Halmi KA.
Neurocognitive evidence favors “topdown”over “bottom up”mechanisms in the
pathogenesis of body size distortions in anorexia nervosa. Eat Weight Disord 2001;
6:140–147.
69. Lauer CJ, Gorzewski B, Gerlinghoff M, Backmund H, Zihl J. Neuropsychological
assessments before and after treatment in patients with anorexia nervosa and
bulimia nervosa. J Psychiatr Res 1999; 33:129–138.
70. Green MW, Elliman NA, Wakeling A, Rogers PJ. Cognitive functioning, weight
change and therapy in anorexia nervosa. J Psychiatr Res 1996; 30:401–410.
71. Wu JC, Hagman J, Buchsbaum MS, Blinder B, Derrfler M, Tai WY, Hazlett E,
Sicotte N. Greater left cerbral hemispheric metabolism in bulimia assessed by
positron emission tomography. Am J Psychiatry 1990; 147:309–312.
72. Delvenne V, Lostra F, Goldman S, Biver F, DeMaertelaer V, AppelboomFondu J,
Schoutens A, Bidaut LM, Luxen A, Mendelwicz J. Brain hypometabolism of
glucose in anorexia nervosa: A PET-scan study. Biol Psychiatry 1995; 37:161–169.
73. Delvenne V, Goldman S, Simon Y, De Maertelaer V, Lotstra F. Brain
hypometabolism of glucose in bulimia nervosa. Int J Eat Disord 1997; 21:313–326.
74. Troop NA, Murphy F, Bramon E, Treasure JL. Disgust sensitivity in eating
disorders: a preliminary investigation. Int J Eat Disord 2000; 27:446–451.
75. Amann B, Schafer M, Sterr A, Arnold S, Grunze H. Central pontine myelinolysis in
a patient with anorexia nervosa. Int J Eat Disord 2001; 30:462–466.
76. Patchell RA, Fellows HA, Humphries LL. Neurologic complications of anorexia
nervosa. Acta Neurol Scand 1994; 89:111–116.
77. MacKenzie JR, LaBan MM, Sackeyfio AH. The prevalence of peripheral
neuropathy in patients with anorexia nervosa. Arch Phys Med Rehab 1989; 70:
827–830.
78. Pirke KM. Central and peripheral noradrenalin regulation in eating disorders.
79. Raymond NC, deZwaan M, Faris PL, Nugent SM, Achard DM, Crosby RD,
Mitchell JE. Pain thresholds in obese binge-eating disorder subjects. Biol Psychiatry
1995; 37:202–204.
80. Brewerton TD, Lydiara RB, Laraia MT, Shook JE, Ballenger JC. CSF
betaendorphin and dynorphin in bulimia nervosa. Am J Psychiatry 1992; 149:
1086–1090.
81. McLoughlin DM, Wassif WS, Morton J, Spargo E, Peters TJ, Russell GF.
Metabolic abnormalities associated with skeletal myopathy in severe anorexia
nervosa. Nutirtion 2000; 16:192–196.
82. McLoughlin DM, Spargo E, Wassif WS, et al. Structural and functional changes in
skeletal muscle in anorexia nervosa. Act Neuropathol 1998; 95: 632.
83. Nussbaum M, Baird D, Sonnenblick M, Cowan K, Shenker IR. Short stature in
anorexia nervosa patients. J Adol Health Care 1985; 6:453–455.
84. Smith FM, Latchford G, Hall RM, Millner PA, Dickson RA. Indications of
disordered eating behaviour in adolescent patients with idiopathic scoliosis. J Bone
Joint Surg Br 2002; 84:392–394.
85. Grinspoon S, Thomas E, Pitts S, Gross E, Mickley D, Miller K, Herzog D,
Klibanski A. Prevalence and predictive factors for regional osteopenia in women
with anorexia nervosa. Ann Intern Med 2000; 133:790–794.
86. Andersen AE, Watson T, Schlechte J. Osteoporosis and osteopenia in men with
eating disorders. Lancet 2000; 355:1967–1968.
87. Lucas AR, Melton LJIII, Crowson CS, O’Fallon WM. Long-term fracture risk
among women with anorexia nervosa: a population-based cohort study. Mayo Clin
Proc 1999; 74:972–977.
88. Hartman D, Crisp A, Rooney B, Rackow C, Atkinson R, Patel S. Bone density of
women who have recovered from anorexia nervosa. Int J Eat Disord 2000; 28:
1007–1112.
89. Kanis JAWHO Study Group. Assessment of fracture risk and its application for
screening post-menopausal osteoporosis. Osteoporos Int 1994; 4:368–381.
90. Soyka LA, Grinspoon S, Levitsky LL, Herzog DB, Klibanksi A. The effects of AN
on bone meabolism in female adolescents. J Clin Endocrinol Metab 1999; 84:
4489–4496.
91. Andersen AE, Woodward PJ, LaFrance N. Bone mineral density of eating disorder
subgroups. Int J Eat Disord 1995; 18:335–342.
92. Klibanski A, Biller BM, Schoenfeld DA, Herzog DB, Saxe VC. The effects of
estrogen administration on trabecular bone loss in young women with AN. J Clin
Endocrinol Metab 1995; 80:898–904.
93. Karlsson MK, Weigall SJ, Duan Y, Seeman E. Bone size and volumetric density in
women with AN receiving estrogen replacement therapy and in women recovered
from AN. J Encocrinol Metab 2000; 85:3177–3182.
94. Wolfert A, Mehler PS. Osteoporosis: prevention and treatment in anorexia
nervosa. Eat Weight Disord 2002; 7:72–81.
95. Bachrach LK, Katzman DK, Litt IF, Guido D, Marcus R. Recovery from
osteopenia in adolescent girls with AN. J Clin Endocrinol Metab 1991; 72:
602–606.
96. Heer M, Mika C, Grzella I, Drummer C, Herpertz-Dahlmann B. Changes in bone
turnover in patients with anorexia nervosa during eleven weeks of inpatient dietary
treatment. Clin Chem 2002; 48:754–760.
97. Grinspoon S, Thomas L, Miller K, Herzog D. Klibanski A Effects of recombinant
human IGF-I and oral contraceptive administration on bone density in anorexia
nervosa. J Clin Endocrinol Meab 2002; 87:2883–2891.
98. Lai KY, de Bruyn R, Lask B, Bryant-Waugh R, Hankins M. Use of pelvic
ultrasound to monitor ovarian and uterine maturity in childhood onset anorexia
nervosa. Arch Dis Child 1994; 71:228–231.
99. Treasure JL, Gordon PA, King EA, Wheeler M, Russell GF. Cystic ovaries: a phase
of anorexia nervosa. Lancet 1985; 2:1379–1382.
100. Treasure JL. The ultrasonographic features in anorexia nervosa and bulimia
nervosa: a simplified method of monitoring hormonal states during weight gain.
J Psychosom Res 1988; 32:623–634.
101. Jahanafar S, Eden JA, Nguyent TV. Bulimia nervosa and polycystic ovary syndrome.
Gynecol Endocrinol 1995; 9:113–117.
102. Johnson JG, Spitzer RL, Williams JB. Health problems, impairment and illnesses
associated with bulimia nervosa and binge eating disorder among primary care and
obstetric gynaecology patients. Psychol Med 2001; 31:1455–1466.
103. Blais MA, Becker AE, Burwell RA, Flores AT, Nussbaum KM, Greenwood DN,
Ekeblad ER, Herzog DB. Pregnancy: outcome and impact on symptomatology in a
cohort of eating-disordered women. Int J Eat Disord 2000; 27: 140–149.
104. Morgan JF, Lacey JH, Sedgwick PM. Impact of pregnancy on bulimia nervosa. Br J
Psychiatry 1999; 174:278.
105. Franko DL, Blais MA, Becker AE, Delinsky SS, Greenwood DN, Flores AT,
Ekeblad ER, Eddy KT, Herzog DB. Pregnancy complications and neonatal
outcomes in women with eating disorders. Am J Psychiatry 2000; 158:1461–1466.
106. James DC. Eating disorders, fertility, and pregnancy: relationships and
complications. J Perinat Neonat Nurs 2001; 15:36–48.
107. Franko DL, Spurrell EB. Detection and management of eating disorders during
pregnancy. Obstet Gynecol 2000; 95:942–946.
108. Stein A, Woolley H, Cooper SD, Fairburn CG. An observational study of mothers
with eating disorders and their infants. J Child Psychol Psychiatry 1994; 35:
733–748.
11
Neurotransmitter Dysregulation in Anorexia
Disorder
Timothy D.Brewerton
Medical University of South Carolina
Charleston, South Carolina, U.S.A.
Howard Steiger
Douglas Hospital
Montreal, Quebec, Canada
The current system of psychiatric diagnosis, DSM-IV (1), addresses two official
eating disorder (ED) syndromes—anorexia nervosa (AN) and bulimia nervosa
(BN)—and a third (still provisional) diagnostic entity—binge eating disorder
(BED). However, BED has all but officially been recognized as a distinct eating
syndrome. AN, BN, and BED are all polysymptomatic syndromes, defined by
maladaptive attitudes and behaviors around eating, weight, and body image, but
typically including “nonspecific” disturbances of self-image, mood, impulse
regulation, and interpersonal functioning. All three syndromes are known to be
associated with significant mortality and morbidity, both medical and
psychiatric (2,3). Despite popular beliefs, there is no convincing evidence that
cultural factors alone cause eating disorders. Indeed, during the past few years
(and especially the last decade) investigations into the role of neurotransmitters
and other neuromodulators in the eating disorders have been highly productive,
and have implicated primary neurotransmitter disturbances in the etiology of both
AN and BN. Furthermore, recent data clearly identify strong genetic factors in
AN and BN, which appear to share common genetic vulnerabilities (4,5) linked
to obsessionality, perfectionism, anxiety, and/or behavioral inhibition (6,7).
One powerful piece of evidence to support monoamine involvement in the
eating disorders is the observation that antidepressant medications can be
beneficial in controlled studies, not only in BN patients but in recovered AN
patients as well (8).
However, it is also clear that some disturbances are consequences of the
abnormal eating practices and nutritional disturbances that characterize these
disorders (9), which in turn exacerbate or perpetuate signs and symptoms (10).
258 BREWERTON AND STEIGER
This perspective, taken together with the disorders’ consequences, challenges,

and costs, compels us toward a better understanding of the biological
mechanisms underlying all stages and types of eating disorders. The
identification of the psychobiological underpinnings of these conditions may be
useful in many ways, including the development of improved medical and
psychopharmacological interventions, improved education and psychotherapy
for patients and their families, and improved prevention efforts at a primary
level.
It must be emphasized that most measurements of neurotransmitter function
provide only a glimpse into the state of the organism at that moment. Sorting
out what is trait and what is state related has been a challenging focus of
neurotransmitter research in the eating disorders.
MONOAMINES
The classical monoaminergic neurotransmitter systems, including serotonin
(5-hydroxytryptamine, 5-HT), norepinephrine (NE), and dopamine (DA), have
been fairly extensively studied in the eating disorders using available techniques
in biological psychiatry. Most of these studies have been conducted during the
active state of illness, during which severe nutritional compromise may
represent an important confound. Dieting and/or semistarvation clearly
depletes central monoamines and leads to altered neurotransmitter levels and
receptor sensitivity in animals and humans (11–15). To avoid this problem, a
more recent strategy has been to study “recovered” patients, i.e., AN and BN
patients who have attained normalization of eating and weight, resumption of
menses and/or normalization of gonadal hormone levels, and abatement of
typical cognitive features to subclinical levels. This strategy attempts to
minimize starvation state-related effects and to reveal potential trait-related
disturbances or vulnerabilities. However, the long-term effects of chronic
malnutrition and disordered eating behaviors on the brain (similar to substance
use disorders) should not be minimized. Studies of transmitter function in
at-risk premorbid individuals as well as nonaffected identical and fraternal
twins, siblings, and other first-degree relatives of ED patients could begin to
confirm trait-related disturbances.
Neurotransmitter function in patients with EDs have been investigated using
a variety of existing techniques and methodologies, each of which has its own
advantages and disadvantages. Studies of cerebrospinal fluid (CSF)
concentrations of the major metabolites have been a popular strategy and
include measures of 5-hydroxyindoleacetic acid (5-HIAA) for 5-HT,
3-methoxy-4hydroxyphenylglycol (MHPG) for NE, and homovanillic acid
(HVA) for DA. Some studies have also examined actual concentrations of 5-HT
NEUROTRANSMITTER DYSREGULATION 259
and NE, but not DA. Such studies measure transmitter metabolism of the whole
brain and spinal cord and lack any anatomical specificity.
Neuroendocrine and other psychobiological response measures have been
studied following acute challenges with various agents, including amino
acid precursors, e.g., L-tryptophan (L-TRP) and 5-hydroxytryptophan (5HTP)
for 5-HT, presynaptic receptor agonists, e.g., dl-fenfluramine (dl-FEN) or
d-fenfluramine (d-FEN) for 5-HT, postsynaptic receptor agonists, e.g.,
m-chlorophenylpiperazine (m-CPP) for 5-HT, and isoproterenol (ISOP) for NE.
Longer term challenges with receptor antagonists, e.g., antipsychotics for DA
and 5-HT, and antidepressants, especially the serotonin-specific reuptake
inhibitors (SSRIs), also illuminate the role of neurotransmitters in the eating
disorders. Acute amino acid precursor depletion, most notably of L-TRP
(16–19), has been another important source of information about the role of
central 5-HT function in eating and related disorders.
Platelet (PLT) and leukocyte studies are possibly reflective of central
neurotransmitter function but are always at least one step removed from
the nervous system, e.g., platelet 5-HT reuptake, 3H-imipramine binding,
3H-paroxetine binding, platelet monoamine oxidase (MAO), platelet 5-HT
content, as well as platelet receptor-mediated aggregation (5-HT2 and

α-adrenergic).
Plasma concentrations of neurotransmitter precursors, e.g., L-TRP,
Ltyrosine (L-TYR), and their competing large neutral amino acids (LNAAs),
neurotransmitters themselves, e.g., NE, DA, and whole-blood serotonin
(WBS), as well as the usual metabolites, MHPG, HVA, and 5-HIAA.
Brain imaging receptor-binding studies are a promising avenue but remain
relatively unexplored in the eating disorders.
For each neurotransmitter, the results from controlled studies in humans will
be reviewed and summarized for both AN and BN. Where applicable,
comparisons between restricting AN patients, bingeing-purging AN patients,
and normal-weight BN patients will be made. Very little work of this nature has
been done in BED patients but when available will be mentioned.
NOREPINEPHRINE
There are a number of reasons to suspect NE involvement in the eating
disorders. Most notably, NE pathways at the level of the hypothalamus are
known to be involved in the initiation of feeding (20). Disturbances in these
pathways may therefore be involved in the pathophysiology of the profoundly
altered feeding behaviors classically associated with the eating disorders. In
addition, NE’s role in the modulation of mood, anxiety, neuroendocrine control,
metabolic rate, sympathetic tone, and temperature make it a likely candidate for
study (21–26). It has been recognized for some time that lowweight anorexic
patients, and to some degree bulimic patients, have reduced body temperature,
blood pressure, pulse, and metabolic rate (25,27,28). Investigations in this area
have shown that low-weight AN patients have reduced measures of plasma,
urinary, and CSF MHPG (27,29–31). In contrast, reports of plasma NE levels in
the eating disorders has been more variable (32,33), and this appears to be
linked not only to weight but to the stresses associated with the illness (25). AN
patients tend to have higher plasma NE levels at admission, which then decrease
as treatment and weight gain progresses (25,34).
When ill, BN patients demonstrate lower values of plasma NE at
baseline (21,28) and in response to abstinence (35), standing (36), test meal
challenge (37), and mental challenge (37). They also have other evidence of
blunted sympathetic activation in response to mental stress (38). However,
despite low baseline plasma NE levels, BN patients show normal responses to
exercise (39) but reduced responses to orthostasis (40).
In AN patients, depression has been found to be significantly worse in
those patients with the lowest ∆ change in plasma NE concentrations to
orthostasis (41). Reduced urinary MHPG levels have also been related to the
presence of comorbid major depression (29,42). It is therefore important in
such studies to control for psychiatric comorbidity.
Like the plasma NE studies, CSF NE levels have been reported to be no
different in AN patients than controls at low weight and after short-
term weight gain, but then significantly lower after weight recovery of at least
6 months (26,31,32). In BN patients, reduced CSF NE levels have been
reported during the active state of the illness (23,43). However, upon long-term
recovery, concentrations of CSF MHPG have been reported to normalize in
both AN and BN (7) despite earlier reports of lower levels (32). Given that CSF
NE concentrations have not yet been reported in long-term (>1 year) recovered
AN or BN patients, the extent to which adrenergic alterations seen in the eating
disorders are trait related remains unclear. Nevertheless, available evidence
suggests exquisite sensitivity of this system to malnutrition or stress.
Challenge studies using the (β-adrenergic agonist isoproterenol in
underweight anorexic patients revealed erratic secretion of plasma NE in
response to increasing doses (24). Bulimic patients demonstrated significantly
increased chronotropic responses to isoproterenol (44). Challenge studies with
adrenergic agents in recovered patients have not been reported.
The number of platelet α2 receptors has been reported to be reduced in both
AN and BN compared to controls (33,45), suggesting increased postsynaptic
receptor sensitivity that is probably secondary to dieting or semistarvation. In
summary, peripheral and central sympathetic nervous activity is reduced in both
AN and BN, although it tends to normalize with recovery. Taken together, the
preponderance of the evidence so far leads to the conclusion that these changes
are a result of chronic starvation or intermittent dieting (26). However, a trait-
related disturbance of the adrenergic system cannot be ruled out at this
time (35).
Studies of adrenergic receptors on human leukocytes have been another
strategy to investigate adrenergic function in the eating disorders. Buckholtz
et al. (46) reported altered β-adrenergic receptor affinity on circulating
lymphocytes of BN patients compared to those of controls. However, in a
similar study of a mixed group of eating disorder patients, Lonati-Galligani and
Pirke (40) reported lower receptor number (Bmax) but normal affinity (Kd) in
low-weight AN patients, whereas both measures were no different from
controls in the BN patients and the weight-recovered AN patients. Gill and
colleagues (47) reported differential changes in α–and β–adrenoceptor linked
(45Ca2+) uptake in platelets from patients with AN, further documenting an
adrenergic disturbance in eating disorder patients. However, the issue of cause
versus effect remains unanswered in platelet and leukocyte studies.
DOPAMINE
DA is also suspect in the neuropathophysiology of the eating disorders given its
reported involvement in the regulation of feeding, mood, activity, perception,
sexual/social behavior, hormone and peptide release, and to some extent
aggression (48–51). Notably, DA is involved in the hedonic reward responses to
eating and its maintenance as well as to other pleasurable activities (52–54).
The majority of studies of DA metabolism in the eating disorders have
consistently shown that low-weight AN patients have reduced measures of
peripheral and central DA activity, including decreased plasma (27) and CSF
HVA (31). In BN patients, reduced CSF HVA levels also have been reported in
BN patients with frequent binge-purge episodes (23,50) but not in those less
severely ill. Furthermore, binge frequency was inversely correlated with CSF
HVA levels in one study (50). Upon long-term recovery, concentrations of CSF
HVA have been reported to normalize in BN (8), whereas a trend for decreased
CSF HVA levels persisted in six restricting AN patients compared to controls
and to bingeing and/or purging AN patients (7). This suggests a possible trait-
related disturbance specific to restricting AN, although this finding needs
replication given the small sample size. These results could also still be due to
nutritional factors given that patients in this study weighed significantly less than
those in the BN group and may still have been at the low end of the normal
weight range.
Anecdotal reports of the successful use of dopaminergic antagonists (typical
antipsychotic agents) in the treatment of AN patients (55) have been generally
followed by equivocal results in controlled studies (56,57). Atypical

antipsychotic agents may show more promise in the adjunctive treatment of AN
given their combined antidopaminergic and antiserotonergic effects (58–60), but
the results of placebo-controlled studies remain to be seen.
Genetic investigations into the role of DA have been limited to the Bal I
DRD3 receptor polymorphisms in which no differences were found
between AN patients and controls (61). However, the polymorphisms of other
genes coding for DA receptors could be tested. Interestingly, Corcos and
colleagues (62) reported significantly lower IgG and IgM autoantibodies to DA
in BN patients compared to controls. There was also a trend for lower levels of
IgM autoantibodies to DA in the eating-disordered group. The relevance of
these findings to the pathophysiology of the eating disorders remains uncertain
but invokes possible autoimmune mechanisms.
SEROTONIN
Several lines of reasoning point to disturbances of 5-HT function in
the pathophysiology and neuropsychopharmacology of the EDs (8,9,63),
including serotonin’s role in feeding (64,65), satiety (66,67), dieting/
fasting (11,12), mood regulation (16), anxiety (68), obsessive-
compulsiveness/perfectionism/behavioral inhibition (69), harm
avoidance (70,71), impulsivity/aggression (72,73), motor activity (74,75),
gender (76,77), seasonality (66,78,79), body image/perception (80), and social
behavior (81–83) (see Table 1).
Reductions in a variety of 5-HT parameters have been consistently reported
in low-weight AN patients. Although no significant differences have been found
in absolute plasma L-TRP levels (84–86), the plasma L-TRP/LNAA ratio is
reduced in the low weight state (30,87,88) but normalizes upon short-term
weight recovery (22,30). In BN, Gendal and Joyce (89) reported that the
L-TRP/LNAA ratio inversely correlated with the desire to binge-eat. In
addition, symptomatic bulimic relapse or worsening of symptoms has been
reported following acute L-TRP depletion in BN (17–19).
Other significant findings include decreased CSF L-TRP levels (90) and
decreased CSF 5-HIAA levels (22,88,91) during low-weight status with
normalization of these levels with short-term weight recovery (STWR, goal
weight maintenance ≥3 weeks). Strikingly, Kaye and colleagues (69,92) have
reported abnormally elevated CSF 5-HIAA levels following long-term weight
recovery (LTWR, goal weight maintenance ≥6–12 months), and interpret these
findings as indicating that AN may correspond to a primary state of excessive
5-HT tone, which is then masked by malnutrition-induced reductions in 5-HT
activity during active illness. In other words, they propose that the
TABLE 1 Monoamines and the Phenomenology of the Eating Disorders
pathophysiology of AN actually involves a hyperserotonergic trait and,

furthermore, postulate that this trait may correspond to behavioral traits of
obsessionality and inhibition. Corroborating the notion of hyperserotonergic
status in AN, Kaye and colleagues have noted long-term weight-restored
anorexics to display elevated 5-HTla receptor binding, measured by positron
emission tomography (PET) (93).
In BN, reduced levels of CSF 5-HIAA are consistently reported only in the
subgroup of patients displaying more frequent binge-purge episodes (23,50).
Suggesting a possible link to severity of bulimic symptomatology, binge
frequency has been found to correlate inversely with CSF 5-HIAA
concentrations (50). In a small pilot study, Brewerton and colleagues (94) have
reported no difference in CSF 5-HT levels between BN patients and controls.
However, upon recovery for at least a year, BN patients have been reported to
have elevated CSF 5-HIAA levels compared to healthy controls (95), much like
those described earlier as being characteristic of long-term recovered anorexics.
As in AN, this finding has been linked to obsessive-compulsive personality
traits, perfectionism, and behavioral inhibition, associated with a hypothetical
tendency toward hyperserotonergic status. However, we note, that the Kaye
et al. study of recovered BN may be confounded by small weight discrepancies
between their (heavier) recovered bulimics and lighter comparison controls.

Such weight differentials could underlie discrepant levels of 5-HT metabolism.
Decreased prolactin (PRL) responses following m-CPP (96–98),
L-TRP (96,97), and fenfluramine (FEN) (99–101) have been reported in AN
and indicate an anatomically specific alteration in 5-HT receptor sensitivity at the
level of the hypothalamus, which could conceivably also occur in other brain
pathways (9). Blunting of PRL following m-CPP persists into short-term weight
recovery, although trends toward normalization of PRL responses, after
refeeding and weight gain, have been reported (97). With at least a year of
recovery, neurohormonal responses to m-CPP normalize in restricting AN
patients (92). Apparently, full normalization of PRL responsivity to
serotonergic agents occurs after full weight restoration, normalization of
hypothalamic-pituitary-gonadal function, and abatement of overt eating
disorder symptoms (7). However, the appetite-suppressing effect of FEN is
significantly diminished in recovered AN patients despite normalization of
hormonal release(102).
Platelet (PLT) studies contribute to the demonstration of serotonergic
dysfunction in AN. Significant increases/reductions in PLT imipramine
(IMI) binding (103), but not PLT 5-HT uptake (103,104) or PLT MAO
content (42), have been reported in low-weight AN patients. However, a more
recent study reported decreased PLT MAO in AN (105), which was inversely
correlated with impulsivity and positively correlated with persistence (which is
similar to rigidity). In a related vein, Finocchiaro and colleagues (106)
conducted a novel study of indole metabolism and reported altered
phytohemagglutinin stimulated, light-induced [3H]thymidine incorporation into
the DNA of peripheral blood mononuclear leukocytes in AN patients compared
to controls. The authors concluded that the white cells of AN patients show a
failure in the regulation of 5-HT and melatonin metabolism in response to light.
As in AN, neurobiological indices in active BN are often consistent with
reduced 5-HT tone. For example, findings in BN show a consistent pattern of
PRL blunting following m-CPP (107–110), fenfluramine (99,101,111–113),
and 5-hydroxytryptophan (5-HPT) (114), but not L-TRP (9,107).
PRL responses following L-TRP are low only in the BN patients with
concurrent major depression, again emphasizing the need to control for
comorbidity. PRL responses following m-CPP are inversely correlated to
baseline cortisol (CORT) (9). Self-reported binge frequency also has been
shown to be inversely correlated to PRL responses following m-CPP (9) and
fenfluramine (101,111,113) in BN patients. Given that this presumed alteration
in hypothalamic postsynaptic 5-HT functioning normalizes with recovery from
BN (8,95,115), these serotonergic abnormalities could be understood to be a
result of bingeing, purging, and/or dieting rather than a cause of these behaviors,
although other vulnerabilities of the 5-HT system may also exist and interact
with these psychosomatic behaviors. There is only one serotonergic challenge
study reported in BED (101), which found that PRL responses following d-FEN
were no different in patients with BED than in controls. This lends support to
the idea that purging, dieting, and weight loss (rather than bingeing per se) have
greater roles in creating the serotonergic abnormalities noted above. Dieting,
bingeing, and vomiting all may affect central 5-HT synthesis (13,14,22,116,117)
and could conceivably result in down-regulation of postsynaptic 5-HT receptors
and blunted PRL responses. In addition, these behaviors may involve
activation of the HPA axis, which in turn appears to dampen 5-HT receptor
sensitivity (9,107). Despite findings linking recovery from BN to normalization
of blunted endocrine responses after 5-HT agonists (95,115), other findings
(based on PET techniques) suggest persistent reductions in postsynaptic 5-HT2a
receptor activity even in fully recovered bulimics (118). Such findings associate
BN with a stable reduction in 5-HT neurotransmission at some central sites—
and present the possibility that such tendencies exist independently of disorder
sequelae in BN patients.
In BN, platelet studies indicate reduced PLT IMI binding (119) and PLT
MAO (120). PLT 5-HT uptake has been reported to be increased in one
study (121) but not another (120). Steiger et al. (110,122) reported reduced
PLT paroxetine binding in groups of BN patients compared to healthy controls.
Possible Trait-Linked Effects

Independently of dietary factors, personality trait variations might explain some
of the variations in 5-HT status seen in eating disorder sufferers. In noneating-
disordered populations, correspondence between 5-HT function and personality
trait variations has been well established. For example, impul sivity has been
consistently linked to decreased 5-HT activity; suicide, fire setting, violence,
and borderline personality disorder (BPD, for which impulsivity is
pathognomonic) have all been linked to decreased 5-HT metabolism (as
indicated by reduced CSF 5-HIAA) (123,124). Likewise, impulsive suicidality
and aggression have been linked to low platelet 5-HT content and reduced PRL
response to 5-HT agonists (123,124). On the opposite side of the same coin,
findings in non eating-disordered samples have (at least inconsistently)
associated anxiety or compulsivity with increased 5-HT tone. For example,
patients with obsessive-compulsive disorder have been reported to display
elevated CSF 5-HIAA (125) and increased PRL response after the 5-HT agonist
fenfluramine (126). Furthermore, the partial 5-HT agonist m-CPP has been
observed to increase obsessionality in obsessive-compulsive patients, and
anxiety in patients with generalized anxiety disorder (127–129). Likewise,
heightened anxiety has been associated with elevated 5-HT activity in

both generalized anxiety disorder (130) and AN (131). Such findings
have encouraged some theorists to propose that “impulsive” and “compulsive”
traits occupy opposite poles of a continuum of 5-HT under- to
overactivation (132,133). While this notion remains controversial, it is
tempting to contemplate the possibility that 5-HT findings in restrictive versus
bulimic ED variants may reflect variations associated with differential loadings
of compulsive or impulsive traits in these ED subgroups.
In keeping with the notion outlined above, various studies report that
personality trait variations account for variations of 5-HT indices in ED
patients, at least when actively eating disordered. Waller and colleagues (134)
observed that hostile bulimics, compared to less hostile ones (by self report),
showed smaller neuroendocrine responses following buspirone (which they
presumed to be a 5-HTla agonist). Likewise, Carrasco and colleagues (135)
observed systematically lower platelet MAO concentrations (taken as a proxy
for reduced 5-HT activity) in bulimics with impulsive or “borderline” traits.
Results of several studies by Steiger and his colleagues are comparable.
In one study, PRL responses after m-CPP were measured in bulimic
women who reported, or who denied, a history of self-mutilative or suicidal
impulsivity (136). (Incidentally, these two groups of women, were quite
comparable on indices of binge and purge frequency and body mass). Compared
to normal eaters, the self-harming bulimics were clearly blunted, as far as 5-HT
function was concerned; the non-self-harming bulimics were not. In other
words, an association was observed between blunting of the m-CPP-stimulated
PRL response and self-destructiveness, comparable to that obtained in non-
eating-disordered populations (137). This observation suggests that hypoactivity
of the 5-HT system in BN may be more strongly linked to self-aggressive
impulsivity than it is to binge-purge symptoms per se. However, in the study by
Brewerton et al. (107), no such differences were found between bulimic
patients with and without a history of suicidality. Another study by
Steiger’s group examined platelet 3H-paroxetine binding in normal women and
in bulimic women, and assessed effects of “nonplanning impulsivity” (i.e., the
tendency to act without considering consequences) (110). Both bulimic groups
displayed reductions in density (Bmax) of paroxetine binding sites. However, in
bulimics, the extent of reduction in binding site density was inversely correlated
with “nonplanning.” In other words, reduced peripheral 5-HT reuptake
corresponded to increased impulsivity. This effect parallels inverse relationships
noted between platelet 5-HT binding and aggressive impulsivity (138) or self-
mutilation (138) in personality-disordered subjects, raising the notion that in BN
we could be observing a constitutional (trait-linked) susceptibility to
underactivity of the 5-HT system. Furthermore, if exacerbated by effects of
dieting, such susceptibilities could cause certain people to become especially

impulsive and/or prone to binge eating.
Taken together, research findings from plasma, CSF, and pharmacological
challenge studies suggest reduced 5-HT synthesis, uptake, and turnover, as well
as altered postsynaptic 5-HT receptor sensitivity during the active phases of
both AN and BN. Consequently, many reported alterations in 5-HT function
appear to be state dependent, although they may have important biological roles
in the perpetuation of symptoms, particularly mood dysregulation, increased
anxiety, obsessionality, impulsivity, self-aggression, and perhaps the resistance
to and difficulty in learning healthier coping strategies (139).
However, to avoid presenting an oversimplified, unidirectional hypothesis of
5-HT alterations in the eating disorders, it is necessary to note some findings
suggesting heightened 5-HT receptor sensitivity at certain central sites in eating
disorder patients with active symptoms. For example, Brewerton (9) reported
enhanced temperature and migraine headache responses to m-CPP but not
L-TRP in BN patients (regardless of the comorbid presence of AN or
MD) (9,140,141). As discussed in detail elsewhere (141), the enhanced migraine-
like HA responses in the BN patients may indicate enhanced 5-HT2 receptor
sensitivity in CNS vascular tissues. Enhanced 5-HT-mediated platelet
aggregation, a 5-HT2 receptor-mediated phenomenon, has also been reported in
BN (142) and AN (99,112,142) and lends further support to this hypothesis.
The normal cortisol responses following m-CPP and L-TRP in AN and BN are
compatible with this view given the involvement of both 5HT1 (facilitative) and
5-HT2 receptors (inhibitive) in cortisol secretion. These presumed alterations
in 5-HT receptor sensitivity, whether primary or secondary, demonstrate that
5-HT receptor sensitivity can be both decreased and increased in the same
subjects depending on the anatomical location of the receptor as well as the
receptor subtype. We (9,143) have argued in favor of a dysregulation
hypothesis of serotonin dysfunction in the eating and related disorders,
proposing that there is a failure in transmitter regulation in the face of a variety
of psychobiological perturbations potentially affecting monoamine function,
including dieting, fasting, purging, substance abuse, excessive exercising,
medical illnesses, family stresses or losses, sociocultural pressures, traumatic
events, puberty, other developmental tasks/challenges, and changes in the
seasons. Certainly, evidence suggests that a model of neurotransmitter
alterations in the eating disorders stated in terms of a unidirectional (high versus
low activity) concept will not be adequate.
Interest in 5-HT activity in the EDs has led to quite a catalogue of studies on
5-HT system genes—controlling activity of 5-HT receptors, tryptophan
hydroxylase (TPH, the rate-limiting enzyme for 5-HT synthesis), and 5-HT
transporter (reuptake) mechanisms (144). Collier et al. (145) reported a
statistically significant 5-HT2A-1438G/A receptor gene polymorphism in a

group of restricting AN patients compared to healthy controls. This finding has
been replicated in at least two other studies in AN (146,147) as well as in
OCD (147), but not in BN (147). Nacmias et al. (146) reported that other
serotonergic polymorphisms of the 5-HT2a as well as those of the 5HT2c
receptors showed no differences in AN patients compared to controls.
Likewise, no differences between AN patients and controls have been reported
for serotonin transporter gene-linked polymorphisms (5-HTTLPR) (148,149),
tryptophan hydroxylase polymorphisms (150), and 5-HTlDbeta and 5-HT7
gene polymorphisms (151).
For BN, there have been various association studies: Studies on 5-HT2c
polymorphisms in BN detect no syndrome-linked associations (144). Similarly,
three of four available studies on the 5-HT2a receptor gene indicate absence of
association with BN (146,147,152). However, a fourth (in a heterogeneous
anorexic-bulimic sample) associates the 5-HT2a “G” allele with proneness to
bulimic symptoms, borderline personality, and generalized impulsivity (153).
Such findings imply that common genetic factors might mediate concurrence of
bulimic eating patterns and traits of a borderline/impulsive type. Yet another
recent study, first to examine the 5-HT transporter gene (promoter region,
5HTTLPR) in BN, indicates a short-allele variation to confer sevenfold risk of
BN (154). The short (s) allele of 5HTTLPR has been linked to reduced
transcription of 5-HT transporter protein, decreased 5-HT reuptake in
lymphoblasts (155), and traits like suicidality (156), neuroticism, and
impulsivity (157). Preliminary findings from our lab provide a second indication
of relevance of the 5-HT transporter (5-HTT) gene to binge eating and
impulsivity (158). Results in 48 women with binge eating syndromes showed
individuals carrying the short (s) allele of 5HTTLPR (either s/s or s/1 genotypes)
to show more impulsivity and lower density of paroxetine-binding sites than did
long (1) allele homozygotes. These results, if they hold up, would cross-validate
(at a genetic level) a link between impulse control problems and
hyposerotonergic status, indicating convergence among impulsive traits, low
5-HT transporter activity, and the s allele.
Evidence from Pharmacological Effects

It is well known that serotonin-specific antidepressant medications can be
beneficial in controlled studies of BN patients (159) but not in low-weight AN
patients (160,161). More recent data indicate a prophylactic effect of fluoxetine
following weight gain in recovered AN patients (8). SSRIs don’t work during
the low-weight state, presumably because of central depletion of 5-HT and
other monoamines with starvation. There is significantly less 5-HT centrally to

be inhibited by SSRIs.
Finally, recent evidence indicates significant antibulimic responses to 5HT3
antagonists, such as ondansetron (162,163). Although the authors attribute this
therapeutic response to the drug’s ability to reduce vagal tone, the role of the
5-HT3 receptor remains intriguing given its antianxiety effects (164). These
findings open important new arenas for future research involving possible
serotonergic-cholinergic mechanisms, which has been a relatively unexplored
area in the eating disorders.
MAO/ISATIN
Isatin, or tribulin, is an endogenous indole associated with stress, which inhibits
MAO (165). Brewerton et al. (94) reported significantly higher CSF
concentrations of isatin in BN patients compared to healthy controls. There
was also a trend for CSF isatin concentrations to be inversely correlated
with CSF concentrations of the serotonin metabolite 5-HIAA (n = 14,
ρ = –0.51, p = 0.06), although CSF isatin levels were not significantly
correlated with CSF MHPG or HVA. The increase in isatin levels has been
hypothesized to be in response to the resultant monoamine depletion secondary
to the effects of the illness on monoaminergic function. As noted previously,
platelet MAO has been reported to be decreased in BN (120) and in AN (105).
This decrease may represent a compensatory change in response to monoamine
depletion during the active state of the disorders.
RELATIONSHIP TO OTHER SYSTEMS

Neurotransmitter systems do not exist in a vacuum but are exquisitely
interdependent with other brain and body systems and the environment as well.
It is important to think about systems (e.g., 5–HT) and their subsystems
(presynaptic, postsynaptic, receptor subtypes) in the context of larger systems
(brain, environment) and interacting systems/subsystems (e.g., NE, DA,
neurohormones, neuropeptides) with complex feedback and counterfeedback
mechanisms at multiple anatomical levels. An extensive discussion of this rather
far-reaching topic is beyond the scope of this chapter but is discussed in more
detail elsewhere (9).
CONCLUSIONS
Taken together, available findings implicate abnormalities of all monoamine
neurotransmitter systems during the active phases of both AN and BN. Upon
normalization of weight and neurohormonal function, most transmitter

anomalies resolve or atleast improve. Some data show persistent particularities
of the 5–HT system, and suggest that observed tendencies may reflect
psychological traits found in both AN and BN, including obsessionality,
perfectionism, high harm avoidance, and behavioral inhibition, on the one
hand, and recklessness, failure to consider consequences of actions,
selfdestructiveness, and behavioral disinhibition, on the other. Furthermore,
some evidence may be consistent with association between greater behavioral
inhibition and excessive 5–HT activity (at some loci in the system), and
behavioral disinhibition and reduced 5–HT neurotransmission (also at some loci
in the system). The findings in question create a case for the idea that any given
individuals’ 5–HT functioning probably varies in function of constitutionally
determined (latent or manifest) personality trait tendencies. In this light, it is
intriguing to contemplate the ways in which constitutional traits associated with
hypoactivity of the 5–HT system (e.g., impulsivity) may predispose to binge
eating—and traits associated with elevated 5–HT tone (like compulsivity or
harm avoidance) may predispose to dietary restriction.
Some evidence suggests prolonged alterations in NE metabolism, but this is
most likely due to persistent low-grade dietary restraint following recovery.
Preliminary data indicate a DA deficit in restricting AN patients, but this result
remains to be replicated in larger samples. Recent findings also emphasize the
importance of neurotransmitter precursor substrate availability to normal brain
function and especially to the process of recovery from an eating disorder.
Future research directions will include further exploration of neurotransmitter-
related gene candidates, in vivo receptor imaging studies, and improved
psychopharmacological interventions based on biological alterations
characteristic of the different stages and features of these dangerous disorders.
REFERENCES
Disorders, 4th ed. Washington, DC: American Psychiatric Press, 1994.
2. Becker AE, Grinspoon SK, Klibanski A, Herzog DB. Eating disorders. N Engl J
Med 1999; 340:1092–1098.
3. Walsh BT, Devlin MJ. Eating disorders: progress and problems. Science 1998; 280:
1387–1390.
4. Lilenfeld LR, Kaye WH, Greeno CG, Merikangas KR, Plotnicov K, Pollice C,
Radhika R, Strober M, Bulik C, Nagy L. A controlled family study of anorexia
nervosa and bulimia nervosa. Arch Gen Psychiatry 1998; 55:603–610.
5. Strober M, Freeman R, Lampert C, Diamond J, Kaye WH. Controlled family
study of anorexia nervosa and bulimia nervosa: evidence of shared liability and
transmission of partial syndromes. Am J Psychiatry 2000; 157:393–401.
6. Halmi KA, Sunday SR, Strober M, Kaplan A, Woodside DB, Fichter M, Treasure
J, Berrettini WH, Kaye WH. Perfectionism in anorexia nervosa: variation by
clinical subtype, obsessionality, and pathological eating behavior. Am J Psychiatry
2000; 157:1799–1805.
7. Kaye W, Strober M, Stein D, Gendall K. New directions in treatment research of
anorexia and bulimia nervosa. Biol Psychiatry 1999b; 45:1285–1292.
8. Kaye WH, Gendall KA, Strober M. Serotonin neuronal function and selective
reuptake inhibitor treatment in anorexia nervosa and bulimia nervosa. Biol
Psychiatry 1998a; 44:825–838.
9. Brewerton TD. Toward a unified theory of serotonin dysregulation in eating and
related disorders. Psychoneuroendocrinology 1995; 20:561–590.
10. Pollice C, Kaye WH, Greeno CG, Weltzin TE. Relationship of depression, anxiety,
and obsessionality to state of illness in anorexia nervosa. Int J Eat Disord 1997; 21:
367–376.
11. Cowen PJ, Clifford EM, Walsh AE, Williams C, Fairburn CG. Moderate dieting
causes 5-HT2c receptor supersensitivity. Psychol Med 1996; 26:1155–1159.
12. Cowen PJ, Smith KA. Serotonin, dieting, and bulimia nervosa. Adv Exp Med Biol
1999; 467:101–104.
13. Goodwin GM, Fairburn CG, Cowen PJ. Dieting changes serotonergic function in
women, not men: implications for the etiology of anorexia nervosa. Psychol Med
1987a; 17:839–842.
14. Goodwin GM, Fairburn CG, Cowen PJ. The effects of dieting and weight loss
upon neuroendocrine responses to tryptophan, clonidine and apomorphine in
volunteers: important implications for neuroendocrine investigations in depression.
Arch Gen Psychiatry 1987b; 44:952–957.
15. Goodwin GM, Cowen PJ, Fairburn CG, Parry-Billings M, Calder PC, Newsholme
EA. Plasma concentrations of tryptophan and dieting. Br Med J 1990; 300:
1499–1500.
16. Deldago PL, Charney DS, Price LH, Aghajanian GK, Landis H, Henninger GR.
Serotonin function and the mechanism of antidepressant action: reversal of
antidepressant-induced remission by rapid depletion of plasma tryptophan. Arch
2000; 47:151–157.
19. Weltzin TE, Fernstrom MH, Fernstrom JD, Neuberger SK, Kaye WH. Acute
tryptophan depletion and increased food intake and irritability in bulimia nervosa.
Am J Psychiatry 1995; 152:1668–1671.
20. Rowland NE, Morien A, Li BH. The physiology and brain mechanisms of feeding.
Nutrition 1996; 12:626–639.
21. Jimerson DC, George DT, Kaye W, Brewerton TD, Goldstein DS.
Norepinephrine regulation in bulimia. In: Hudson JI, Pope HG, eds. Psychobiology
of Bulimia. Washington, DC: American Psychiatric Press, 1987: 145–156.
22. Kaye WH, Gwirtsman HE, George DT, Jimerson DC, Ebert MH. CSF 5HIAA
concentrations in anorexia nervosa: reduced values in underweight subjects
normalize after weight gain. Biol Psychiatry 1988b; 23:102–105.
23. Kaye WH, Ballenger JC, Lydiard RB, Stuart GW, Laraia MT, O’Neil P, Fossey
MD, Stevens V, Lesser S, Hsu G. CSF monoamine levels in normalweight bulimia:
evidence for abnormal noradrenergic activity. Am J Psychiatry 1990; 147:225–229.
24. Kaye WH, George DT, Gwirtsman HE, Jimerson DC, Goldstein DS, Ebert MH,
Lake CR. Isoproterenol infusion test in anorexia nervosa: assessment of pre- and
post-beta-noradrenergic receptor activity. Psychopharmacol Bull 1990; 26:
355–359.
25. Lesem MD, George DT, Kaye WH, Goldstein DS, Jimerson DC. State-related
changes in norepinephrine regulation in anorexia nervosa. Biol Psychiatry 1989; 25:
509–512.
26. Pirke KM. Central and peripheral noradrenalin regulation in eating disorders.
27. Gross HA, Lake CR, Ebert MH, Ziegler MG, Kopin IJ. Catecholamine metabolism
in primary anorexia nervosa. J Clin Endocrinol Metab 1979; 49:805–809.
28. Obarzanek E, Lesem MD, Goldstein DS, Jimerson DC. Reduced resting
metabolicrateinpatientswithbulimianervosa.ArchGenPsychiatryl991;48:456^t62.
29. Halmi KA, Dekirmenjian H, Dav JM, Casper R, Goldberg S. Catecholamine
30. Johnston JL, Leiter LA, Burrow GN, Garfinkel PE, Anderson GH. Excretion of
urinary catecholamine metabolites in anorexia nervosa: effect of body composition
and energy intake. Am J Clin Nutr 1984; 40:1001–1006.
31. Kaye WH, Ebert MH, Raleigh M, Lake CR. Abnormalities in CNS monoamine
32. Kaye WH, Jimerson DC, Lake CR, Ebert MH. Altered norepinephrine metabolism
following long-term weight recovery in patients with anorexia nervosa. Psychiatry
Res 1985; 14:333–342.
33. Luck P, Mikhailid DP, Dashwood MR, Barradas MA, Sever PS, Dandona P,
Wakeling A. Platelet hyperaggregability and increased alpha-adrenoceptor density
in anorexia nervosa. J Clin Endocrinol Metab 1983; 57:911–914.
34. Pahl J, Pirke KM, Schweiger U, Warnhoff M, Gerlinghoff M, Brinkmann W,
Berger M, Krieg C. Anorectic behavior, mood, and metabolic and endocrine
adaptation to starvation in anorexia nervosa during inpatient treatment. Biol
Psychiatry 1985; 20:874–887.
35. Kaye WH, Gwirtsman HE, George DT, Jimerson DC, Ebert MH, Lake CR.
Disturbances of noradrenergic systems in normal weight bulimia: relationship to
diet and menses. Biol Psychiatry 1990; 27:4–21.
36. Pirke KM, Jorg P, Schweiger U, Warnhoff M. Metabolic and endocrine indices of
starvation in bulimia: a comparison with anorexia nervosa. Psychiatry Res 1985; 15:
33–39.
37. Pirke KM, Kellner M, Philipp E, Laessle R, Krieg JC, Fichter MM. Plasma
norepinephrine after a standardized test meal in acute and remitted patients with
anorexia nervosa and in healthy controls. Biol Psychiatry 1992; 31:1074–1077.
38. Koo-Loeb JH, Pedersen C, Girdler SS. Blunted cardiovascular and catecholamine
stress reactivity in women with bulimia nervosa. Psychiatry Res 1998; 80:13–27.
39. Pirke KM, Eckert M, Ofers B, Goebl G, Spyra B, Schweiger U, Tuschl RJ, Fichter
MM. Plasma norepinephrine response to exercise in bulimia, anorexia nervosa, and
controls. Biol Psychiatry 1989; 25:799–802.
40. Lonati-Galligani M, Pirke KM. Beta 2-adrenergic receptor regulation in circulating
mononuclear leukocytes in anorexia nervosa and bulimia. Psychiatry Res 1986; 19:
189–198.
41. Laessle RG, Schweiger U, Pirke KM. Mood and orthostatic norepinephrine
response in anorexia nervosa. Psychiatry Res 1988; 24:87–94.
42. Biederman J, Herzog DB, Rivinus TM, Ferber RA, Harper GP, Onsulak PJ,
Schildkrautt JJ. Urinary MHPG in anorexia nervosa patients with and without a
concomitant major depressive disorder. J Psychiatr Res 1984; 18:149–160.
43. Kaye WH, George DT, Gwirtsman HE, Jimerson DC, Goldstein DS, Ebert MH,
Lake CR. Isoproterenol infusion test in anorexia nervosa: assessment of pre- and
post-beta-noradrenergic receptor activity. Psychopharmacol Bull 1990; 26:
355–359.
44. George DT, Kaye WH, Goldstein DS, Brewerton TD, Jimerson DC. Altered
norepinephrine regulation in bulimia: effects of pharmacological challenge with
isoproterenol. Psychiatry Res 1990; 33:1–10.
45. Heufelder A, Warnhoff M, Pirke KM. Platelet alpha 2-adrenoceptor and adenylate
cyclase in patients with anorexia nervosa and bulimia. J Clin Endocrinol Metab
1985; 61:1053–1060.
46. Buckholtz NS, George DT, Davies AO, Jimerson DC, Potter WZ. Lymphocyte
beta-adrenergic receptor modification in bulimia. Arch Gen Psychiatry 1988; 45:
479–182.
47. Gill J, DeSouza V, Wakeling A, Dandona P, Jeremy JY. Differential changes in
alpha- and beta-adrenoceptor linked [45Ca2 +] uptake in platelets from patients
with anorexia nervosa. J Clin Endocrinol Metab 1992; 74:441–446.
48. Engstrom G, Alling C, Blennow K, Regnell G, Traskman-Bendz L.
Reduced cerebrospinal HVA concentrations and HVA/5-HIAA ratios in suicide
attempters: monoamine metabolites in 120 suicide attempters and 47 controls. Eur
Neuropsychopharmacol 1999; 9:399–405.
49. Hoebel BG. Brain neurotransmitters in food and drug reward. Am J Clin Nutr
1985; 42:1133–1150.
50. Jimerson DC, Lesem MD, Kaye WH, Brewerton TD. Low serotonin and
dopamine metabolite concentrations in CSF from bulimic patients with frequent
binge episodes. Arch Gen Psychiatry 1992; 49:132–138.
51. Kaye WH, Guido KWF, Frank GK, McConaha C. Altered dopamine activity after
recovery from restricting anorexia nervosa. Neuropsychopharmacology 1999; 21:
503–506.
52. Schultz W. Reward signaling by dopamine neurons. Neuroscientist 2001; 7:
293–302.
53. Hoebel BG, Hernandez L, Schwartz DH, Mark P, Hunter GA. Microdialysis
studies of brain norepinephrine, serotonin, and dopamine release during ingestive
behavior. The Psychobiology of Human Eating Disorders. Ann N Y Acad Sci 1989;
575:71–193.
54. Dayan P, Balleine BW. Reward, motivation, and reinforcement learning. Neuron
2002; 36:285–298.
55. Dally P, Sargant W. Treatment and outcome of anorexia nervosa. Br Med J 1966;
2:793–795.
56. Vandereycken W, Pierloot R. Pimozide combined with behavior therapy in the short-
term treatment of anorexia nervosa. A double-blind placebo-controlled cross-over
study. Acta Psychiatr Scand 1982; 66:445–450.
57. Vandereycken W. Neuroleptics in the short-term treatment of anorexia nervosa: a
double-blind placebo-controlled, cross-over trial with sulpride. Br J Psychiatry
1984; 144:288–292.
58. Hansen L. Olanzapine in the treatment of anorexia nervosa. Br J Psychiatry 1999;
175:592.
59. Jensen VS, Mejlhede A. Anorexia nervosa: treatment with olanzapine. Br J
Psychiatry 2000; 177:87.
60. LaVia M, Gray N, Kaye WH. Case reports of olanzapine treatment of anorexia
61. Bruins-Slot L, Gorwood P, Bouvard M, Blot P, Ades J, Feingold J, Schwartz JC,
Mouren-Simeoni MC. Lack of association between anorexia nervosa and D3
dopamine receptor gene. Biol Psychiatry 1998; 43:76–78.
62. Corcos M, Atger F, Levy-Soussan P, Avrameas S, Guilbert B, Cayol V, Jeammet P.
Bulimia nervosa and autoimmunity. Psychiatry Res 1999; 87:77–82.
63. Kaye WH, Weltzin TE. Serotonin activity in anorexia and bulimia nervosa:
relationship to the modulation of feeding and mood. J Clin Psychiatry 1991; 52
(suppl):41–48.
64. Dourish CT, Cooper SJ, Gilbert F, Coughlan J, Iversen SD. The 5-HT1A agonist
8-OH-DPAT increases consumption of palatable wet mash and liquid diets in the
rat. Psychopharmacology 1988; 94:58–63.
65. De Vry J, Schreiber R. Effects of selected serotonin 5-HT(l) and 5-HT(2) receptor
agonists on feeding behavior: possible mechanisms of action. Neurosci Biobehav
Rev 2000; 24:341–353.
66. Brewerton TD, Murphy DL, Jimerson DC. Testmeal responses following m-
chlorophenylpiperazine and L-tryptophan in bulimics and controls.
67. Leibowitz SF, Alexander JT. Hypothalamic serotonin in control of eating behavior,
meal size, and body weight. Biol Psychiatry 1998; 44:851–864.
68. Anderson IM, Mortimore C. 5-HT and human anxiety: evidence from studies using
acute tryptophan depletion. Adv Exp Med Biol 1999; 467:43–55.
69. Kaye WH, Gwirtsman HE, George DT, Ebert MH. Altered serotonin activity in
anorexia nervosa after long-term weight restoration. Arch Gen Psychiatry 1991a;
48:556–562.
70. Brewerton TD, Hand LD, Bishop ER. The Tridimensional Personality
Questionnaire in eating disorder patients. Int J Eat Disord 1993; 14:213–218.
71. Waller DA, Gullion CM, Petty F, Hardy BW, Murdock MV, Rush AJ.
Tridimensional Personality Questionnaire and serotonin in bulimia nervosa.
72. Linnoila M, Virkhunen M, Scheinin M, Nuutila A, Rimon R, Goodwin FK. Low
cerebrospinal fluid 5-hydroxyindoleacetic acid concentration differentiates
impulsive from nonimpulsive violent behavior. Life Sci 1983; 33:2609–2614.
73. Coccaro EF, Siever LJ, Klar H, Maurer G, Cochrane K, Cooper TB, Mohr RC,
Davis KL. Serotonergic studies in affective and personality disorder patients:
correlates with suicidal and impulsive aggressive behavior. Arch Gen Psychiatry
1989; 46:587–599.
74. Brewerton TD, Stellefson EJ, Hibbs N, Hodges EJ, Cochrane CE. A comparison of
eating disorder patients with and without compulsive exercising. Int J Eat Disord
1995; 17:413–416.
75. Epling F, Pierci D. Activity-based anorexia: a biological perspective. Int J Eat
Disord 1988; 7:475–485.
76. Carlsson M, Svensson K, Eriksson E, Carlsson A. Rat brain serotonin: biochemical
and functional evidence for a sex difference. J Neural Transm 1985; 63:297–313.
77. Goodwin GM, Fraser S, Stump K, Fairburn CG, Elliott JM, Cowen PJ. Dieting
and weight loss in volunteers increases the number of alpha2adrenoceptors and
5-HT receptors on blood platelets without effect on [3H]imipramine binding.
J Affect Disord 1987; 12:267–274.
78. Brewerton TD. Seasonal variation of serotonin function in humans: research and
clinical implications. Ann Clin Psychiatry 1989; 1:153–164.
79. Brewerton TD, Berrettini W, Nurnburger J, Linnoila M. An analysis of seasonal
fluctuations of CSF monoamines and neuropeptides in normal controls: findings
with 5-HIAA and HVA. Psychiatry Res 1988; 23:257–265.
80. Goldbloom DS, Olmsted MP. Pharmacotherapy of bulimia nervosa with fluoxetine:
assessment of clinically significant attitudinal change. Am J Psychiatry 1993; 50:
770–774.
81. Raleigh MJ, McGuire MT, Brammer GL, Yuwiler A. Social and
environmental influences on blood serotonin concentrations in monkeys. Arch Gen
Psychiatry 1984; 41:405–410.
82. Raleigh MJ, Brammer GL, McGuire MT, Yuwiler A. Dominant social status
facilitates the behavioral effects of serotonergic agonists. Brain Res 1985; 348:
274–282.
83. McQuire MT, Raleigh MJ. Serotonin-behavior interactions in vervet monkeys.
84. Russell GF. The nutritional disorder in anorexia nervosa. J Psychosom Res 1967;
11:141–149.
85. Coppen AJ, Gupta RK, Eccleston EG, Wood KM, Wakeling A, de Sousa VF.
Plasma tryptophan in anorexia nervosa. Lancet 1976; 1:961.
86. Hassanyeh F, Marshall EF. Measures of serotonin metabolism in anorexia nervosa.
Acta Psychiatr Scand 1991; 84:561–563.
87. Askenazy F, Candito M, Caci H, Myquel M, Chambon P, Darcourt G, Puech AJ.
Whole blood serotonin content, tryptophan concentrations, and impulsivity in
anorexia nervosa. Biol Psychiatry 1998; 43:188–195.
metabolism between nonbulimic and bulimic patients with anorexia nervosa. Am J
Psychiatry 1984; 141:1598–1601.
89. Gendall KA, Joyce PR. Meal-induced changes in tryptophan:LNAA ratio: effects on
craving and binge eating. Eat Behav 2000; 1:53–62.
90. Gerner RH, Cohen DJ, Fairbanks L, Anderson GM, Young JG, Scheinin M,
Linnoila M, Shaywitz BA, Hare TA. CSF neurochemistry of women with anorexia
nervosa and normal women. Am J Psychiatry 1984; 141:948–949.
91. Gillberg C. Low dopamine and serotonin levels in anorexia nervosa. Am J
Psychiatry 1983; 140:948–949.
92. Kaye WH. Persistent alterations in behavior and serotonin activity after recovery
from anorexia and bulimia nervosa. Ann N Y Acad Sci 1997; 817:162–178.
93. Kaye W, Frank G. Gene-environment interactions: Brain and behavior in anorexia
nervosa. Paper presented at the annual meeting of the Eating Disorder Research
Society, November 20–23, 2002, Charleston, South Carolina.
94. Brewerton TD, Zealberg JL, Lydiard RB, Glover V, Sandler M, Ballenger JC. CSF
isatin is elevated in bulimia nervosa. Biol Psychiatry 1995; 37:481–483.
95. Kaye WH, Greeno CG, Moss H, Fernstrom J, Fernstrom M, Lilenfeld LR,
Weltzin TE, Mann JJ. Alterations in serotonin activity and psychiatric symptoms
after recovery from bulimia nervosa. Arch Gen Psychiatry 1998; 55:927–935.
96. Brewerton TD, Brandt HA, Lesem DT, Murphy DL, Jimerson DC. Serotonin in
eating disorders. In: Coccaro E, Murphy D, eds. Serotonin in Major Psychiatric
Disorders. Washington, DC: American Psychiatric Press, 1990:153–184.
97. Brewerton TD, Jimerson DC. Studies of serotonin function in anorexia nervosa.
98. Hadigan CM, Walsh BT, Buttinger C, Hollander E. Behavioral and
neuro endocrine responses to metaCPP in anorexia nervosa. Biol Psychiatry 1995;
37:504–511.
99. Halmi KA, McBride PA, Sunday SR. Serotonin responsivity and hunger and satiety
in eating disorders. Primary and Secondary Eating Disorders: A
Psychoneuroendocrine and Metabolic Approach. Proceedings of the 2nd
International Symposium on Disorders of Eating Behaviour, Pavia, Italy, September
15–19, 1992:123–131.
100. Monteleone P, Brambilla F, Bortolot F, La Rocca A, Maj M. Prolactin response to

d-fenfluramine blunted in people with anorexia nervosa. Br J Psychiatry 1998; 172:
439–442.
101. Monteleone P, Brambilla F, Bortolotti G, Maj M. Serotonergic dysfunction across
the eating disorders: relationship to eating behaviour, purging behaviour,
nutritional status and general psychopathology. Psychol Med 2000; 30:1099–1110.
102. Ward A, Brown N, Lightman S, Campbell IC, Treasure S. Neuroendocrine,
appetitive and behavioural responses to d-fenfluramine in women recovered from
103. Weizman R, Carmi M, Tyano S, Apter A, Rehavi M. High affinity [3H]imipramine
binding and serotonin uptake to platelets of adolescent females suffering from
anorexia nervosa. Life Sci 1986b; 38:1235–1242.
104. Zemishlany Z, Modai I, Apter A, Jerushalmy Z, Samuel E, Tyano S. Serotonin
(5-HT) uptake by blood platelets in anorexia nervosa. Acta Psychiatr Scand 1987;
75:127–130.
105. Diaz-Marsa M, Carrasco JL, Hollander E, Cesar J, Saiz-Ruiz J. Decreased platelet
monoamine oxidase activity in female anorexia nervosa. Acta Psychiatr Scand2000;
101:226–230.
106. Finocchiaro LM, Polack E, Nahmod VE, Glikin GC. Cultured peripheral blood
mononuclear leukocytes from anorexia nervosa patients are refractory to visible
light. Life Sci 1995; 57:559–569.
107. Brewerton TD, Mueller EA, Lesem MD, Brandt HA, Quearry B, George DT,
Murphy DL, Jimerson DC. Neuroendocrine responses to m-chlorophenylpi-
perazine and L-tryptophan in bulimia. Arch Gen Psychiatry 1992b; 49:852–861.
108. Levitan RD, Kaplan AS, Joffe RT, Levitt AJ. Hormonal and subjective responses to
intravenous meta-chlorophenylpiperazine in bulimia nervosa. Arch Gen Psychiatry
1997; 54:521–527.
109. Steiger H, Gauvin L, Israel M, Koerner N, Ng Ying Kin NMK, Paris K, Young SN.
110. Steiger H, Young SN, Kin NM, Koener N, Israel M, Lageix P, Paris J. Implications
of impulsive and affective symptoms for serotonin function in bulimia nervosa.
Psychol Med 2001; 31:85–95.
111. Jimerson DC, Wolfe BE, Metzger ED, Finkelstein DM, Cooper TB, Levine JM.
Decreased serotonin function in bulimia nervosa. Arch Gen Psychiatry 1997; 54:
529–534.
112. McBride PA, Anderson GM, Khait VD, Sunday SR, Halmi KA. Serotonergic
responsivity in eating disorders. Psychopharmacol Bull 1991; 27:365–372.
113. Monteleone P, Brambilla F, Bortolot F, Ferraro C, Maj M. Plasma prolactin
response to D-fenfluramine blunted in bulimic patients with frequent binge
episodes. Psychol Med 1998; 28:975–983.
114. Goldbloom DS, Garfinkel PE, Katz R, Brown GM. The hormonal response to
intravenous 5-hydroxytryptophan in bulimia nervosa. J Psychosom Res 1996; 40:
289–297.
115. Wolfe BE, Metzger ED, Levine JM, Finkelstein DM, Cooper TB, Jimerson DC.
Serotonin function following remission from bulimia nervosa.
116. Fernstrom JD. Dietary effects on brain serotonin synthesis: relationship to appetite
regulation. Am J Clin Nutr 1985; 42:1072–1082.
117. Kaye WH, Gwirtsman HE, Brewerton TD, George DT, Jimerson DC, Wurtman
RJ. Bingeing behavior and plasma amino acids: a possible involvement of brain
serotonin in bulimia. Psychiatry Res 1988; 23:31–43.
118. Kay WH, Frank GK, Meltzer CM, Price JC, McConaha CW, Crossan PJ, Klump
K, Rhodes L. Altered serotonin 2A receptor activity in women who have recovered
from bulimia nervosa. Am J Psychiatry 2001; 158:1152–1155.
119. Marazziti D, Macchi E, Rotondo A, Placidi GF, Cassano GB. Involvement of
serotonin system in bulimia. Life Sci 1988; 43:2123–2126.
120. Hallman J, Sakurai E, Oreland L. Blood platelet monoamine oxidase activity,
serotonin uptake and release rates in anorexia and bulimia patients and in healthy
controls. Acta Psychiatr Scand 1989; 81:73–77.
121. Goldbloom DS, Hicks LK, Garfinkel PE. Platelet serotonin uptake in bulimia
nervosa. Biol Psychiatry 1988; 28:644–647.
122. Steiger H, Leonard S, Kin NY, Ladouceur C, Ramdoyal D, Young SN. Childhood
abuse and platelet tritiated-paroxetine binding in bulimia nervosa:
implicationsofborderlinepersonalitydisorder.JClinPsychiatry2000;61:428–435.
123. Asberg M, Schalling D, Träskman-Bendz L, Wägner A. Psychobiology of
suicide, impulsivity and related phenomena. In: Meltzer HY, ed.
Psychopharmacology:Third Generation of Progress. NewYork: Raven Press, 1987:
655–688.
124. Coccaro EF, Siever LJ, Klar HM, Cochrane K, Cooper TB, Mohs RC, Davis KL.
Serotonergic studies in patients with affective and personality disorders: correlates
with suicidal and impulsive aggressive behaviour. Arch Gen Psychiatry 1989; 46:
587–599.
125. Swedo SE, Leonard HL, Krusei MJP, Rettew DC, Listwak SJ, Berrettini W,
Stipetic M, Hamburger S, Gold PW, Potter WZ, Rapoport JL. Cerebrospinal fluid
neurochemistry in children and adolescents with obsessive-compulsive disorder.
126. Fineberg NA, Roberts A, Montgomery SA, Cowen PJ. Brain 5-HT function in
obsessive-compulsive disorder. Prolactin responses to d-fenfluramine. Br J
Psychiatry 1998; 171:280–282.
127. Altemus M, Swedo SE, Leonard HL, Richter D, Rubinow DR, Potter WZ,
Rapaport JL. Changes in cerebrospinal fluid neurochemistry during treatment of
obsessive-compulsive disorder with clomipramine. Arch Gen Psychiatry 1994; 51:
846–849.
128. Germine RH, Goddard AW, Woods SW, Charnet DS, Henninger GR. Anger and
anxiety response to ra-chlorophenylpiperazine in generalized anxiety disorder. Biol
Psychiatry 1992; 32:457–461.
129. Mundo E, Bellodi L, Smeraldy E. Effects of acute intravenous clominpramine in

obsessive-compulsive symptoms and response to chronic treatment. Biol Psychiatry
1995; 28:525–531.
130. Garvey MJ, Nowyes R Jr, Woodman C, Laukes C. Relationship of generalized
anxiety symptoms to urinary 5-hydroxyindoleacetic acid and vanyllylmandelic acid.
Psychol Res 1995; 57:1–5.
131. Askenazy F, Candito M, Caci H, Myquel M, Chambon P, Darcourt G, Puech AJ.
Whole blood serotonin content, tryptophan concentrations, and impulsivity in
anorexia nervosa. Biol Psychiatry 1998; 43:188–195.
132. Cloninger CR, Svrakic DM, Przybeck TR. A psychobiological model of
temperament and character. Arch Gen Psychiatry 1993; 50:975–990.
133. Hollander E. Treatment of obsessive-compulsive spectrum disorder with SSRIs. Br
J Psychiatry 1998; 35(suppl):7–12.
134. Waller DA, Sheinberg AL, Gullion C, Moeller FG, Cannon DS, Petty F, Hardy
BW, Orsulak P, Ruch AJ. Impulsivity and neuroendocrine response to buspirone in
bulimia nervosa. Biol Psychiatry 1996; 39:371–374.
135. Carrasco JL, Diaz-Marsa M, Hollander E, Cesar J, Saiz-Ruiz J. Decreased platelet
monoamine oxidase activity in female bulimia nervosa. Eur Neuropsychopharmacol
2000; 10:113–117.
136. Steiger H, Koerner NM, Engleberg M, Israel M, Ng Ying Kin NMK, Young SN. Self-
destructiveness and serotonin function in bulimia nervosa. Psychiatry Res2001;
103:15–26.
137. Coccaro EF, Kavoussi RJ, Sheline YI, Lish JD, Cszwenansky JG. Impulsive
aggression in personality disorder correlates with tritiated paroxetine binding in the
platelet. Arch Gen Psychiatry 1996; 53:531–536.
138. Simeon D, Stanley B, Frances A, Mann JJ, Winchel R, Stanley M. Selfmutilation in
personality disorders: psychological and biological correlates. Am J Psychiatry
1992; 149:221–226.
139. Riedel WJ, Klaassen T, Deutz NE, van Someren A, van Praag HM. Tryptophan
depletion in normal volunteers produces selective impairment in memory
consolidation. Psychopharmacology 1999; 141:362–369.
140. Brewerton TD, Murphy DL, Mueller EA, Jimerson DC. The induction of migraine-
like headaches by the serotonin agonist, w-chlorophenylpiperazine. Clin Pharmacol
Ther 1988; 43:605–609.
141. Brewerton TD, Murphy DL, Lesem MD, Brandt HA, Jimerson DC. Headache
responses to m-chlorophenylpiperazine and L-tryptophan in bulimia nervosa.
Headache 1992; 32:217–222.
142. Spigset O, Andersen T, Hagg S, Mjondal T. Enhanced platelet serotonin
5HT2A receptor binding in anorexia nervosa and bulimia nervosa. Eur
Neuropsychopharmacol 1999; 9:469–473.
143. Steiger H. Eating disorders and the serotonin connection: states, traits and
developmental effects. J Psychiatry Neurosci. In press.
144. Hinney A, Remschimidt H, Hebebrand J. Candidate gene polymorphisms in eating
disorders. Eur J Pharmacol 2000; 410:147–159.
145. Collier DA, Arranz MJ, Mupita D, Brown N, Treasure J. Association between the
5-HT2A receptor gene polymorphism and anorexia nervosa. Lancet 1997; 350:
412.
gene polymorphisms in anorexia nervosa and bulimia nervosa. Neurosci Lett 1999;
277:134–136.
5-HT2A promoter polymorphism—1438G/A, anorexia nervosa, and
obsessivecompulsive disorder. Lancet 1998; 351:1785.
148. Hinney A, Barth N, Ziegler A, vonPrittwitz S, Hamann A, Hennighausen K, Pirke
KM, Heils A, Rosenkranz K, Roth H, Coners H, Mayer H, Herzog W, Siegfried
A, Lehmkuhl G, Poustka F, Schmidt MH, Schafer H, Grzeschik KH, Lesch KP,
Lentes KU, Remschmidt H, Hebebrand J. Serotonin transporter gene-linked
polymorphic region, allele distributions in relationship to body weight and in
anorexia nervosa. Life Sci 1997; 61:295–303.
149. Sundaramurthy D, Pieri LF, Gape H, Markham AF, Campbell DA. Analysis of
serotonin transporter gene linked polymorphism (5-HTTLPR) in anorexia nervosa.
Am J Med Genet 2000; 96:53–55.
150. Han L, Nielsen DA, Rosenthal NE, Jefferson K, Kaye W, Murphy D, Altemus M,
Humphries J, Cassano G, Rotondo A, Virkkimen M, Linnoila M, Goldman D. No
coding variant of the tryptophan hydroxylase gene detected in seasonal affective
disorder, obsessive-compulsive disorder, anorexia nervosa, and alcoholism. Biol
Psychiatry 1999; 45:615–619.
151. Hinney A, Herrmann H, Lohr T, Rosenkranz K, Ziegler A, Lehmkuhl G, Poustka
F, Schmidt MH, Mayer H, Siegfried W, Remschmidt H, Hebebrand J. No evidence
for an involvement of alleles of polymorphisms in the serotonin1Dbeta and 7
receptor genes in obesity, underweight or anorexia nervosa. Int J Obes Relat
Metab Disord 1999; 23:760–763.
152. Ziegler A, Hebebrand J, Gorg R, Rosenkranz K, Fichter MM, HerpertzDahlmann
B, Remschimidt H, Hinney A. Further lack of association between the 5-HT2A
gene promoter polymorphism and susceptibility to eating disorders and a meta-
analysis pertaining to anorexia nervosa. Mol Psychiatry 1999; 4:410–412.
153. Nishiguchi N, Matsuchita S, Suzuki K, Murayama M, Shirakawa O, Higuchi S.
eating disorders in Japanese patients. Biol Psychiatry 2001; 50:123–128.
154. Di Bella DD, Catalano M, Cavallini MC, Riboldi C, Bellodi I. Serotonin
transporter linked polymorphic region in anorexia nervosa and bulimia nervosa.
Mol Psychiatry 2000; 5:233–241.
155. Heils A, Teufel A, Petri S, Stober G, Riederer P, Bengel D. Lesch KP: Allelic
variations of human serotonin transporter gene expression. J Neurochem 1996; 66:
2621–2624.
156. Bondy B, Erfurth A, de Jonge S, Kriiger M, Meyer H. Possible association of the
short allele of the serotonin transporter promoter gene polymorphism (5HTTLPR)
with violent suicide. Mol Psychiatry 2000; 5:193–195.
157. Lesch KP, Wolozin BL, Murphy DL, Reiderer P. Primary structure of the human
platelet serotonin uptake site: identity with the brain serotonin transporter. J
Neurochem 1993; 60:2319–2322.
158. Steiger H, Joober R, Israel M, Bruce K, NG Ying Kin NMK, Gauvin L, Young SN,
Joncas J, Torkaman-Xehi A. A polymorphism in the promoter region of the
serotonin transporter gene (5-HTTLPR) corresponds to impulsivity and reduced
paroxetine binding in eating-disordered and normal-eater woman. Presented at the
Annual Meeting of the Eating Disorders Research Society, Charleston, South
Carolina, November 23, 2002.
159. Fluoxetine Bulimia Nervosa Collaborative Group. Fluoxetine in the treatment of
bulimia nervosa: a multicenter, placebo-controlled, double-blind trial. Arch Gen
Psychiatry 1992; 49:139–147.
160. Attia E, Haiman C, Walsh BT, Flater SR. Does fluoxetine augment the inpatient
treatment of anorexia nervosa? Am J Psychiatry 1998; 155:548–551.
161. Strober M, Pataki C, Freeman R, DeAntonio M. No effect of adjunctive fluoxetine
on eating behavior or weight phobia during the inpatient treatment of anorexia
nervosa: an historical case-control study. J Child Adol Psychopharmacol 1999; 9:
195–201.
162. Faris PL, Kim SW, Meller WH, Goodale RL, Hofbauer RD, Oakman SA, Howard
LA, Stevens ER, Eckert ED, Hartman BK. Effect of ondansetron, a 5-HT3 receptor
antagonist, on the dynamic association between bulimic behaviors and pain
thresholds. Pain 1998; 77:297–303.
163. Faris PL, Kim SW, Meller WH, Goodale RL, Oakman SA, Hofbauer RD, Marshall
AM, Daughters RS, Banerjee-Stevens D, Eckert ED, Hartman BK. Effect of
decreasing afferent vagal activity with ondansetron on symptoms of bulimia
nervosa: a randomised, double-blind trial. Lancet 2000; 355:792–797.
164. Roychoudhury M, Kulkarni SK. Anti-anxiety profile of ondansetron, a selective 5-
HT3 antagonist, in a novel animal model. Methods Find Exp Clin Pharmacol 1997;
19:107–111.
165. Glover V, Halket JM, Watkins PJ, Clow A, Goodwin BL, Sandler M. Isatin:
identity with the purified endogenous monoamine oxidase inhibitor tribulin.
J Neurochem 1988; 51:656–659.
12
Neuroendocrine and Neuropeptide
Dysregulation in Anorexia Nervosa, Bulimia
Ursula F.Bailer and Walter H.Kaye
Western Psychiatric Institute and Clinic, University of Pittsburgh
School of Medicine
NEUROENDOCRINOLOGY
Abnormal hormone profiles and responses to challenge are closely related to the
“starvation” status of anorexia nervosa (AN) and bulimia nervosa (BN) patients.
Hormone abnormalities may also be present, but to a lesser extent, in normal-
weight women with BN. The presence of starvation in AN is evident from the
weight loss, but it may not be recognized in normal-weight bulimics. Although
bulimic women often maintain a normal weight, they do so by restricting food
intake when not bingeing and purging, and they may have monotonous and
poorly balanced meals. Starvation-induced depletion of hepatic glycogen stores
results in free fatty acids and ketone bodies replacing glucose as the primary
energy source. This shift from glycogenolysis to lipolysis and ketogenesis is
associated with an increase in free fatty acids and their metabolites.
(β-Hydroxybutyric acid levels are elevated in both AN and BN (1), indicating
that bulimic patients are nutritionally depleted in spite of their normal body
weight.
The relationship of starvation and eating disorders to neuroendocrine
function is most clearly seen for the pituitary-gonadal axis.
Secondary amenorrhea is one of the criteria for AN in postmenarcheal women,
and oligomenorrhea occurs in about 50% of bulimics. The secondary
amenorrhea is a direct result of altered gonadotropin secretion. Serum sex
hormonebinding globulin may be increased, and both estrogen and testosterone
are decreased (2). The luteinizing hormone response to luteinizing
hormonereleasing hormone stimulation is blunted, but the follicle-stimulating
hormone response is usually normal.
With reference to the hypothalamic-pituitary-adreno-cortical (HPA) axis, it
is well known that plasma cortisol is increased at all times of the day and night,
but its circadian rhythm is preserved in terms of amplitude and timing.
284 BAILER AND KAYE
Stimulation and suppression tests of the HPA axis have been conducted mainly
in AN, and they are in accord with the baseline hormone findings.
Adrenocorticotropic hormone (ACTH) response to corticotropinreleasing
hormone (CRH) administration is reduced, undoubtedly secondary to enhanced
negative feedback on the pituitary corticotrophs exerted by elevated circulating
cortisol. The cortisol response to ACTH administration is increased, suggesting
increased secretory capacity of the adrenal cortex. The low-dose
dexamethasone suppression test is abnormal in 50–90% of anorexics and in 20%
to 60% of bulimics, depending on the weight loss. Because dexamethasone acts
primarily at the pituitary, ACTH and cortisol escape from dexamethasone
suppression, suggesting increased suprapituitary stimulation of corticotrophs by
CRH and vasopressin. Taken together, the pituitary-adrenocortical findings
indicate a mild to moderate activation of this hormone axis in AN and BN.
Interestingly, the abnormalities in AN and in reduced-weight BN (3,4) are
strikingly similar to those occurring in 30–50% of patients with major
depression, although malnutrition, and not mood disturbances, are likely to be
most contributory.
Obese women with binge eating disorder do not show abnormalities of
dexamethasone suppression, either before or after weight loss (5). Data are not
available on prolactin, growth hormone (GH) secretion, or 24-h cortisol
secretion in subjects with binge eating disorder (6). Among obese binge eaters
who do not purge, no abnormalities of glucose or insulin have been reported
relative to weight-matched controls (7). However, sophisticated methodologies
to evaluate cephalic-phase insulin release (CPIR) and glucose metabolism have
not been applied to this population.
With reference to the pituitary-thyroid axis, starvation leads to considerably
decreased plasma free triiodothyronine (T3) concentrations, along with
somewhat decreased plasma free thyroxine (T4) and increased plasma reverse
T3 concentrations. This represents the “euthyroid sick syndrome” hormone
profile (8,9). The decreased circulating T3 helps reduce energy expenditure and
minimizes muscle protein catabolism into amino acids for gluconeogenesis.
Cerebrospinal fluid (CSF) thyrotropin-releasing hormone also appears to be
reduced in AN (10). When bingeing, bulimic patients generally have normal
thyroid indices with perhaps reduced T3 and thyroidstimulating hormone
concentrations; however, when they become abstinent, their pituitary-thyroid
axis function resembles that of anorexic patients (11–13). Thyroid hormone
levels are similar in obese binge eaters and weightmatched controls (7,14).
Insulinlike growth factor, type I (IGF-1) concentrations are low in both AN
and BN, and circulating GH is increased, perhaps owing to diminished feedback
of IGF-1 on GH secretion. Circulating prolactin is usually unchanged in AN and
may be reduced in BN. Prolactin responses to serotonergic challenges such as
NEUROENDOCRINE AND NEUROPEPTIDE DYSREGULATION 285
meta-chlorophenylpiperazine, fenfluramine, L-tryptophan, and 5-OH-

tryptophan are diminished in both AN and BN.
NEUROPEPTIDES
The past decade has witnessed accelerating basic research on the role of
neuropeptides in the regulation of feeding behavior and obesity. The mechanisms
for controlling food intake involve a complicated interplay between peripheral
systems (including gustatory stimulation, gastrointestinal peptide secretion, and
vagal afferent nerve responses) and central nervous system (CNS)
neuropeptides and/or monoamines. Thus, studies in animals show that
neuropeptides, such as cholecystokinin, the endogenous opioids (such as
β-endorphin), and neuropeptide Y regulate the rate, duration, and size of meals,
as well as macronutrient selection (15,16). In addition to regulating eating
behavior, a number of CNS neuropeptides participate in the regulation of
neuroendocrine pathways. Thus, clinical studies have evaluated the possibility
that CNS neuropeptide alterations may contribute to dysregulated secretion of
the gonadal hormones, cortisol, thyroid hormones, and growth hormone in the
eating disorders (17,18).
While there are relatively few studies to date, most of the neuroendocrine
and neuropeptide alterations apparent during symptomatic episodes of AN and
BN tend to normalize after recovery. This observation suggests that most of the
disturbances are consequences rather than causes of malnutrition, weight loss,
and/or altered meal patterns. Still, an understanding of these neuropeptide
disturbances may shed light on why many people with AN or BN cannot easily
“reverse” their illness. In AN, malnutrition may contribute to a downward
spiral sustaining and perpetuating the desire for more weight loss and dieting.
Symptoms such as increased satiety, obsessions, and dysphoric mood may be
exaggerated by these neuropeptide alterations and thus contribute to the
downward spiral. In addition, mutual interactions between neuropeptide,
neuroendocrine, and neurotransmitter pathways may contribute to the
constellation of psychiatric comorbidity often observed in these disorders. Even
after weight gain and normalization of eating patterns, many individuals who
have recovered from AN or BN have physiological, behavioral, and
psychological symptoms that persist for extended periods. Menstrual cycle
dysregulation, for example, may persist for several months after weight
restoration. The following sections provide a brief overview of studies of
neuropeptides in AN and BN.
286 BAILER AND KAYE
Corticotropin-Releasing Hormone
When underweight, patients with AN have increased plasma cortisol secretion,
which is thought to be at least in part a consequence of hypersecretion of
endogenous CRH (3,19–21). In that the plasma and CSF measures return to
normal, it appears likely that activation of the HPA axis is precipitated by
weight loss. The observation of increased CRH activity is of great theoretical
interest in AN since intracerebroventricular CRH administration in
experimental animals produces many of the physiological and behavioral
changes associated with AN, including markedly decreased eating behavior (22),
hypothalamic hypogonadism (23), decreased sexual activity (24), and hyper
activity (25).
Opioid Peptides
Studies in laboratory animals raise the possibility that altered endogenous opioid
activity might contribute to pathological feeding behavior in eating disorders
since opioid agonists generally increase, and opioid antagonists decrease, food
intake (26). State-related reductions in concentrations of CSF (3-endorphin and
related opiate concentrations have been found in both underweight AN and ill
BN subjects (27–29). In contrast, using the T lymphocyte as a model system,
Brambilla et al. (30) found elevated β-endorphin levels in AN, although the
levels were normal in BN (31). If β-endorphin activity is a facilitator of feeding
behavior, then reduced CSF concentrations could reflect decreased central
activity of this system, which then maintains or facilitates inhibition of feeding
behavior in the eating disorders.
A disturbance in CNS opioid function may also contribute to the
neuroendocrine abnormalities in AN and BN (e.g., disturbances in HPA and
pituitary-gonadal axis function) (32,33). Brain opioid pathways inhibit ACTH
and cortisol release in humans, and they suppress pulsatile gonadotropin
secretion in rats and in sexually mature humans. Underweight anorexics
frequently have a blunted response of LH secretion to opiate antagonists (34),
and weight restoration tends to normalize this response. The failure of opioid
antagonists to increase luteinizing hormone secretion in underweight anorexics
suggests that another neurotransmitter system (or systems) may be responsible
for this neuroendocrine disturbance.
Vasopressin and Oxytocin

In addition to the effects of vasopressin on HPA axis regulation and freewater
clearance by the kidney and the effects of oxytocin during the puerperium, these
structurally related neuropeptides are distributed throughout the CNS and

function as long-acting neuromodulators of complex behaviors. The effects of
vasopressin appear to be reciprocal to those of oxytocin; central administration
of vasopressin to rats enhances memory consolidation and retrieval, whereas
administration of oxytocin disrupts memory (35).
In addition to abnormally high CSF vasopressin concentrations and impaired
osmoregulation of plasma vasopressin (36), AN patients have reduced CSF
oxytocin concentrations and impaired plasma oxytocin responses to
stimulation (37). Underweight anorexics also have an impaired plasma oxytocin
response to challenging stimuli (38). These abnormalities tend to normalize
after weight restoration, suggesting that they are secondary to malnutrition or
abnormal fluid balance, or both. In underweight anorexics, low CNS oxytocin
might interact with high CNS vasopressin to enhance the retention of cognitive
distortions of the aversive consequences of eating, thereby reinforcing these
patients’ perseverative preoccupation with the adverse consequences of food
intake.
Normal-weight bulimics were found to have elevated CSF vasopressin
concentrations but normal CSF oxytocin both on admission and after 1 month
of nutritional stabilization and abstinence from bingeing and purging. In these
patients as well, CNS vasopressin might contribute to their obsessional
preoccupation with the aversive consequences of weight gain (39).
Neuropeptide Y and Peptide YY

Neuropeptide Y (NPY) and Peptide YY (PYY) are of considerable theoretical
interest since they are among the most potent endogenous stimulants of feeding
behavior in the CNS (16,26,40). PYY is more potent than NPY in stimulating
food intake; both are selective for carbohydrate-rich foods. Underweight
individuals with AN have been shown to have elevations of CSF NPY but
normal PYY (41). Clearly, elevated NPY does not result in increased feeding in
underweight individuals with AN; however, the possibility that increased NPY
activity underlies the obsessive and paradoxical interest in dietary intake and
food preparation is a hypothesis worth exploring. On the other hand, CSF levels
of NPY and PYY have been reported to be normal in women with BN when
measured while subjects were acutely ill. Although levels of PYY increased
above normal when subjects were reassessed after one month of abstinence from
bingeing and vomiting, levels of the peptides were similar to control values in
long-term recovered individuals (42). More recently, it has been reported that
the plasma concentration of NPY was lower in AN patients than in controls,
whereas BN patients had elevated NPY levels (43). Additional studies to assess
the potential behavioral correlates of these findings are needed.
288 BAILER AND KAYE
Cholecystokinin
Cholecystokinin (CCK) is a peptide secreted by the gastrointestinal system in
response to food intake. Release of CCK is thought to be one means of
transmitting satiety signals to the brain by way of vagal afferents (44). In parallel
to its role in satiety in rodents, exogenously administered CCK reduces food
intake in humans. The preponderance of data suggests that patients with BN, in
comparison to controls, have diminished release of CCK following ingestion of
a standardized test meal (45–48). Measurements of basal CCK values in blood
lymphocytes and in CSF also appear to be decreased in patients with
BN (31,49). It has been suggested that the diminished CCK response to a meal
may have a role in diminished postingestive satiety observed in BN. The
CCK response in BN patients was found to return to normal following
treatment (45).
Studies of CCK in AN have yielded less consistent findings. Some studies
have found elevations in basal levels of plasma CCK (46,50), as well as increased
peptide release following a test meal (46,51). One study found that blunting of
CCK response to an oral glucose load normalized in AN patients after
partial restoration of body weight (50). Other studies have found that
measures of CCK function in AN were similar to or lower than control
values (30,47,52,53). Further studies are needed to evaluate the relationship
between altered CCK regulation and other indices of abnormal gastric function
in symptomatic BN and AN patients (54).
Leptin
Leptin, the protein product of the ob gene, is secreted predominantly by adipose
tissue cells and acts in the CNS to decrease food intake, thus regulating body fat
stores. In rodent models, defects in the leptin coding sequence resulting in
leptin deficiency or defects in leptin receptor function are associated with
obesity. In humans, serum and CSF concentrations of leptin are positively
correlated with fat mass in individuals across a broad range of body weight,
including obesity (55,56). Thus, obesity in humans is not thought to be a result
of leptin deficiency per se, although rare genetic deficiencies in leptin
production have been associated with familial obesity (57).
Underweight patients with AN have consistently been found to have
significantly reduced serum leptin concentrations in comparison to normal
weight controls (53,58–61). Based on studies in laboratory animals, it has been
suggested that low leptin levels may contribute to amenorrhea and other
hormonal changes in the disorder (60). Although the reduction in fasting serum
leptin levels in AN is correlated with reduction in body mass index (BMI), there
has been some discussion of the possibility that leptin levels in AN patients may
be higher than expected based on the extent of weight loss (62,63). Mantzoros
et al. (60) reported an elevated CSF to serum leptin ratio in AN compared to
controls, suggesting that the proportional decrease in leptin levels with weight
loss is greater in serum than in CSF. A longitudinal investigation during
refeeding in AN patients has shown that CSF leptin concentrations reach normal
values before full weight restoration, possibly as a consequence of the relatively
rapid and disproportionate accumulation of fat during refeeding (60). This
finding led the authors to suggest that premature normalization of leptin
concentration might contribute to difficulty in achieving and sustaining a normal
weight in AN. Plasma and CSF leptin levels appear to be similar to control
values in long-term recovered AN subjects (42).
Recent studies indicate that patients with BN, in comparison to carefully
matched controls, have significantly decreased leptin concentrations in serum
samples obtained after overnight fast (43,62,64–66). Initial findings in individuals
who have achieved sustained recovery from BN, when compared to controls
with closely matched percent body fat, suggest that serum leptin levels remain
decreased. This finding may be related to evidence for a persistent decrease in
activity in the hypothalamic-pituitary-thyroid axis in long-term recovered BN
individuals. These alterations could be associated with decreased metabolic rate
and a tendency to weight gain, contributing to the preoccupation with body
weight characteristic of BN.
Ghrelin
Ghrelin was originally discovered in the rat and the human stomach, and
stimulates GH secretion in rodents. This petide that antagonizes leptin action has
a role in the regulation of feeding behavior and energy metabolism in the
CNS (67). Ghrelin-producing neurons are located in the hypothalamus,
whrereas ghrelin receptors are expressed in various regions of the brain.
Intracerebroventricular injections of ghrelin strongly stimulated feeding
in rats and increased body weight gain. In addition, it has been reported
that fasting plasma ghrelin concentrations in humans are negatively correlated
with BMI (68,69), percentage body fat and fasting leptin and insulin
concentrations (70), which have an important role in the pathophysiology of
AN (71). In the latter study it could be shown that ghrelin was elevated in AN
patients and returned to normal levels after weight recovery. The possibility of
ghrelin resistance in cachectic states as caused by eating disorders could
be suggested. Fasting plasma ghrelin concentrations in patients with BN were
significantly higher than those in controls (69), although the BMIs between
bulimics and controls were not significantly different, suggesting that not only
290 BAILER AND KAYE
BMI, but also abnormal eating behavior with bingeing and purging, had an
influence on circulating ghrelin level in BN patients.
Gastrin-Releasing Peptide
Human gastrin-releasing peptide (GRP) is a 27-amino-acid peptide that shares
similar decapeptide with bombesin (BBS) (72). Peripheral and central
administration of GRP attenuates food intake in mammals and humans (73,74).
In the CNS, distinct BBS-like receptor subtypes have been identified in brain
tissue such as the bed nucleus of the stria terminalis, the olfactory tubercle, the
putamen, and the neocortex, with a neuromedin B- and a GRPpreferring
subtype (75,76). Both subtypes have been implicated in the modulation of BBS-
like peptide-induced food suppression (77). CSF GRP was significantly lower in
recovered bulimic patients (>1 year, normal weight, and regular menstrual
cycles, no bingeing or purging) compared to normal controls and recovered
anorectic patients (78). Lower CSF GRP in this group could be a trait-related
disturbance that might add to hyperphagic behavior, and thus to the
pathophysiology of this illness.
CONCLUSIONS AND PERSPECTIVES

The increase in understanding of neuropeptide modulation of appetite and
weight control also resulted in new insights into endocrine and neuropepetide
disturbances in AN and BN. Obviously, there are still many methodological
problems to be taken into consideration when interpreting the endocrino logical
observations. Animal models that focus on one facet of behavior, such as motor
activity or sexual receptivity, are not necessarily suitable models for AN. The
serum concentrations of monoamines and peptides reflect pituitary secretion
but may provide a limited perspective on higher brain function. CSF measures
reflect some general pool of chemicals but offers limited understanding of
specific pathways. Minor weight changes in patients with AN are associated with
significant responses in serum catecholamines, leptin, cortisol, gonadotropin,
and GH, indicating that the timing of the respective investigations is of critical
importance and may be a cause of discrepant findings in several studies (18).
Among individuals who engage in binge eating without purging, few
biological correlates of binge eating have been identified. Neuroendocrine
function does not differ from weight-matched controls in nonpurging binge
eaters (6).
Determining whether abnormalities are a consequence or a potential
antecedent of pathological feeding behavior is a major question in eating disorders.
When studying patients who had recovered from their eating disorder, any
persistent psychobiological abnormalities might be trait related and potentially

have contributed to the pathogenesis of the disorder.
Last but not least, most models describe only one or two specific systems out
of many, while our knowledge about the interactions between these systems is
limited and, at the present time, it is not possible to map the sum of numerous
interactive pathways.
REFERENCES
1. Pirke KM, Pahl J, Schweiger U. Metabolic and endocrine indices of starvation in
bulimia: a comparison with anorexia nervosa. Psychiatry Res 1985; 15:33–39.
2. Tomova A, Kumanov P, Kirilov G. Factors related to sex hormone binding
globulin concentrations in women with anorexia nervosa. Psychosom Med 1995;
40:499–506.
3. Gold PW, Gwirtsman H, Avgerinos PC, Nieman LK, Gallucci WT, Kaye W,
Jimerson D, Ebert M, Rittmaster R, Loriaux DL. Abnormal hypothalamicpituitary-
adrenal function in anorexia nervosa. Pathophysiologic mechanisms in underweight
and weight-corrected patients. N Engl J Med 1986; 314:1335–1342.
4. Fichter MM, Pirke KM, Pollinger J, et al. Disturbances in the
hypothalamopituitary-adrenal and other neuroendocrine axes in bulimia. Biol
Psychiatry 1990; 27:1021–1037.
5. Yanovski SZ, Yanovski JA, Gwirtsman HE, Bernat A, Gold PG, Chrousos GP.
Normal dexamethasone suppression in obese binge and nonbinge eaters with rapid
weight loss. J Clin Endocrinol Metab 1993; 76:675–679.
6. Yanovski SZ. Biological correlates of binge eating. Addict Behav 1995; 20:
705–712.
7. Adami GF, Gandolfo P, Campostano A, Cocchi F, Bauer B, Scopinaro N. Obese
binge eaters: metabolic characteristics, energy expenditure, and dieting. Psychol
Med 1995; 25:195–198.
8. Wartofsky L, Burman KD. Alterations in throid function in patients with systemic
illness: the “euthyroid sick syndrome.” Endocr Rev 1982; 3:164–217.
9. Altemus M, Hetherington M, Kennedy B, et al. Thyroid function in bulimia
nervosa. Psychoneuroendocrinology 1996; 21:249–261.
10. Lesem MD, Kaye WH, Bissette G, et al. Cerebrospinal fluid TRH
immunoreactivity in anorexia nervosa. Biol Psychiatry 1994; 35:48–53.
11. Devlin MJ, Walsh BT, Kral JG, et al. Metabolic abnormalities in bulimia nervosa.
12. Altemus M, Hetherington M, Flood M. Decrease in resting metabolic rate during
abstinence from bulimic behavior. Am J Psychiatry 1991; 148:1071–1072.
13. Spalter AR, Gwirtsman HE, Demitrack MA, et al. Throid function in bulimia
292 BAILER AND KAYE
14. Wadden TA, Foster GD, Letizia KA, Wilk JE. Metabolic, anthropometric, and
psychological characteristics of obese binge eaters. Int J Eat Disord 1993; 14:
127–135.
15. Morley JE, Bhmdell JE. The neurobiological basis of eating disorders: some
formulations. Biol Psychiatry 1988; 23:53–78.
16. Schwartz MW, Woods SC, Porte D Jr, Seeley RJ, Baskin DG. Central nervous
system control of food intake. Nature 2000; 404:661–671.
17. Jimerson DC, Wolfe BE, Naab S. In: Anorexia nervosa and bulimia nervosa. CE,
Brumback RA, eds. Textbook of Pediatric Neuropsychiatry Coffee Washington,
DC: American Psychiatric Press, 1998:563–578.
18. Stoving RK, Hangaard J, Hansen-Nord M, Hagen C. A review of endocrine
changes in anorexia nervosa. J Psychiatr Res 1999; 33:139–152.
19. Kaye WH, Gwirtsman HE, George DT, Ebert MH, Jimerson DC, Tomai TP,
Chrousos GP, Gold PW. Elevated cerebrospinal fluid levels of immunoreactive
corticotropin-releasing hormone in anorexia nervosa: relation to state of nutrition,
adrenal function, and intensity of depression. J Clin Endocrinol Metab 1987; 64:
203–208.
20. Walsh BT, Roose SP, Katz JL, Dyrenfurth I, Wright L, Vande Wiele R, Glassman
AH. Hypothalamic-pituitary-adrenal-cortical activity in anorexia nervosa and
bulimia. Psychoneuroendocrinology 1987; 12:131–140.
21. Licinio J, Wong ML, Gold PW. The hypothalamic-pituitary-adrenal axis in
anorexia nervosa. Psychiatry Res 1996; 62:75–83.
22. Glowa JR, Gold PW. Corticotropin releasing hormone produces profound
anorexigenic effects in the rhesus monkey. Neuropeptides 1991; 18:55–61.
23. Rivier C, Vale W. Influence of corticotropin releasing factor on reproductive
functions in the rat. Endocrinology 1984; 114:914–921.
24. Sirinathsinghji DJ, Rees LH, Rivier J. Corticotropin-releasing factor is a potent
inhibitor of sexual receptivity in the female rat. Nature 1983; 305:232–235.
25. Sutton RE, Koob GF, LeMoal M. Corticotropin-releasing factor produces
behavioral activation in rats. Nature 1982; 297:331–333.
26. Morley JE, Levine AS, Gosnell BA, Mitchell JE, Krahn DD, Nizielski SE. Peptides
and feeding. Peptides 1985; 6:181–192.
27. Kaye WH, Berrettini WH, Gwirtsman HE, Chretien M, Gold PW, George DT,
Jimerson DC, Ebert MH. Reduced cerebrospinal fluid levels of immunoreactive
pro-opiomelanocortin related peptides (including beta-endorphin) in anorexia
28. Lesem MD, Berrettini W, Kaye WH, Jimerson DC. Measurement of CSF
dynorphin A 1–8 immunoreactivity in anorexia nervosa and normal-weight bulimia.
Biol Psychiatry 1991; 29:244–252.
29. Brewerton TD, Lydiard RB, Laraia MT, Shook JE, Ballenger JC. CSF
betaendorphin and dynorphin in bulimia nervosa. Am J Psychiatry 1992; 149:
1086–1090.
30. Brambilla F, Brunetta M, Peirone A, Perna G, Sacerdote P, Manfredi B, Panerai

AE. T-lymphocyte cholecystokinin-8 and beta-endorphin concentrations in eating
disorders: I. Anorexia nervosa. Psychiatry Res 1995; 59:43–50.
31. Brambilla F, Brunetta M, Draisci A, Manfredi B, Panerai AE, Peirone A, Perna G,
Sacerdote P. T-lymphocyte cholecystokinin-8 and beta-endorphin in eating
disorders: II. Bulimia nervosa. Psychiatry Res 1995; 59:51–56.
32. Grossman A. Brain opiates and neuroendocrine function. Clin Endocrinol Metab
1983; 12:725–746.
33. Pfeiffer A, Herz A. Endocrine actions of opioids. Horm Metab Res 1984; 16:
386–397.
34. Armeanu M, Berkhout GMJ, Schoemaker J. Pulsatile luteinizing hormone secretion
in hypothalamic amenorrhea, anorexia nervosa, and polycystic ovarian disease
during naltrexone treatment. Fertil Steril 1992; 57:762–770.
35. Bohus B, Kovacs GL, De Weid D. Oxytocin, vasopressin and memory: opposite
effects on consolidation and retrial process. Brain Res 1978; 157:414–417.
36. Nishita JK, Ellinwood EH Jr, Rockwell WJ. Abnormalities in the response of plasma
arginine vasopressin durinh hypertonic saline infusion in patients with eating
disorders. Biol Psychiatry 1989; 26:73–86.
37. Demitrack MA, Lesem MD, Listwak SJ. CSF oxytocin in anorexia nervosa and
bilimia nervosa: clinical and pathophysiological considerations. Am J Psychiatry
1990; 147:882–886.
38. Chiodera P, Volpi R, Capretti L. Effect of estrogen or insulin-induced
hypoglycemia on plasma oxytocin levels in bulimia and anorexia nervosa.
Metabolism 1991; 40:1226–1230.
39. Gold PW, Kaye WH, Robertson GL, et al. Abnormalities in plasma and
cerebrospinal-fluid arginine vasopressin in patients with anorexia nervosa. N Engl J
Med 1983; 308:1117–1123.
40. Kalra SP, Dube MG, Sahu A, Phelps CP, Kalra PS. Neuropeptide Y secretion
increases in the paraventricular nucleus in association with increased appetite for
food. Proc Natl Acad Sci USA 1991; 88:10931–10935.
41. Kaye WH, Berrettini W, Gwirtsman H, George DT. Altered cerebrospinal fluid
neuropeptide Y and peptide YY immunoreactivity in anorexia and bulimia nervosa.
42. Gendall K. Leptin, neuropeptide Y, and peptide YY in long-term recovered eating
disorder patients. Biol Psychiatry 1999; 46:292–299.
43. Baranowska B, Wolinska-Witort E, Wasilewska-Dziubinska E, Roguski K,
Chmielowska M. Plasma leptin, neuropeptide Y (NPY) and galanin concentrations
in bulimia nervosa and in anorexia nervosa. Neuroendocrinol Lett 2001; 22:356–
358.
44. Gibbs J, Young RC, Smith GP. Cholecystokinin decreases food intake in rats.
J Comp Physiol Psychology 1973; 84:488–495.
45. Geracioti TD Jr, Liddle RA. Impaired cholecystokinin secretion in bulimia nervosa.
N Engl J Med 1988; 319:683–688.
294 BAILER AND KAYE
46. Phillipp E, Pirke KM, Kellner MB, Krieg JC. Disturbed cholecystokinin secretion
in patients with eating disorders. Life Sci 1991; 48:2443–2450.
47. Pirke KM, Kellner MB, Friess E, Krieg JC, Fichter MM. Satiety and
cholecystokinin. Int J Eat Disord 1994; 15:63–69.
48. Devlin MJ, Walsh BT, Guss JL, Kissileff HR, Liddle RA, Petkova E. Post prandial
cholecystokinin release and gastric emptying in patients with bulimia nervosa. Am J
Clin Nutr 1997; 65:114–120.
49. Lydiard RB, Brewerton TD, Fossey MD, Laraia MT, Stuart G, Beinfeld MC,
Ballenger JC. CSF cholecystokinin octapeptide in patients with bulimia nervosa and
in normal comparison subjects. Am J Psychiatry 1993; 150:1099–1101.
50. Tamai H, Takemura J, Kobayashi N, Matsubayashi S, Matsukura S, Nakagawa T.
Changes in plasma cholecystokinin concentrations after oral glucose tolerance test
in anorexia nervosa before and after therapy. Metab Clin Exp 1993; 42:581–584.
51. Harty RF, Pearson PH, Solomon TE, McGuigan JE. Cholecystokinin, vasoactive
intestinal peptide and peptide histidine methionine responses to feeding in anorexia
nervosa. Regul Peptides 1991; 36:141–150.
52. Geracioti TD Jr, Liddle RA, Altemus M, Demitrack MA, Gold PW. Regulation of
appetite and cholecystokinin secretion in anorexia nervosa. Am J Psychiatry 1992;
149:958–961.
53. Baranowska B, Radzikowska M, Wasilewska-Dziubinska E, Roguski K, Borowiec
M. Disturbed release of gastrointestinal peptides in anorexia nervosa and in
obesity. Diabetes Obes Metab 2000; 2:99–103.
54. Geliebter A, Melton PM, McCray RS, Gallagher DR, Gage D, Hashim SA. Gastric
capacity, gastric emptying, and test-meal intake in normal and bulimic women. Am
J Clin Nutr 1992; 56:656–661.
55. Considine RV, Considine EL, Williams CJ, Hyde TM, Caro JF. The hypothalamic
leptin receptor in humans: identification of incidental sequence polymorphisms and
absence of the db/db mouse and fa/fa rat mutations. Diabetes 1996; 45:992–994.
56. Schwartz MW, Peskind E, Raskind M, Boyko EJ, Porte D Jr. Cerebrospinal fluid
leptin levels: relationship to plasma levels and to adiposity in humans. Nature Med
1996; 2:589–593.
57. Farooqi IS, Keogh JM, Kamath S, Jones S, Gibson WT, Trussell R, Jebb SA, Lip
GY, O’Rahilly S. Partial leptin deficiency and human adiposity. Nature 2001; 414:
34–35.
58. Hebebrand J, van der Heyden J, Devos R, Kopp W, Herpertz S, Remschmidt H,
Herzog W. Plasma concentrations of obese protein in anorexia nervosa. Lancet
1995; 346:1624–1625.
59. Grinspoon S, Gulick T, Askari H, Landt M, Lee K, Anderson E, Ma Z, Vignati L,
Bowsher R, Herzog D, Klibanski A. Serum leptin levels in women with anorexia
nervosa. J Clin Endocrinol Metab 1996; 81:3861–3863.
60. Mantzoros C, Flier JS, Lesem MD, Brewerton TD, Jimerson DC. Cerebrospinal
fluid leptin in anorexia nervosa: correlation with nutritional status and potential
role in resistance to weight gain. J Clin Endocrinol Metab 1997; 82: 1845–1851.
61. Eckert ED, Pomeroy C, Raymond N, Kohler PF, Thuras P, Bowers CY. Leptin in
anorexia nervosa. J Clin Endocrinol Metab 1998; 83:791–795.
62. Frederich R, Hu S, Raymond N, Pomeroy C. Leptin in anorexia nervosa and bulimia
nervosa: importance of assay technique and method of interpretation. J Lab Clin
Med 2002; 139:72–79.
63. Jimerson DC. Leptin and the neurobiology of eating disorders. J Lab Clin Med
2002; 139:70–71.
64. Brewerton TD, Lesem MD, Kennedy A, Garvey WT. Reduced plasma leptin
concentration in bulimia nervosa. Psychoneuroendocrinology 2000; 25:649–658.
65. Jimerson DC, Mantzoros C, Wolfe BE, Metzger ED. Decreased serum leptin in
bulimia nervosa. J Clin Endocrinol Metab 2000; 85:4511–4514.
66. Monteleone P, Bortolotti F, Fabrazzo M, La Rocca A, Fuschino A, Maj M. Plasma
leptin response to acute fasting and refeeding in untreated women with bulimia
nervosa. J Clin Endocrinol Metab 2000; 85:2499–2503.
67. Nakazato M, Murakami N, Date Y, Kojima M, Matsuo H, Kangawa K, Matsukura
S. A role for ghrelin in the central regultion of feeding. Nature 2001; 409:
194–198.
68. Shiiya T, Nakazato M, Mizuta M, Date Y, Mondal MS, Tanaka M, Nozoe S,
Hosoda H, Kangawa K, Matsukura S. Plasma ghrelin levels in lean and obese
humans and the effect of glucose on ghrelin secretion. J Endocrinol Metab 2002;
87:240–244.
69. Tanaka M, Naruo T, Muranaga T, Yasuhara D, Shiiya T, Nakazato M, Matsukura S,
Nozoe S. Increased fasting plasma ghrelin levels in patients with bulimia nervosa.
Eur J Endocrinol 2002; 146:R1-R3.
70. Tschop M, Weyer C, Tataranni AP, Devanarayan V, Ravussin B, Heiman ML.
Circulating ghrelin levels are decreased in human obesity. Diabetes 2001; 50:
707–709.
71. Otto B, Cuntz U, Fruehauf E, Wawarta R, Folwaczny C, Riepl RL, Heiman ML,
Lehnert P, Fichter M, Tschöp M. Weight gain decrease elevated plasma ghrelin
concentrations of patients with anorexia nervosa. Eur J Endocrinol 2001; 145:
669–673.
72. Brown M, Marki W, Rivier J. Is gastrin releasing peptide mammalian bombesin?
LifeSci 1980; 27:125–128.
73. Bray GA. Nutrient intake is modulated by peripheral petide aministration. Obesity
Res 1995; 3(Suppl. 4):569S-572S.
74. Flynn FW. Bombesin-like peptides in the reglation of ingestive behavior. Ann NY
Acad Sci 1994; 739:120–134.
75. Ladenheim EE, Jensen RT, Mantey SA, Moran TH. Distinct distributions of two
bombesin receptor subtypes in the rat central nervous system. Brain Res 1992; 593:
168–178.
76. Wolf SS, Moody TW. Receptors for GRP/bombesin-like peptides in the rat
forebrain. Peptides 1985; 6(Suppl. 1):111–114.
296 BAILER AND KAYE
77. Ladenheim EE, Wirth KE, Moran TH. Receptor subtype mediation of feeding
suppression by bombesin-like peptides. Pharmacol Biochem Behav 1996; 54:
705–711.
78. Frank GK, Kaye WH, Ladenheim EE, McConaha C. Reduced gastrin releasing
peptide in cerebrospinal fluid after recovery from bulimia nervosa. Appetite 2001;
37:9–14.
13
Neuroimaging of the Eating Disorders
Janet Treasure
Thomas Guy House
London, England
Rudolf Uher
Institute of Psychiatry
London, England
It has been an unfortunate fact that eating disorders are all too often dismissed as
trivial complaints of spoilt princesses who are merely being stubborn and
willful. However, it will be difficult to continue to hold such a position in the
face of the evidence that is coming from research into brain imaging. Pictures
can be worth a thousand words, and the brain images, which show both
structural and functional brain changes, cannot be easily dismissed. Food
provokes a unique pattern of brain activation in people with eating disorders.
Moreover, food-induced activation of the orbitofrontal cortex continues after
recovery. Also structural abnormalities persist after recovery. Are these effects
scars from the illness or markers of a neurodevelopmental diathesis? The
answers may have implications for clinical evaluation, diagnostics, treatment,
and outcome. We therefore argue that brain imaging is not just an interesting
application of modern technology but an essential and exciting tool that can be
used to clarify the confusing tapestry of eating disorders.
Interest in imaging applied to eating disorders began with the increased
resolution offered by computed tomography (CT). This revealed that the brain,
in parallel with all other organs of the body, was shrunken especially in AN.
However, this was somewhat dismissed as a nonspecific side effect of starvation
and added to the list of numerous clinical consequences. However, the second
phase of this research, which relates to brain function, can be and has been
hypothesis driven. We predict that this may have a more profound effect on
thinking in the field.
GENERAL ASPECTS OF METHODOLOGY

There are several aspects of the technology and methodology of brain imaging
that are of importance in understanding and interpreting research in this field.
298 TREASURE AND UHER
For example, the level of resolution and the balance between cost, utility, and
acceptability varies among the different technologies. Functional magnetic
resonance scanning has the best resolution and does not involve injections of
radioactive substances; however, the design at the moment is limited to rapid
changes in blood flow. The other two technologies allow for investigation into
the chemistry of the brain but the time course and resolution are slower. Each of
the technologies is best suited to answering different questions, and there can be
useful synergy between the approaches.
Scanning experiments have to be carefully designed. They are expensive and
can be difficult to interpret. The experimental conditions (both internal and
external) need to be clearly defined. For example, if food is used as the
paradigm under investigation it is important to control for metabolic factors
such as time since the last meal. In earlier studies, all patients with an eating
disorder (anorexia nervosa or bulimia) were analyzed as a group with no attempt
to subcategorize according to eating behaviors. However, bingeing/disinhibited
eating appears to be associated with a distinct pattern of brain activation to food
whereas weight causes less of a differential response. Thus, the
subcategorization as defined in DSM-IV has some utility.
As in all aspects of science it helps to have a clear, specific hypothesis. Here
we have an advantage over those working in different fields of psychopathology
as the brain physiology and biochemistry of appetite control has been clearly
defined in animal models. Furthermore, it has been possible to develop this
research further in man. Thus, it is possible to develop testable hypotheses in
relationship to food, drawing from this body of research.
The statistical techniques used to analyze clusters of activation are complex,
but in the end all methods need to follow the basic principles of eliminating type
1 and 2 errors and adjusting for multiple testing. Unfortunately, in many
scanning studies, especially those in the first wave, these principles were
neglected. However, in later studies the statistical methods are more robust.
In this chapter we have focused on those studies that are of higher quality in
terms of all the features outlined above. However, as this is a new line of
investigation we have also included some studies, that would fail these quality
criteria but are important first steps into this area.
BACKGROUND: BRAIN DEVELOPMENT

Eating disorders have their onset at a crucial time for development: biologically,
cognitively, and socially. Adolescence is a critical phase of brain development.
Although the total brain size does not increase significantly from the age of 5 a
large number of changes occur before mature, adult function emerges. White
matter volume increases significantly during childhood and adulthood, as axons
NEUROIMAGING OF THE EATING DISORDERS 299
are gradually myelinated. A reciprocal decrease in gray matter volume occurs

[for review, see (1)]. A wave of synaptic proliferation takes place in the frontal
lobes at around the time of puberty. This is followed by dramatic reductions in
cortical-cortical connectivity in especially in frontal regions during
adolescence (2). This synaptic pruning is thought to be the central
neurodevelopment event, which leads to full adult functioning. The gender
dimorphism in brain development is of particular relevance for eating disorders.
Interestingly there is differential growth in the areas of the brain implicated in
appetite control. For example, the amygdala is smaller in females whereas the
hippocampus and caudate nucleus are larger.
These developmental changes in brain structure are associated with changes
in brain function. For instance, the reaction time for emotional recognition
increases during adolescence. This is thought to be due to a relative inefficiency
in frontal circuitry prior to the pruning of excess synaptic contacts (3).
Furthermore, links between structure and function have been demonstrated. For
example, the speed with normal children switch attentional set is linked to the
size of the anterior cingulate gyrus (4). In conclusion, the timing of the onset of
eating disorders during this rapid flux in brain structure and function may have
profound implications for both the etiology and course of the illness. In
particular, the frontal lobes with their key role in appetite as well as cognitive
and emotional control are of great relevance.
BRAIN STRUCTURE AND EATING DISORDERS
Anorexia Nervosa
Reduced total cerebral volume is a consistent finding in AN (5–9). The majority
of these studies were undertaken before AN was subdivided into the restricting
and the binge-purge subtypes, and often the generic term “eating disorders” was
used so it is not easy to relate anatomical changes to symptoms. The greater the
absolute weight loss and the faster the rate of weight loss, the smaller the brain
volume. Two small longitudinal studies examined the structural changes in the
brain of adolescents after full weight gain (10,11). Both found persistent deficits
in gray matter (cell bodies of neurons and glial cells), although there was
recovery of white matter (mainly myelinated axons). This supports the finding
of gray matter deficits in people who have made a full recovery from their
eating disorder (12). One postmortem study reported that there was a
reduction in basal dendritic fields and dendritic spine density (13). Similar
findings have been reported in schizophrenia and mood disorders where these
changes are most apparent in those with a positive family history (14).
Brain shrinkage may be localized to specific regions. For instance, the

pituitary gland (15) and the amygdala-hippocampal complex are decreased in
size (16).
Bulimia Nervosa
The cerebral sulci are also widened in BN although the cerebral ventricles are
normal in size as are the thalamus and midbrain areas (17–19). One magnetic
resonance imaging (MRI) study found that the inferior frontal lobe cortex was
reduced in size (18). To our knowledge there are no data on brain structure in
binge eating disorder.
Brain Biochemistry
Brain biochemistry assessed using proton magnetic resonance spectroscopy is
perturbed in people with eating disorders. For example, lipid signals in the
frontal lobe were reduced to half of those seen in the normal population (20).
Lipid levels correlated with body mass index.
Brain Damage and Eating Symptoms

Reports that relate eating disorder symptoms and syndromes with structural
changes in the brain have been recently reviewed (21). Lesions in the prefrontal
and temporal cortices and mesiotemporal structures and hypothalamus
predominantly on the right-hand side are particularly linked with eating
disorder symptoms. These case studies are of interest in testing and generating
hypotheses about the central control of appetite, nutrition, and metabolism;
however, in the majority of cases of AN there is no macroscopic brain lesion.
BRAIN FUNCTION IN EATING DISORDERS

In the absence of specific, structural abnormalities in eating disorders there has
been interest in examining how the brain works. Several studies have examined
the activity of the brain at rest. However, the resting state in the context of
these experimental procedures probably varies between individual and site
depending on the subjective reaction to the experimental maneuver and the
precise metabolic state. Therefore, studies in the resting state can be difficult to
interpret. Other investigators have studied brain activity in response to emotional
and eating disorder-specific stimuli such as food and body shape.
Studies on Eating Disorders in the Resting State

Several studies have examined people with eating disorders in the resting state
using positron emission tomography (PET). As might be expected given the
poor nutritional state in AN with low glucose and high levels of ketones, there
is a global reduction in cerebral glucose metabolism most marked in the frontal
and parietal areas. However, glucose metabolism is relatively increased in the
basal ganglia (22–24). There have also been reports of regional, relative
hypoperfusion using single photon emission computed tomography (SPECT). A
study in children and adolescents with AN found poor flow in the temporal
lobe (25). Adults with the restricting form of AN had decreased flow in the
frontal and anterior cingulate region (26,27). Flow in the thalamus, amygdala,
and hippocampus was increased (27). This differential pattern of resting activity
was not found in those with the binge-purge subtype of AN. In BN no clear
findings have been found in resting blood flow in either the acute state (22,28)
or after recovery (28,29).
In conclusion, there may be basal differences in brain activity in people with
eating disorders, but this appears to vary according to clinical features such as
age, symptom pattern, and metabolic state.
The Brain in Relation to Food and Appetite

One of the obvious paradigms to be investigated in functional studies in people
with eating disorders is the response to food. This system has been studied both
in animals and in humans. We are therefore digressing from the focus on people
with eating disorders to discuss what is known about the central control of
appetite.
Animal Models
The control of food intake is of central importance to all living beings and
requires a complex integration between the individual and the environment.
The central information processing pathways for food stimuli in the visual,
olfactory, and gustatory modalities in the primate brain have been studied in
depth. Rolls (30,31) has summarized the findings from his meticulous series of
experiments involving electrophysiological recording from individual neu rons.
The initial processing of food stimuli occurs in the inferior temporal visual
cortex (visual), olfactory bulb and piriform cortex (smell), and nucleus of the
solitary tract, thalamus, and insula (taste). These areas project to the amygdala,
orbitofrontal cortex, lateral hypothalamus, and striatum where the reward value
of the stimuli is calibrated (Fig. 1).
FIGURE 1 Schematic illustration of the processing of food-related stimuli. It is based on

data from Rolls (30, 31).
The reward value assigned to food-related stimuli is modulated by integrated

information from peripheral metabolism and the gastrointestinal tract.
Dopaminergic projections to the ventral striatum and other limbic structures
play a major role in food reward- and food-driven behavior (32). The lateral
hypothalamus, in particular the perifornical area, is also implicated in the
modulation of the rewarding aspect of food. For example, weight loss (25%
body weight) enhances the rewarding effect of stimulation in this area whereas
leptin administration produces the opposite effect and attenuates the effectiveness
of the rewarding stimulus for up to 4 days (33). Thus, this particular reward
system is only responsive to long-term shifts in energy balance. Acute food
deprivation for up to 48 h does not have an effect on these neurons. This
suggests that in states of chronic starvation with low leptin, such as AN, areas of
the brain associated with food reward should become more sensitive to food
stimuli.
In conclusion, research into the central reaction to food needs to consider
both the background metabolic state and the hedonic properties.
Brain Imaging of the Appetite Control System in

Humans
Food
Extrinsic food cues in a variety of sensory modalities lead to brain activation.
Visual images of food contrasted with nonfood produced changes in activation in
the left temperoinsular cortical region. The level of activation correlated with
the subjective rating of hunger (34,35). Taste and olfactory stimuli activated the
orbitofrontal cortex (36).
Hunger and Satiety

Several studies have investigated brain activation in response to various tates of
hunger and satiation and the sites of activation are shown in Table 1. Tataranni
and colleagues have done a series of studies in which they have used PET
scanning after a 36 h fast in men (37) and women (42) at normal weight and in
people with obesity (44). The design of this study was limited in that there was
no direct evidence that the activation in the fasting state was related to hunger.
Others have used different designs to ensure that focus of brain activity is on
food or appetite in the fasting state. Thus, Morris and Dolan had an interesting
design in that they used a cognitive task (memory for food/nonfood items) as a
marker for hunger (induced by fasting for 16 h) (39). They thus were able to
correlate activation in PET scanning in various brain areas both with hunger and
performance on this task. LaBar used visual foodrelated cues (38). In summary,
many of these studies find that the orbitofrontal cortex, anterior cingulate,
amygdala, insula, and hypothalamus are more activated in states of hunger
(Fig. 2).
Studies examining the process of satiation are easier to design in that they
follow differences in activation after food consumption (for summary, see
Table 1). Several studies show increased activation in the dorsolateral and
anteror medial prefrontal cortex, posterior insula, and hypothalamus after
satiation (Fig. 3). This change in activation during satiation parallels changes in
metabolic factors. The increase in insulin is linked to changes in cerebral blood
flow in the hypothalamus, insula, and orbitofrontal cortex (40). Quantitative
rather then qualitative differences were found between genders in that the
women appeared to have less hunger and were more satiated. In the hungry
state men had greater activation in the frontotemporal and paralimbic areas.
After satiation women had greater activation in the dorsolateral prefrontal
cortex and the occipital and parietal sensory association areas (42).
TABLE 1 Changes in Brain Activation Relating to States of Hunger and Satiation

FIGURE 2 Areas of the brain activated in states of hunger (the orexogenic network).
The vertical lines are areas activated by hunger only and the crosshatched areas are those
activated by both hunger and satiety.
FIGURE 3 Areas of the brain activated by satiation (the anorexogenic network). The
horizontal lines are areas activated by hunger only and the cross-hatched areas are those
activated by both hunger and satiety.
There are hedonic as well as metabolic components to the process of

satiation. Sensory-specific satiety is the process by which a food substance
becomes less rewarding after a substantial amount of it has been eaten
independent of the calorie content eaten and the metabolic state (45). Several
loci in the orbitofrontal and medial prefrontal cortex are activated by olfactory
food stimuli in a sensory-specific manner, i.e., they are activated only to food
that has not been eaten to satiety (46). This emphasizes the role of the
orbitofrontal and medial prefrontal cortex in modulating the reward associated
with food.
In summary, the human central appetite control system is composed of an
orexigenic (i.e., appetite promoting) network and an anorexigenic (i.e.
inhibitory conrol) circuit; the balance between these two subsystems
determines eating behavior. Both external (visual, taste, or olfactory) and
internal appetite cues are associated with changes in blood flow in the
orexogenic system. The orexigenic network (consisting of orbitofrontal and
insular cortices, amygdala, and hypothalamus) activates with fasting and
promotes feeding behavior. The inhibitory, anorexigenic circuit consists of
anterior ventromedial and dorsolateral prefrontal cortices and acts to
terminate eating, probably by direct inhibition of the orexigenic
system (8,12,15,16, 47,48).
Food Challenges in Eating Disorders

External and internal stimuli (food cues or eating) have been used as specific
challenges in people with an eating disorder. The first specific-challenge,
neuroimaging study, using SPECT, reported a significant increase in perfusion of
the frontal lobes in reaction to eating a cake in people with AN (49). In a later
study people with AN were shown to have the opposite patterns of regional
cerebral blood flow (rCBF) to those with BN: in BN there was increased
frontotemporal perfusion in the resting state which decreased when the patient
ate a cake; in the AN group decreased perfusion at rest was followed by
increased perfusion after eating (again most markedly in frontal and temporal
lobe) (50). More recently, the diagnostic subgroups of bingeing-purging,
anorexia nervosa (BPAN), and restricting anorexia nervosa (RAN) were also
found to have contrasting patterns of brain activity in response to food. BPAN
but not RAN showed increases in glucose metabolism in the right prefrontal and
parietal regions in response to the sight of a custard cake (51). The importance
of specifying between the different subtypes of eating disorders may explain the
negative results of an earlier study (52).
Karhunen and colleagues (53), also using SPECT, found that exposure to
food was associated with different changes in the cerebral blood flow in obese
women who binge-ate compared to either normal-weight or obese controls

with no binge eating. The binge eating group had increased flow in the left
hemisphere, especially in the frontal and prefrontal regions. There were strong
correlations between frontal flow and hunger during exposure to the food. In a
different study following a person with binge eating disorder over time in the
binge eating state there was a bilateral increase in global blood flow whereas in
the restrictive phase flow on the right-hand side of the brain was diminished
relative to that on the left (54). This suggests that food stimuli elicit frontal/
temporal activation in the bulimic disorders across the weight spectmm. This
pattern is different from that found in people with restricting AN who have
reduced activation. Unfortunately, the resolution of SPECT studies is not
sufficient to distinguish specifically between the orexogenic and anorexogenic
pathways.
Recently, functional MRI (fMRI) has been introduced into eating disorder
research, bringing the potential of better spatial-temporal resolution. High-
calorie drinks versus low calorie drinks elicited activity in the orexogenic circuit
including amygdala, prefrontal cortex, and insula (55). On the other hand,
there was no differential activation in this circuit when high- and low-calorie
foods were used both in MRI (Ellison, personal commimication) and in
PET (34). One possible explanation for this finding is that there is no dose-
response relationship between activation of the orexogenic circuit to visual cues
of foods that differ in terms of calorie content. The hedonic properties between
foods, which may vary among individuals, may have a greater impact on
activation than calorie content. Thus food/nonfood or liked/not-liked food may
be a better experimental design for such studies.
In an fMRI study using food/nonfood stimuli (color photographs), the medial
orbitofrontal cortex was activated by food in AN patients but not in healthy
controls (56). The same paradigm was used to investigate the response to food
cues in various subtypes of eating disorder and in people who have recovered
from AN (57). The results from this study are summarized in Table 2 and
Figure 3. In response to food stimuli, patients with eating disorder (either AN
or BN) recruit the orbitofrontal cortex and the anterior cingulate instead of the
lateral prefrontal cortex, inferior parietal lobule, precuneus, and posterior
cerebellum, which are activated in the comparison group; most of these
differences are left sided. The anterior cingulate is not activated by food in those
currently ill with restricting AN. Activation of the dorsolateral prefrontal area
differs across eating disorder subtypes and is reduced in BN. People who had
recovered from AN activated the same lateral prefrontal area as did the healthy
comparison group. Failure to activate this area with food may be considered as a
state marker of an eating disorder Figure 4.
TABLE 2 Brain Regions Significantly Differentially Activated in ED and Its Subgroups as

Compared to Controls
Brain regions: MO-PFC, medial and orbital prefrontal cortex; ACC, anterior cingulate
cortex; PCC, posterior cingulate cortex; L-PFC, lateral prefrontal cortex; IPL, inferior
parietal lobule; cerebell, cerebellum. ↑ significantly more active than in controls; ↓
significantly less active than in controls.
BODY IMAGE CHALLENGE

The concept of body image is less easy to study than food as there are no animal
models on which to base experimental work in order to test a hypothesis.
Damasio conceptualizes body image as a complex function of a widely
distributed network spanning structures from the brainstem to the associative
cortices (58).
In a preliminary study people with AN were challenged with an image of
their own body distorted to maximal unacceptability. There was activation of the
right amygdala, right fusiform, and brainstem. The authors interpreted this to
be activation of the fear network (59). Thus, it may be possible to examine this
controversial aspect of the psychopathology of eating disorders in more detail
using scanning techniques.
NEUROTRANSMISSION
Neuroimaging techniques have recently been used to examine the
neuropharmacology of eating disorders. The monoamines dopamine and
serotonin have an important role in feeding behavior in animals and humans.
Dopamine is linked to reward and it may modulate the hedonic response to
food. Serotonin inhibits feeding. The serotonin hypothesis of eating disorders
has as its essence the idea that abnormal 5-HT function may be a trait
vulnerability for eating disorders.
FIGURE 4 Areas of the brain activated by visual food stimuli in people with eating
disorders. It is based on functional fMRI studies (56, 57).
People with bulimia nervosa were found to have a 17% reduction in the
levels of serotonin transporter in the hypothalamus and thalamus and a 15%
reduction in levels of dopamine transporter in the striatum (60). A reduction in
serotonin transporter was also found in obese binge eaters (61).
In people with obesity an inverse relationship was found between BMI and
striatal dopamine receptor availability. Thus, this preliminary evidence suggests
that abnormal monamine systems may be associated with disinhibited eating.
The serotonin theory has also been tested by examining 5-HT receptor levels
in people who have recovered from an eating disorder. Reduced levels of
5-HT2A binding were found in the mesial temporal cortex (including the
amygdala and hippocampus) and to a lesser degree in the cingulate of women
with a history of restricting AN (62). The mechanism underlying this
phenomenon is unknown. The reduced receptor binding may be a process
compensating for increased extracellular 5-HT. (Earlier research had indeed
found increased 5-HT in the CSF of people who had recovered from AN; 63).
A similar study was undertaken with a smaller group of patients who had
recovered from BN. These people had reduced 5-HT2A in the lateral and
medial orbitofrontal cortex (64). Furthermore, this group of patients did not
have the expected age-related decline in 5-HT2A receptors. In conclusion,
abnormal levels of 5-HT2A receptor are found in various sites of the orexogenic
network after recovery from an eating disorder.
WHAT CAN WE UNDERSTAND FROM THE

FUNCTIONAL STUDIES?
Across the spectrum of eating disorders, in response to either extrinsic or
intrinsic food-related cues, there are changes in activity in the frontal and limbic
areas that form the orexogenic and anorexogenic systems. The functional
imaging studies implicate, in particular, the orbitofrontal cortex and the lateral
prefrontal cortex as key areas of differential activity in people with eating
disorders. The functional response to food differs between the subtypes of
eating disorders in both SPECT and fMRI studies. People with bulimic symptoms
had reduced activation of the dorsolateral prefrontal cortex.
Interestingly, people who had recovered from restricting AN had some
activation (in the orbitofrontal cortex and anterior cingulate) similar to that
found in people with a current eating disorders but also activated parts of the
brain responsive to food cues in non-eating disorder controls (such as the apical
and lateral prefrontal cortex). This suggests that there are both state and trait
functional differences in response to food in people with eating disorders.
LINKING STRUCTURE AND FUNCTION

The orbitofrontal cortex and lateral frontal areas are of key importance in many
other aspects of brain function beside the control of appetite (65), e.g.,
psychopathology (65) and personality (66). It is interesting to examine whether
this can enrich theories that explain the mechanisms that underlie eating disorders.
The Orbitofrontal Cortex

The orbitofrontal cortex is involved in the appraisal and processing of subjectively
salient and self-related stimuli and is critically involved in the motivational
control of goal-directed behavior (67,68). Thus, apart from the primary
reinforcers, such as food and touch, in humans, secondary reinforcers, such as
money and drugs of abuse (e.g., cocaine) activate this area (69).
Activation in the orbitofrontal cortex occurs to both positive and negative
rewards (70). There is some preliminary evidence of anatomical specificity
between positive and negative reinforcement for touch (36). However, the
resolution of fMRI may be insufficient to distinguish between positive and
negative rewards for other stimuli such as food because in animals positive and
negative rewards are computed on individual neurons.
There are interesting parallels between people with damage to the
orbitofrontal cortex and people with AN. For example, people with ventral
frontal lobe damage have impaired identification of facial and vocal expression.
This disability correlates with the degree of alteration of their emotional
experience (68). Impaired identification of facial and vocal expression is present
also in AN (71). Animals with lesions of the prefrontal cortex experience an
interceptive agnosia, i.e., an inability to determine how the body is reacting
when faced with contrasting behavioral options. This resonates with the clinical
descriptions of alexithymia (72,73) and poor interoceptive awareness in AN (74).
The Lateral Prefrontal Cortex

The dorsolateral prefrontal circuit is involved in executive functions (75). For
example, lesions in the right lateral prefrontal cortex produce defects in
working memory (76). These areas are activated in set shifting tasks (77). This
is of interest as we have found people with eating disorders to be impaired in
these types of tasks. Abnormal activation in these circuits has also been found in
depression, obsessive-compulsive disorder, schizophrenia, and substance
abuse (75).
POSSIBLE MECHANISMS UNDERPINNING THE

STRUCTURAL AND FUNCTIONAL BRAIN
ABNORMALITIES IN EATING DISORDERS
The state and trait abnormalities found in both brain structure and function have
implications for etiological models. The trait abnormalities may result from
neurodevelopmental mechanisms or from damage or “scaring” from the illness.
The neurodevelopmental theory posits that programming either from genetic
mechanisms or from early environmental experiences causes the structural and
functional abnormalities in the brain. These may be linked to variations in
neurotransmitter function, e.g., 5-HT.
Eating Behaviors as Part of a Neurodevelopmental

Process
Research into the development of the central control of appetite is burgeoning.
Many of the genetic causes of obesity that are now being described result from
abnormalities in the central control of appetite with overeating as a
consequence. For example, one of the commonest monogenic causes of obesity
relates to mutations in the melanocortin receptor (78). It means that
phenotypes of obesity that focus on abnormal eating behaviors may be more
pertinent than those with a focus on metabolic abnormalities.
Eating behaviors run in families. In a study on the Amish the heritability for
disinhibition of eating was found to be 40% and for restraint 30% (79). The
mechanisms underpinning these effects are uncertain. Extreme eating behaviors
may be caused by increased hypothalamic drive, but also they may be moderated
by variations in olfactory and gustatory sensory perception (80) or in the reward
system (81).
Also, early environmental effects may set the level of activity of the appetite
system, a type of metabolic imprinting or programming (82,83). One such theory
is that poor fetal growth with malnutrition early in development may predispose
to a thrifty phenotype (84), whereas poor nutrition later in pregnancy or in
early development may protect people from developing obesity (85,86).
Continuities in the patterns of eating persist over childhood. For example, poor
sucking in infancy is associated with picky eating at ages of 3.5 and 5.5 (87).
Anorexia Nervosa as a Neurodevelopmental

Disorder
Gillberg has suggested that eating disorders, in particular AN, should be
considered to be a neurodevelopmental disorder (88). He bases this argument
on the presence of neurological soft signs. He found dysdiadochokinesis in some

16-year-olds with AN, which was still present when they were 21. We have also
found that dysdiadokinesis is present in a proportion of adults with AN. This is
present in the context of other abnormalities in neuropsychological testing such
as a difficulty in set shifting tasks that also remain after recovery (89,90).
Early environmental adversities, such as maternal stress and anxiety during
pregnancy, perinatal hypoxic injury, or postinfective autoimmune reactions, are
associated with aberrant brain development (91,92). People with AN have a
higher exposure to these risk factors. Maternal anxiety, complications during
pregnancy, birth injury, being small for date, and a seasonal birth pattern have
all been implicated as risk factors in case control studies (93–96). Links with
later infective etiology have also been made such as the association between AN
and pediatric autoimmune neuropsychiatric disorder (PANDAS) (97,98).
Further research that links functional neuroimaging with research into genetic
and childhood risk factors is needed to develop these ideas further.
Brain Abnormalities in Anorexia Nervosa

Consequent on the Illness (Scars)
Severe malnutrition during the critical period of adolescence may arrest brain
maturation or cause the dysfunctional development of brain structure and
function. An obvious possibility is that it is caused by vitamin or essential fatty
acid deficiencies. Alternatively, it may result from an epiphenomenona of AN,
such as high levels of cortisol, which has been implicated in the
neuropathological changes in mood disorders (99) and schizophrenia. No clear
relationships between structural and functional brain changes and any putative
etiological factor have been reported. However, these studies often use data
from cross-sectional measures, and it is likely that it is the integrated effect of
deficiencies over time, that is the relevant factor.
CONCLUSION
The information produced by brain scanning has implications for models used in
attempts to explain the etiology and maintenance of eating disorders. The
findings are consistent with a neurodevelopmental etiologic process. However,
there is also substantial evidence to suggest that many of the changes represent
acquired damage. Eating disorders occur at a critical time of brain development,
and toxic effects from malnutrition or stress may cause atrophic or dystrophic
changes. There is no doubt that these studies have generated many new ideas
about symptoms, causes, and consequences of eating disorders, and further
research linking genetics, precise definitions of the phenotypes of eating
disorders, and scanning has great potential to increase our knowledge about
these confusing conditions.
SUMMARY
• The technologies used to image the brain have a different balance of

strengths and weaknesses. Each of them is suited to answer different types of
questions.
• The onset of eating disorders coincides with a transitional flux in brain
structure and function that may have implications for both etiology and
prognosis.
• Cerebral volume is reduced and the gray matter loss may persist in AN.
Smaller structural changes are seen in BN.
• Lesions in the prefrontal and temporal cortices and mesiotemporal
structures and hypothalamus predominantly on the right-hand side are
particularly linked with eating disorder symptoms. However, in the majority
of cases of AN there is no macroscopic brain lesion.
• The central control of appetite, nutrition, and metabolism involves pathways
that recognize food in the environment and modulate the response to these
cues depending-on the current metabolic state and expected hedonic effect.
• The orbito frontal cortex, anterior cingulate, amygdala, insula and
hypothalamus form the orexogenic system, which is activated by hunger.
• The dorsolateral and anterior medial prefrontal cortex, posterior insula and
the hypothalamus form the anorexogenic system activated by satiation.
• The orbitofrontal cortex and the lateral prefrontal cortex are key areas of
differential activity between people with eating disorders and those without
in one fMRI study.
• In studies using SPECT the pattern of activation in response to food cues
differed between people who binge ate in comparison to pure restrictors.
People with bulimic symptoms had reduced activation of the dorso lateral
prefrontal cortex in one fMRI study.
• People who had recovered from restricting anorexia nervosa had some
activation (in the orbitofrontal cortex and anterior cingulate) similar to that
found in people with a current eating disorders but also activated parts of the
brain responsive to food cues in non eating disorder controls (such as the
apical and lateral prefrontal cortex).
• There are two theories that may account for abnormal brain structure or
function in people with eating disorders. (a) Neurodevelopmental disorders
and brain abnormalities precede the onset of the illness. (b) The brain
abnormalities are consequent to the illness and result from starvation or

other aspects of the illness.
ACKNOWLEDGMENTS
This study was supported by grant QLKl-1999–916 from the
European Commission Framework V program http://www.cordis.lu/life/
home.html), the Nina Jackson Eating Disorders Research Charity, and a
Wellcome Trust Travelling Fellowship for Dr. R.Uher.
REFERENCES
1. Durston S, Hulshoff Pol HE, Casey BJ, Giedd JN, Buitelaar JK, van Engeland H.
Anatomical MRI of the developing human brain: what have we learned? J Am Acad
Child Adol Psychiatry 2001; 40(9): 1012–1020.
2. Huttenlocher PR. Synaptic density in human frontal cortex: developmental changes
and the effects of aging. Brain Res 1979; 163(2): 195–205.
3. McGivern R, Andersen J, Byrd D, Mutter K, Reilly J. Cognitive efficiency on a
match to sample task decreases at the onset of puberty in children. Brain Cogn
2002; 50(1):73.
4. Casey BJ, Trainor R, Giedd J, Vauss Y, Vaituzis CK, Hamburger S, et al. The role
of the anterior cingulate in automatic and controlled processes: a developmental
neuroanatomical study. Dev Psychobiol 1997; 30(l):61–69.
5. Dolan RJ, Mitchell J, Wakeling A. Structural brain changes in patients with
anorexia nervosa. Psychol Med 1988; 18(2):349–353.
6. Kohn MR, Ashtari M, Golden NH, Schebendach J, Patel M, Jacobson MS, et al.
Structural brain changes and malnutrition in anorexia nervosa. Ann N Y Acad Sci
1997; 817:398–399.
7. Kingston K, Szmukler G, Andrewes D, Tress B, Desmond P. Neuropsychological
and structural brain changes in anorexia nervosa before and after refeeding. Psychol
Med 1996; 26(l):15–28.
8. Krieg JC, Pirke KM, Lauer C, Backmund H. Endocrine, metabolic, and cranial
computed tomographic findings in anorexia nervosa. Biol Psychiatry 1988; 23(4):
377–387.
9. Swayze VW, Andersen A, Arndt S, Rajarethinam R, Fleming F, Sato Y, et al.
Reversibility of brain tissue loss in anorexia nervosa assessed with a computerized
Talairach 3-D proportional grid. Psychol Med 1996; 26(2):381–390.
anorexia nervosa. J Pediatr 1996; 129(6):794–803.
11. Golden NH, Ashtari M, Kohn MR, Patel M, Jacobson MS, Fletcher A, et al.
Reversibility of cerebral ventricular enlargement in anorexia nervosa,
demonstrated by quantitative magnetic resonance imaging. J Pediatr 1996; 128(2):

296–301.
12. Lambe EK, Katzman DK, Mikulis DJ, Kennedy SH, Zipursky RB. Cerebral gray
matter volume deficits after weight recovery from anorexia nervosa. Arch Gen
Psychiatry 1997; 54(6):537–542.
13. Neumarker KJ, Dudeck U, Meyer U, Neumarker U, Schulz E, Schonheit B.
Anorexia nervosa and sudden death in childhood: clinical data and results obtained
from quantitative neurohistological investigations of cortical neurons. Eur Arch
Psychiatry Clin Neurosci 1997; 247(l):16–22.
14. Rosoklija G, Toomayan G, Ellis SP, Keilp J, Mann JJ, Latov N, et al.
Structural abnormalities of subicular dendrites in subjects with schizophrenia and
mood disorders: preliminary findings. Arch Gen Psychiatry 2000; 57(4):349–356.
15. Doraiswamy PM, Krishnan KR, Figiel GS, Husain MM, Boyko OB, Rockwell WJ,
et al. A brain magnetic resonance imaging study of pituitary gland morphology in
anorexia nervosa and bulimia. Biol Psychiatry 1990; 28(2): 110–116.
16. Giordano GD, Renzetti P, Parodi RC, Foppiani L, Zandrino F, Giordano Gm, et
al. Volume measurement with magnetic resonance imaging of
hippocampusamygdala formation in patients with anorexia nervosa. J Endocrinol
Invest 2001; 24(7):510–514.
17. Krieg JC, Backmund H, Pirke KM. Cranial computed tomography findings in
bulimia. Acta Psychiatr Scand 1987; 75(2): 144–149.
18. Hoffman GW, Ellinwood EH Jr, Rockwell WJ, Herfkens RJ, Nishita JK, Guthrie
LF. Cerebral atrophy in bulimia. Biol Psychiatry 1989; 25(7):894–902.
19. Hoffman GW, Ellinwood EH Jr, Rockwell WJ, Herfkens RJ, Nishita JK, Guthrie
LF. Brain Tl measured by magnetic resonance imaging in bulimia. Biol Psychiatry
1990; 27(1):116–119.
20. Roser W, Bubl R, Buergin D, Seelig J, Radue EW, Rost B. Metabolic changes in the
brain of patients with anorexia and bulimia nervosa as detected by proton magnetic
resonance spectroscopy. Int J Eat Disord 1999; 26(2): 119–136.
21. Uher R, Treasure J, Campbell IC. Neuranatomical bases of eating disorders. In:
H’haenen HA, den Boer JA, Willner P, eds. Biological Psychiatry. John Wiley &
Sons: Chichester, 2002:1173–1180.
22. Delvenne V, Goldman S, De MV, Lotstra F. Brain glucose metabolism in eating
disorders assessed by positron emission tomography. Int J Eat Disord 1999; 25(l):
29–37.
23. Delvenne V, Goldman S, Simon Y, De MV, Lotstra F. Brain hypometabolism of
glucose in bulimia nervosa. Int J Eat Disord 1997; 21(4):313–320.
24. Herholz K, Krieg JC, Emrich HM, Pawlik G, Beil C, Pirke KM, et al. Regional
cerebral glucose metabolism in anorexia nervosa measured by positron emission
tomography. Biol Psychiatry 1987; 22(1):43–51.
25. Gordon I, Lask B, Bryant-Waugh R, Christie D, Timimi S. Childhood-onset
anorexia nervosa: towards identifying a biological substrate. Int J Eat Disord 1997;
22(2): 159–165.
26. Naruo T, Nakabeppu Y, Deguchi D, Nagai N, Tsutsui J, Nakajo M, et al.

Decreases in blood perfusion of the anterior cingulate gyri in anorexia nervosa
restricters assessed by SPECT image analysis. BMC Psychiatry 2001; 1(1):22.
27. Takano A, Shiga T, Kitagawa N, Koyama T, Katoh C, Tsukamoto E, et al. Abnormal
neuronal network in anorexia nervosa studied with I-123-IMP SPECT. Psychiatry
Res 2001; 107(1):45–50.
28. Andreason PJ, Altemus M, Zametkin AJ, King AC, Lucinio J, Cohen RM.
Regional cerebral glucose metabolism in bulimia nervosa. Am J Psychiatry 1992;
149(11):1506–1513.
29. Frank GK, Kaye WH, Greer P, Meltzer CC, Price JC. Regional cerebral blood
flow after recovery from bulimia nervosa. Psychiatry Res 2000; 100(l):31–39.
30. Rolls ET, Baylis LL. Gustatory, olfactory, and visual convergence within the
primate orbitofrontal cortex. J Neurosci 1994; 14(9):5437–5452.
31. Rolls ET, Critchley HD, Browning A, Hernadi I. The neurophysiology of taste and
olfaction in primates, and umami flavor. Ann N Y Acad Sci 1998; 855:426–437.
32. Wise RA, Rompre PP. Brain dopamine and reward. Annu Rev Psychol 1989; 40:
191–225.
33. Fulton S, Woodside B, Shizgal P. Modulation of brain reward circuitry by leptin.
Science 2000; 287(5450): 125–128.
34. Gordon CM, Dougherty DD, Fischman AJ, Emans SJ, Grace E, Lamm R, et al.
Neural substrates of anorexia nervosa: a behavioral challenge study with positron
emission tomography. J Pediatr 2001; 139(l):51–57.
35. Gordon CM, Dougherty DD, Rauch SL, Emans SJ, Grace E, Lamm R, et al.
Neuroanatomy of human appetitive function: a positron emission tomography
investigation. Int J Eat Disord 2000; 27(2): 163–171.
36. Francis S, Rolls ET, Bowtell R, McGlone F, O’Doherty J, Browning A, et al. The
representation of pleasant touch in the brain and its relationship with taste and
olfactory areas. Neuroreport 1999; 10(3):453–459.
37. Tataranni PA, Gautier JF, Chen K, Uecker A, Bandy D, Salbe AD, et al.
Neuroanatomical correlates of hunger and satiation in humans using positron
emission tomography. Proc Natl Acad Sci USA 1999; 96 AB(8):4569–4574.
38. LaBar KS, Gitelman DR, Parrish TB, Kim YH, Nobre AC, Mesulam MM. Hunger
selectively modulates corticolimbic activation to food stimuli in humans. Behav
Neurosci 2001; 115(2):493–500.
39. Morris JS, Dolan RJ. Involvement of human amygdala and orbitofrontal cortex in
hunger-enhanced memory for food stimuli. J Neurosci 2001; 21(14):5304–5310.
40. Liu Y, Gao JH, Liu HL, Fox PT. The temporal response of the brain after eating
revealed by functional MRI. Nature 2000; 405(6790): 1058–1062.
41. Del Parigi A, Gautier JF, Chen K, Salbe AD, Ravussin E, Reiman E, et al.
Neuroimaging and obesity: mapping the brain responses to hunger and satiation in
humans using positron emission tomography. Ann N Y Acad Sci 2002; 967:
389–397.
42. Del Parigi A, Chen K, Gautier JF, Salbe AD, Pratley RE, Ravussin E, et al. Sex
differences in the human brain’s response to hunger and satiation. Am J Clin Nutr
2002; 75(6):1017–1022.
43. Small DM, Zatorre RJ, Dagher A, Evans AC, Jones-Gotman M. Changes in brain
activity related to eating chocolate: from pleasure to aversion. Brain 2001; 124;
(Pt9):1720–1733.
44. Gautier JF, Chen K, Uecker A, Bandy D, Frost J, Salbe AD, et al. Regions of the
human brain affected during a liquid-meal taste perception in the fasting state: a
positron emission tomography study. Am J Clin Nutr 1999; 70(5):806–810.
45. Raynor HA, Epstein LH. Dietary variety, energy regulation, and obesity. Psychol
Bull 2001; 127(3):325–341.
46. O’Doherty J, Rolls ET, Francis S, Bowtell R, McGlone F, Kobal G, et al. Sensory-
specific satiety-related olfactory activation of the human orbitofrontal cortex.
Neuroreport 2000; ll(4):893–897.
anorexia nervosa [see comments]. J Pediatr 1996; 129(6):794–803.
48. Kingston K, Szmukler G, Andrewes D, Tress B, Desmond P. Neuropsychological
and structural brain changes in anorexia nervosa before and after refeeding. Psychol
Med 1996; 26(l):15–28.
49. Nozoe S, Naruo T, Nakabeppu Y, Soejima Y, Nakajo M, Tanaka H. Changes in
regional cerebral blood flow in patients with anorexia nervosa detected through
single photon emission tomography imaging. Biol Psychiatry 1993; 34(8):578–580.
50. Nozoe S, Naruo T, Yonekura R, Nakabeppu Y, Soejima Y, Nagai N, et al.
Comparison of regional cerebral blood flow in patients with eating disorders. Brain
Res Bull 1995; 36(3):251–255.
51. Naruo T, Nakabeppu Y, Sagiyama K, Munemoto T, Homan N, Deguchi D, et al.
Characteristic regional cerebral blood flow patterns in anorexia nervosa patients
with binge/purge behavior. Am J Psychiatry 2000; 157(9): 1520–1522.
52. Krieg JC, Lauer C, Leinsinger G, Pahl J, Schreiber W, Pirke KM, et al. Brain
morphology and regional cerebral blood flow in anorexia nervosa. Biol Psychiatry
1989; 25(8): 1041–1048.
53. Karhunen LJ, Vanninen EJ, Kuikka JT, Lappalainen RI, Tiihonen J, Uusitupa MI.
Regional cerebral blood flow during exposure to food in obese binge eating
women. Psychiatry Res 2000; 99(l):29–42.
54. Hirano H, Tomura N, Okane K, Watarai J, Tashiro T. Changes in cerebral blood
flow in bulimia nervosa. J Comput Assist Tomogr 1999; 23(2):280–282.
55. Ellison Z, Foong J, Howard R, Bullmore E, Williams S, Treasure J. Functional
anatomy of calorie fear in anorexia nervosa. Lancet 1998; 352(9135):! 192.
56. Uher R, Murphy T, Brammer M, Dalgleish T, Phillips M, Ng V, et al. Functional
neural correlates of eating disorders, personal communication, 2002
57. Uher R. Recovery and chronicity in anorexia nervosa: brain activity associated with
differential outcomes. Biol Psychiatry. (In press).
58. Damasio AR. How the brain creates the mind. Sci Am 1999; 281(6):112–117.
59. Seeger G, Braus DF, Ruf M, Goldberger U, Schmidt MH. Body image distortion
reveals amygdala activation in patients with anorexia nervosa—a functional
magnetic resonance imaging study. Neurosci Lett 2002; 326(l):25–28.
60. Tauscher J, Pirker W, Willeit M, de Zwaan M, Bailer U, Neumeister A, et al.
[1231] Beta-CIT and single photon emission computed tomography reveal reduced
brain serotonin transporter availability in bulimia nervosa. Biol Psychiatry 2001; 49
(4):326–332.
61. Kuikka JT, Tammela L, Karhunen L, Rissanen A, Bergstrom KA, Naukkarinen H,
et al. Reduced serotonin transporter binding in binge eating women.
Psychopharmacology (Berl) 2001; 155(3):310–314.
62. Frank GK, Kaye WH, Meltzer CC, Price JC, Greer P, McConaha C, et al.
Reduced 5-HT2A receptor binding after recovery from anorexia nervosa. Biol
Psychiatry 2002; 52(9):896–906.
63. Kaye WH, Gwirtsman HE, George DT, Ebert MH. Altered serotonin activity in
anorexia nervosa after long-term weight restoration. Does elevated
cerebrospinal fluid 5-hydroxyindoleacetic acid level correlate with rigid and
obsessive behavior? Arch Gen Psychiatry 1991; 48(6):556–562.
64. Kaye WH, Frank GK, Meltzer CC, Price JC, McConaha CW, Crossan PJ, et al.
Altered serotonin 2A receptor activity in women who have recovered from bulimia
nervosa. Am J Psychiatry 2001; 158(7): 1152–1155.
65. Joseph R. Frontal lobe psychopathology: mania, depression, confabulation,
catatonia, perseveration, obsessive compulsions, and schizophrenia. Psychiatry
1999; 62(2): 138–172.
66. Cloninger RC. Functional neuroanatomy and brain imaging of personality and its
disorders. In: D’haenen H, den Boer JA, Willner P, eds. Biological Psychiatry.
Chichester: John Wiley & Sons Ltd, 2002:1377–1385.
67. Damasio AR. The somatic marker hypothesis and the possible functions of the
prefrontal cortex. Philos Trans R Soc Lond B Biol Sci 1996; 351(1346):
1413–1420.
68. Hornak J, Rolls ET, Wade D. Face and voice expression identification in patients with
emotional and behavioural changes following ventral frontal lobe damage.
Neuropsychologia 1996; 34(4):247–261.
70. Rolls ET. The orbitofrontal cortex and reward. Cereb Cortex 2000; 10(3):
284–294.
71. Kucharska-Pietura K, Nikolaou V, Masiak M, Treasure J. The recognition of
emotion in faces and voice in anorexia nervosa. Int J Eat Disord. In press.
72. Troop NA, Schmidt UH, Treasure JL. Feelings and fantasy in eating disorders: a
factor analysis of the Toronto Alexithymia Scale. Int J Eat Disord 1995; 18(2):
151–157.
73. Cochrane CE, Brewerton TD, Wilson DB, Hodges EL. Alexithymia in the eating
disorders. Int J Eat Disord 1993; 14(2):219–222.
75. Tekin S, Cummings JL. Frontal-subcortical neuronal circuits and clinical

neuropsychiatry: an update. J Psychosom Res 2002; 53(2):647–654.
76. Bechara A, Damasio H, Damasio AR. Emotion, decision making and the
orbitofrontal cortex. Cereb Cortex 2000; 10(3):295–307.
77. Konishi S, Hayashi T, Uchida I, Kikyo H, Takahashi E, Miyashita Y. Hemispheric
asymmetry in human lateral prefrontal cortex during cognitive set shifting. Proc
Natl Acad Sci USA 2002; 99(11):7803–7808.
78. Farooqi IS, Yeo GS, Keogh JM, Aminian S, Jebb SA, Butler G, et al. Dominant and
recessive inheritance of morbid obesity associated with melanocortin 4 receptor
deficiency. J Clin Invest 2000; 106(2):271–279.
79. Steinle NI, Hsueh WC, Snitker S, Pollin TI, Sakul H, St Jean PL, et al. Eating
behavior in the Old Order Amish: heritability analysis and a genome-wide linkage
analysis. Am J Clin Nutr 2002; 75(6): 1098–1106.
80. Lindemann B. Receptors and transduction in taste. Nature 2001; 413 (6852):219–
225.
81. Pelchat ML. Of human bondage: food craving, obsession, compulsion, and
addiction. Physiol Behav 2002; 76(3):347–352.
82. Barker DJ. In utero programming of chronic disease. Clin Sci (Lond) 1998; 95(2):
115–128.
83. Waterland RA, Garza C. Potential mechanisms of metabolic imprinting that lead to
chronic disease. Am J Clin Nutr 1999; 69(2): 179–197.
84. Hales CN, Barker DJ. The thrifty phenotype hypothesis. Br Med Bull 2001; 60:
5–20.
85. Ravelli AC, Der Meulen JH, Osmond C, Barker DJ, Bleker OP. Obesity at the age
of 50 y in men and women exposed to famine prenatally. Am J Clin Nutr 1999; 70
(5):811–816.
86. Ravelli GP, Stein ZA, Susser MW. Obesity in young men after famine exposure in
utero and early infancy. N Engl J Med 1976; 295(7):349–353.
87. Jacobi C, Agras WS, Bryson S, Hammer LD. Behavioral validation, precursors, and
concomitants of picky eating in childhood. J Am Acad Child Adol Psychiatry 2003;
42(l):76–84.
88. Gillberg C, Rastam M, Gillberg IC. Anorexia nervosa: physical health and
neurodevelopment at 16 and 21 years. Dev Med Child Neurol 1994; 36(7):
567–575.
89. Tchanturia, Morris RG, Brecelj Anderluh, B, Nikolaou V, Treasure J. An
examination of set shifting in anorexia nervosa before and after weight gain and in
full recovery and links to childhood and adult OCPD Traits. Br J Psychiatry. In
press.
90. Tchanturia K, Serpell L, Troop N, Treasure J. Perceptual illusions in eating
disorders: rigid and fluctuating styles. J Behav Ther Exp Psychiatry 2001; 32(3):
107–115.
91. Lou HC, Hansen D, Nordentoft M, Pryds O, Jensen F, Nim J, et al. Prenatal
stressors of human life affect fetal brain development. Dev Med Child Neurol
1994; 36(9):826–832.
92. O’Connor TG, Heron J, Golding J, Beveridge M, Glover V. Maternal antenatal

anxiety and children’s behavioural/emotional problems at 4 years. Report from the
Avon Longitudinal Study of Parents and Children. Br J Psychiatry 2002; 180:
502–508.
93. Cnattingius S, Hultman CM, Dahl M, Sparen P. Very preterm birth, birth trauma,
and the risk of anorexia nervosa among girls. Arch Gen Psychiatry 1999; 56(7):
634–638.
94. Eagles JM, Andrew JE, Johnston MI, Easton EA, Millar HR. Season of birth in
females with anorexia nervosa in Northeast Scotland. Int J Eat Disord 2001; 30(2):
167–175.
95. Foley DL, Thacker LR, Aggen SH, Neale MC, Kendler KS. Pregnancy and
perinatal complications associated with risks for common psychiatric disorders in a
population-based sample of female twins. Am J Med Genet 2001; 105(5):
426–431.
96. Shoebridge P, Gowers SG. Parental high concern and adolescent-onset anorexia
nervosa. A case-control study to investigate direction of causality. Br J Psychiatry
2000; 176:132–137.
97. Sokol MS, Ward PE, Tamiya H, Kondo DG, Houston D, Zabriskie JB. D8/17
expression on B lymphocytes in anorexia nervosa. Am J Psychiatry 2002; 159(8):
1430–1432.
98. Sokol MS. Infection-triggered anorexia nervosa in children: clinical description of
four cases. J Child Adol Psychopharmacol 2000; 10(2): 133–145.
99. Cotter DR, Pariante CM, Everall IP. Glial cell abnormalities in major psychiatric
disorders: the evidence and implications. Brain Res Bull 2001; 55(5):585–595.
14
Molecular Biology of Anorexia Nervosa,
Bulimia Nervosa, Binge Eating Disorder, and
Obesity
Dorothy Grice*
University of Pennsylvania
Philadelphia, Pennsylvania, U.S.A.
Eating disorders reflect a wide range of abnormal eating behaviors and

unhealthy attitudes toward food, food consumption, body shape, and body size.
Eating disorders such as anorexia nervosa (AN) and bulimia nervosa (BN) have
been well defined clinically and generally fall under the purview of psychiatric
disorders [see Diagnostic and Statistical Manual of Mental Disorders, 4th ed. DSM-IV
(l)], although less well-studied disorders such as binge eating disorder (BED)
and eating disorder-not otherwise specified (ED-NOS), are also found in this
domain. With the rising appreciation of the biological and genetic basis of
psychiatric disorders and, specific to this chapter, eating disorders, many
researchers have sought to identify the genetic contributions to complex
disorders such as these. This chapter will briefly review molecular genetic
approaches to the study of complex disorders and focus on recent molecular
genetic studies of eating disorders. The preponderance of this literature is based
on studies of AN and BN: however, when possible, studies that examine other
phenotypes, such as BED, will be included. Lastly, we will consider how
genetic research on obesity may inform studies in the psychiatrically defined
eating disorders.
A critical aspect of genetic studies is determining an accurate phenotype,
e.g., the clinical, behavioral, or biological indicators that comprise core
feature(s) of the disorder under study. Phenotypes for eating disorders have
emerged from decades, if not centuries, of in-depth and thoughtful clinical
examinations and interviews. Of the DSM-IV eating disorders, AN and BN are
the most studied. AN is a serious psychiatric disorder characterized by severely
restricted eating, obsessive fears of weight gain, and distorted attitudes toward
eating, food, and body image. As symptoms progress, individuals with AN fail
* Current affiliation: University of Medicine and Dentistry of New Jersey, New Jersey
Medical School, Newark, New Jersey, U.S.A.
324 GRICE
to maintain a minimally normal body weight (or fail to progress through

developmentally appropriate growth phases). Often presenting as a waxing-and-
waning disorder, AN typically has its onset during adolescence. BN, a related
eating disorder, is characterized by recurrent episodes of binge eating followed
by various inappropriate compensatory behaviors of weight control (such as self-
induced vomiting), and a persistent and pathological preoccupation with body
weight and shape. Similar to AN, BN usually has its onset during adolescence,
not uncommonly after an episode of dieting and/or food restriction. The vast
majority of individuals diagnosed with AN or BN are female, with the lifetime
prevalence of AN estimated to be 0.5–1.6% (2,3) and of BN 1.0–3.0 % (4,5).
In women, AN has the highest mortality of the psychiatric disorders (6). AN and
BN have high rates of co-occurrence. During their lifetime, almost 40% of
individuals with AN will also meet criteria for BN (7–9). From this perspective,
AN and BN can be considered heritable eating disorders that likely have both
diseasespecific and overlapping genetic components.
GENETICS
Genetic study of disease yields evidence about the transmission and underlying
biology of the illness, two important factors related to treatment and
prevention. Psychiatric genetics is a very young field and thus far genes for
psychiatric disorders have been difficult to isolate. Eating disorders are complex
disorders, as are most psychiatric diagnoses. Genetic, biological,
developmental, and sociocultural forces contribute to vulnerabilities to and
etiologies of these disorders. These various influences are thought to have direct
and indirect effects, including synergistic influences, on the onset and
continuation of pathological eating disorders. Although clinical descriptions of
anorexic syndromes can be found in the literature dating back to the early
1870s (10,11), it is only in the past few decades that the genetic basis of eating
disorders has come to be appreciated and examined.
Understanding the genetic contributors to eating disorders presents
particular challenges. For example, there are strong social and cultural
presentations of ideal body weight and shape that influence an individual’s self-
perception of body size, in addition to any existing inherent genetic
vulnerabilities in this regard. Inheritance patterns of eating disorders in families,
useful in performing some genetic analyses, are often unclear and do not follow
traditional mendelian patterns. The often secretive and isolated nature of
pathological behaviors associated with eating disorders, particularly when
affected individuals display a normal or near-normal body mass index (BMI),
may compound the difficulty in identifying clinical cases and family patterns of
affectedness. These features, compounded by variability in clinical expression,
MOLECULAR BIOLOGY 325
can obscure phenotype classification. Lastly, from the clinical perspective there
is a spectrum of eating disorder diagnoses and pathology. AN and BN are
considered the most severe eating disorder diagnoses, each with diagnostic
subtypes; however, there is a residual category in DSM-IV that identifies eating
disorders not otherwise specified (ED-NOS), including subthreshold AN, BN,
and BED. Beyond DSM-IV diagnostic categories, many other pathological eating
behaviors and attitudes are associated with disruptions in psychological and/or
physical health, although the specific genetic factors that may contribute to these
states are largely unexamined (12).
GENETIC METHODS
Traditional methods of determining the role of genetic factors in the etiology of
a given disorder are twin, family, and adoption studies. Heritability (the
proportion of the population variance of a trait that can be explained by genetic
transmission) can be estimated from these types of studies. Once heritability is
established for a given disorder, genetic linkage analysis, association studies, and
candidate gene surveys are used to search for genes that underlie a particular
disorder. The bases for sound genetic studies include homogeneous phenotype
classification, unbiased ascertainment of index patients (probands), and accurate
clinical assessments. As discussed above, accurate assessment and definition of
AN and BN phenotypes requires particular attention. Study design should also
include appropriate ascertainment strategies to identify probands and families
and mimimize ascertainment bias (e.g., epidemiological studies or, lacking that,
multiple recruitment strategies to minimize ascertainment influence from a
single referral source) and to gather sufficient numbers of subjects for sound
statistical analysis. In addition to defined homogeneous phenotypes, it is also
important that the specific clinical assessment of study subjects be free from
inadvertent bias, such as can occur when subjects are assessed by multiple
researchers or phenotyped without structured criteria.
IDENTIFYING GENES
Several genetic analysis methods have been used to search for susceptibility
genes for eating disorders. Genetic association analyses, including casecontrol
studies, of candidate genes can identify gene variants that are statistically
associated with illness by comparing allele frequencies of presumably
biologically relevant genes in affected individuals (cases) and control individuals.
In a case-control study, when the allele frequencies of affected individuals are
compared to those of nonfamily controls, the control population should be
carefully selected since there are natural variations in gene frequencies that occur
326 GRICE
between ethnic groups. This phenomenon, known as population stratification,

can introduce errors into gene frequency analyses. Differences in allele
frequencies between an affected group and a mismatched control group can be
misinterpreted as an indication of genetic association, or lack thereof, when in
fact the results are due to population stratification. An alternative to case-
control association studies are family-based (rather than population-based)
association methods.
One example of a family-based method is the transmission disequilibrium
test (TDT) (13). The TDT measures linkage in the presence of association. In
the TDT method, candidate gene allele determinations (genotypes) are done on
affected individuals and their biological parents (about whom no clinical or
diagnostic data is needed). From this group of families, in order to have
unambiguous determination of allele transmissions from parent to child, only
those families with parents having heterozygous genotypes are selected for
further analysis, (Because the child’s genotype is derived from one allele
transmitted from the mother and one allele transmitted from the father, in the
case of a homozygous parent it is not possible to resolve the exact allele
transmission pattern.) The transmission frequency of the targeted allele is then
calculated in the affected subjects and compared to the frequency of its
nontransmission. In this way, the parental allele group of nontransmitted alleles
becomes a perfectly matched genetic control for the transmitted alleles,
bypassing the issue of population stratification.
Candidate gene studies focus on specific, known genes that are hypothesized
to underlie the biology of the disorder being studied; thus, by design, candidate
gene studies do not function at the genomic level. In contrast, genetic linkage
analysis is atheoretical about specific gene involvement in a disorder. Instead,
this method relies on anonymous genetic markers dispersed at fairly close
intervals through the entire genome, making no assumptions about regions of
interest. Typically in a genetic linkage study, multigenera tional families are
identified in which there are multiple affected individuals. Family members are
phenotyped and then genotyped with a set of genetic markers. These markers
are usually short tandem repeats (STRs) that are located at about 10-cM
intervals through the genome. STRs are chosen because they have significant
variability but no functionality. In a commonly used linkage method, STR
marker alleles that display a higher frequency of sharing (than expected by
chance) in affected family members are presumed to be linked to a nearby
susceptibility gene for the disorder under study. The linkage signal is measured
as an LOD score [Iog10 of the likelihood ratio that two loci are linked versus
nonlinked (i.e. independent assortment)]. Once linkage is found [LOD
score >3.6, the presently accepted threshold for linkage in a complex
trait (14)], additional markers are used to narrow the region of interest and
positional cloning is used to identify genes for further study.
In each of these molecular genetic approaches, proband and family
assessments for phenotypical classification must be rigorous and controlled,
using instruments with good reliability and validity. Confidentiality must be
scrupulously guarded, both between family members and in general, with
adequate precautions taken to blind data and ensure protection of subjects’
privacy. The importance of control group selection is discussed above. Also
central to study integrity is sample size. Current hypotheses suggest that as
complex traits many psychiatric disorders are caused by several to many
susceptibility genes, each of small to modest effect, that likely interact to render
susceptibility to the disorder. Since this genetic susceptibility is parsed over
several to many genetic loci, large study populations are needed to generate
sufficient power to detect these genes (15,16). In uncommon disorders such as
eating disorders, large sample sizes are very difficult to achieve such that
multiple collaborative sites are necessary to generate a sufficient patient
population for genetic studies.
CANDIDATE GENE STUDIES IN EATING

DISORDERS
In order to identify candidate genes for a genetic study, a significant amount
must be known about the underlying biology of the disorder of interest. The
serotonergic, dopaminergic, leptinergic, noradrenergic, and opioid systems are
implicated in eating behaviors, appetite, weight control, mood, obsessionality,
impulse regulation, and energy regulation—all features of the clinical
presentation of eating disorders. Alterations in these biological systems are
found during an acute phase of illness or in the midst of treatment or recovery.
While causal links can be inferred from these findings (and, in many cases,
supported by studies of animal models), it should be done with the caveat that
these are not prospective clinical studies. The following section will summarize
candidate gene studies (organized by gene class when appropriate) in eating
disorders and offer some perspectives on the often-mixed findings that have
emerged.
SEROTONINERGIC GENES IN EATING DISORDERS

Serotonin (5-HT) produces its effects by acting at a number of membranebound
serotonin receptors, which, for the most part, belong to the G-proteincoupled
receptor superfamily (for a review, see 17). The major regulators of the
serotonergic system in the brain are tryptophan hydroxylase, the 5-HT
328 GRICE
transporter, and 5-HT receptors. The 5-HT receptor family is composed of at

least 14 distinct types and is one of the most complex families of
neurotransmitter receptors. In addition to being found in the central nervous
system, 5-HT receptors are also located in the peripheral nervous system as
well as other tissues, including the cardiovascular system, blood, and gut.
Serotonin dysregulation has been implicated in many psychiatric disorders,
including eating disorders, major depression, schizophrenia, obsessive-
compulsive disorder, as well as other anxiety disorders such as panic disorder
and social phobia. Beyond psychiatric disorders, serotonin is also thought to play
a role in migraine headaches, hypertension, irritable bowel syndrome, and
vomiting. Given the fundamental role of 5-HT in a multitude of biological and
behavioral functions related to eating behaviors, considerable interest is focused
on the role serotonergic system candidate genes may play in eating disorders.
(Table 1 summarizes findings related to the 5-HT2A receptor nucleotide–1438
variations; Table 2 summarizes all other association studies of serotonergic
genes in eating disorders.)
THE 5-HT2 RECEPTORS
5-HT2A Subclass
The 5-HT2 receptor class is composed of the 5-HT2A, 5-HT2B, and 5-HT2C
receptors. These receptors demonstrate 46–50% sequence homology and have
been mapped to chromosome 13ql4-q21, 2q36.3–37.1, and Xq24,
respectively. The 5-HT2A receptor is expressed in many peripheral and central
tissues. In the central nervous system, the cortex, basal ganglia, and claustrum
are primary sights for 5-HT2A expression. In genetic studies of eating disorders,
the 5-HT2A receptor is the most extensively studied serotonin-related gene.
Several polymorphisms have been reported in the 5-HT2A receptor gene. The
variant that has received the most attention is a polymorphism in the promoter
(regulatory) region of the gene in which there is an A or G at nucleotide–1438.
A series of association studies (both population based using obese,
underweight, normal-weight, and non-psychiatrically ill controls, and family
based using the TDT) have examined the serotonin receptor 5-HT2A promoter
polymorphism—1438A/G. Five studies have found a positive association
between the 5-HT2A -1438A allele and eating disorders (18–22). The two
studies that focused solely on AN phenotypes found evidence for a positive
association between the—1438A allele and AN (all subtypes) (18,19) and/or
the AN restricting subtype (19). Three studies examined a broader range of
asdsadasd
TABLE 1 Association Studies (Population-Based and Family-Based) of Eating Disorders and the 5-HT2A Promoter Polymorphism -1438A/G
330 GRICE
a TDT method applied to a sample subset.

b In the AN group, restricting and binge-purge subtypes were studied.
c RAN subtype significantly associated with–1438A allele, or the–1438 AA genotype.
d Significant association (p < 0.01) was found between–1438G allele and BN.
*,** Denotes comparison groups that yielded a p value ≤ 0.05. NS, Nonsignificant; p > 0.05.
TABLE 2 Association Studies of Eating Disorders and Serotonin-Related Genes3
aStudies of the 5-HT2A -1438 A/G promoter polymorphism are summarized in Table 1.
*,**,*#,## Denotes comparison group that yielded a p value ≤ 0.05 and associated authorship. NS, Nonsignificant; p > 0.05.
332 GRICE
eating disorder phenotypes (20–22). One study demonstrated a significant

association of the -1438A allele with AN but not BN (20). One study found an
association between the -1438A allele and the AN restricting subtype but not
AN binge-purge subtype or BN (21). And the third study found a positive
association between the -1438A allele in AN, the AN restricting subtype, and
the BN purging subtype but not in binge eating disorder (22). In contrast to the
above positive findings, five additional studies have examined the 5-HT2A—
1438A/G polymorphism and found no significant associations with AN and/or
BN using a variety of population-based controls (23–26), the family-based
TDT (27), as well as subphenotypes (the AN restricting subtype, age of illness
onset, minimal lifetime BMI; 25,27). Interestingly, in a study of Japanese patients,
the alternative G allele at nucleotide -1438 was associated with BN (but not
AN) (28); this study further suggests an association between binge eating
and/or purging behavior and comorbid borderline personality disorder with
increased frequency of the G allele, although this type of analysis has not been
replicated.
Finally, four other polymorphisms in the 5-HT2A receptor gene have been
examined in eating disorders. Two variants are in the coding region of the gene
(Thr25Asn and His452Tyr) and two are located in the noncoding region of the
gene (102 T/C and 516 C/T). A range of comparison groups were used in
these analyses (including the TDT on a subset of individuals); however, none of
these studies produced a significant finding in AN or BN (21,23).
At this stage, it is difficult to draw clear interpretations from the studies of
5-HT2A polymorphisms and eating disorders. All in all, evidence is accumulating
that the AN restricting subtype may be genetically distinct from other AN
subtypes or BN (19–22). This theory is supported by clinical studies (29,30) as
well as recent genetic linkage study of AN (31). Other issues that may cloud our
understanding of the role of the 5-HT2A gene in eating disorders have been
discussed above. For example, in molecular genetic studies, it is important to
use large enough sample sizes (and control groups) to garner sufficient power to
detect genes of small to modest effect, and the effect of a particular candidate
gene is also an estimated value. We do not definitively know if all of these
studies were sufficiently powerful to detect association (or lack thereof) with
confidence, although several studies did have robust sample sizes.
In part to address some of these methodological issues, a meta-analysis to
examine association between the 5-HT2A -1438 polymorphism and AN was
done and found no significant relationship between -1438A/G polymorphism
and AN (25). Genetic heterogeneity and ethnic admixture leading to population
stratification could have obscured positive findings from this metaanalysis.
However, using the family-based TDT to control population stratification bias,
a multicenter collaborative study identified a large number of AN subjects
(n>300) but found neither an association between AN and 5HT2A -1438A allele
nor evidence of an interaction of this allele with age of onset of AN or minimum
lifetime BMI (27). Although the sample size in this study is the largest to date,
we still do not know for sure if it generates sufficient power when the odds ratio
associated with the A allele is small. Of note, however, is a follow-up study
based in part on this sample that used a quantitative TDT approach to suggest
that the A allele may act as a modifying factor, delaying age of AN onset (32).
Thus, the definitive role of the 5-HT2A polymorphisms in eating disorders
remains unclear but of great interest. There are some very interesting findings
and leads to follow, particularly related to the AN restricting subtype and the
5-HT2A -1438A allele. We must await verification of these results, likely
through large-scale controlled studies, before firm conclusions can be drawn
about the ultimate relevance of the -5-HT2A gene to the genetics of eating
disorders.
5-HT2C Subclass
Similar to the 5-HT2A receptor, the 5-HT2C receptor has been implicated
in the regulation of feeding-related behaviors such as satiety and eating
continuity (33). Supportive evidence also comes from mouse models in which
transgenic mice lacking the 5-HT2C receptor demonstrate abnormal feeding
behavior (34). A small number of studies have examined allele frequencies of
the 5-HT2C Cys23Ser polymorphism in subjects with a range of eating disorder
phenotypes. Although no association was found between Cys23Ser variants and
DSM-IV diagnoses of AN (neither restricting nor purging subtypes) (21),
BN (21,35), or BED (35), there was a significant association between the Ser23
allele and a clinically defined group of underweight teenage girls (some of whom
met criteria for AN) (36). If confirmed these data suggest that 5-HT2C may be
involved in the regulation of food intake or proneness to weight loss rather than
an eating disorder diagnosis per se.
5-HT, Subclass
The 5-HTi receptor class comprises five receptor subtypes (5-HT1A, 5-HT1B,
5-HT1D, 5-HT1E and 5-HT1F), which share 40–63% sequence homology in
humans (17). The 5-HT1Dβ, receptor, also known as 5-HT1B, has been mapped
to chromosome 6ql3. 5-HT1Dβ is expressed in the central nervous
system, particularly in the basal ganglia, striatum, and frontal cortex, and is
thought to function as an autoreceptor as well as a heteroreceptor that effects
release of acetylcholine, glutamate, dopamine, noradrenaline, and
γ–aminobutyric acid (for review, see 37). Population-based studies of the
334 GRICE
5-HT1Dβ receptor gene have examined its possible role in both AN and BN. In a
study of AN individuals, the Phel24Cys variant was studied using obese and
underweight individuals as comparison but found no significant association (38).
A study of the 5-HT1BG681C polymorphism examined the distribution of
genotypes in women with BN (39). When the minimum and maximum lifetime
BMIs were compared across the three genotypic groups (G/G, G/C, and CC)
both the G/C and C/C genotypes were associated with lower minimum
lifetime BMIs (and unrelated to lifetime AN rates), perhaps suggesting a role for
5-HT1B in a subphenotype not defined by DSM-IV schemas.
5-HT7 Subclass
The 5-HT7 receptor is located on chromosome 10q23.3–24.4 and shows less
than 50% homology to other receptors in the 5-HT family. 5-HT7 receptors are
found in the limbic and thalamocortical brain regions, regions implicated in
central regulation of feeding behavior (40). These localizations, together with
evidence that atypical antipsychotics have a high affinity for 5-HT7 (41), make it
an attractive candidate gene for psychiatric disorders. However, the sole study
of the 5-HT7 receptor in eating disorders examined the Pro279Leu
polymorphism in AN subjects and found no significant association between
allele frequencies in AN compared to obese and underweight controls (38).
OTHER SEROTONERGIC-RELATED GENES
Serotonin Transporter
The serotonin transporter is an integral component in regulating serotonin
function in the central nervous system and is the site of action for the
serotonin reuptake inhibitors. A 44-base-pair insertion/deletion (5-HTTLPR)
polymorphism in the promoter region of the 5-HT transporter gene (5-HTT)
confers differential transcription rates on gene expression. The so-called short
variant of 5-HTTLPR (the deletion variant) is associated with significant
decreases in 5-HTT expression and 5-HT uptake (42), making 5-HTT a
candidate gene of great interest for many psychiatric disorders, including eating
disorders. Several groups have examined this variant, using populationbased
methods and a variety of control groups, and found conflicting results regarding
the role of 5-HTT in a range of eating disorders phenotypes. One positive
association was found between the 5-HTTLPR deletion (short) variant in AN
compared to obese and normal-weight controls (43). Three studies did not
support this finding (44–46), although one group found a positive association
between this same variant and BN (46). Also of note is a positive association
between the 5-HTTLPR insertion (long) variant and abnormal eating
behaviors (defined by high scores on the Eating Attitudes Test) in nonclinical
subjects (47). Since these studies have not been replicated, as with other
candidate genes, we will have to await further research on 5HTTLPR to show
us how, if at all, this gene influences susceptibility to abnormal eating behaviors
or clinically defined eating disorder diagnoses.
Tryptophan Hydroxylase
Since tryptophan hydroxylase (TPH) is the rate-limiting enzyme in the serotonin
synthesis pathway it has been seen as a potential candidate gene in eating
disorders. TPH is mapped to chromosome 11p15.3-p14. In the one published
study on TPH and AN, there was no evidence for association (48).
CANDIDATE GENES BEYOND SEROTONIN

Beyond the serotonergic hypotheses underlying the biology and heritability of
AN, other genes from a variety of biological classes have been examined (49–63)
primarily through the case-control association method. Due to methodological
issues (e.g., sample sizes, risk of population stratification) these studies may
have limited power to detect differences between affected individuals and
controls. In addition, it is important to recall that although the genes chosen do
have theoretical or functional roles in eating behavior, weight regulation, or
related phenomenon, these studies are exploratory and the actual role of these
genes in eating disorder phenotypes is, for the most part, not completely
understood. With these caveats, however, there are several positive findings
associating eating disorder phenotypes with specific genes, e.g., estrogen
receptor β (52,61), uncoupling protein 2,3 (UCP-2/UCP-3) (56), catechol
O-methyltransferase (58), agouti-related protein (59), small-conductance
calcium-activated potassium channel 3 (hSKCa3) (60), norepinephrine
transporter gene (62), and melanocortin-4 receptor gene (MC-4r) (63), which
is reviewed briefly below and summarized in Table 3.
Estrogen Receptors
There are two types of estrogen receptors: ER-α and ER-β mapped to
chromosomes 6q25.1 and 14q23.2, respectively. ERs belong to the nuclear
hormone receptor family and are found in brain areas involved in regulation of
sdgfsdfsdf
TABLE 3 Association Studies of Eating Disorders and Candidate Genes Displayed in Chronological Order3
336 GRICE
a Studies of the 5-HT A-1438 A/G promoter polymorphism and other serotonin-related genes are summarized in Table 1 and Table 2,
2
respectively.
b RFLP, restriction fragment length polymorphism.
c LEGLUR, leptin gene-linked upstream region.
d VNTR, variable number of tandem repeats.
e Microsatellite markers that flank the gene.
f hSKCa3, small-conductance calcium-activated potassium channel 3.
NS, Nonsignificant; p>0.05.

338 GRICE
food intake (for review, see 64). Since high estrogen levels are known to inhibit
food intake (65) and in the brain, ERs colocalize with corticotropinreleasing
factor (CRF), it is theorized that interactions between estrogen and CRF may
modulate the hypothalamic-pituitary-adrenal axis and be involved in the
molecular basis of feeding, satiety, and, perhaps, eating disorders, especially
AN (66). Two studies have looked specifically at the association between ERs
and eating disorders (52,61). ER-(3 was screened for variations in subjects
with a range of weight extremes (extremely obese, AN, BN, healthy
underweight) (52). Although five different sequence variants were identified,
the association results were negative except for a suggestive association between
a G/A polymorphism at nucleotide 1082 and the AN phenotype. A second
study examined ER-α and ER-β in AN subjects (61). There were no significant
findings related to ER-α, but in ER-β the heterozygous genotype GA at
nucleotide 1082 was associated with AN.
Uncoupling Protein-2/Uncoupling Protein-3

Uncoupling protein-2 and uncoupling protein-3 (UCP-2/UCP-3) are involved
in resting metabolic rate and energy expenditure (67,68). The UCP-2/UCP-3
locus is on chromosome 11q13; UCP-2 has been linked to obesity and
hyperinsulinemia, making it an interesting candidate gene for other abnormal
weight disorders (69). In a study of AN subjects and markers located in the
UCP-2/UCP-3 locus a significant association between a marker (D11S911) and
AN was found (56). Further work in this area will elucidate if UCPs are
involved in the genetic basis of eating disorders beyond obesity.
Catechol-O-Methyltransferase
Catechol-O-methyltransferase (COMT) catalyzes the transfer of a methyl group
to catecholamines (dopamine, epinephrine, and norepinephrine) and is the major
degradative pathway of these neurotransmitters. COMT has been mapped to
chromosome 22q 11.21. In humans there is a functional polymorphism
(Val/Met 158) that determines high and low enzymatic activity. COMT has
been considered a candidate gene for several psychiatric disorders and associated
with positive findings in a subset of these studies (e.g., substance use disorders,
schizophrenia, attention deficit disorder). Relevant to this summary, the Val/
Met polymorphism was studied in a family-based association study and the high-
activity allele was significantly associated with AN, indicating a possible role in
eating disorders (58).
Melanocortinerglc System Genes

The central melanocortinergic system is involved in regulation of food intake,
body weight, and eating behavior and is a major pathway in leptin signaling (for
review, see 70). There are at least two melanocortin receptors; melanocortin-3
receptor (MC3-r) and melanocortin-4 receptor (MC4-r). Hormones in this
system are the melanocortins, products of the proopiomelanocortin gene
(POMC), that act as MC receptor agonists, such as ACTH, β endorphin, and
MSH, and the orexigenic neuropeptide agouti-related protein (AGRP) that acts
as an MC receptor antagonist. In this system, leptin stimulates POMC
expression and suppresses AGRP expression in brain regions involved in feeding
and satiety. Leptin also regulates other important hormones involved in food
intake, neuroendocrine balance, and energy expenditure, e.g., neuropeptide Y
(NPY, another orexigenic neuropeptide), thyrotropin-releasing hormone, and
corticotropin-realeasing hormone. Many components of the central
melanocortinergic system are candidate genes for disordered weight and eating
phenotypes. MC-4r (mapped to chromosome 18q22), POMC (mapped to
chromosome 2p23.3), AGRP (mapped to chromosome 16q22), NPY Yl
(mapped to chromosome 4q31.3-q32), NPY Y5 (mapped to chromosome
4q31-q32), and the leptin gene (mapped to chromosome 7q32.1) have been
studied in eating disorder phenotypes in population-based and family-based
association studies. From these studies, one positive association was found
between AGRP and AN (59), and one subject with BN (out of 81 screened) had
a haplo insufficiency mutation in MC4-r (63).
OTHER CANDIDATE GENES
hSKCa3
hSKCa3 is a small-conductance calcium-activated potassium channel involved in
regulating neural excitability. hSKCa3 has been mapped to chromosome 1q21.
In one model, small-conductance calcium-activated potassium channels
may contribute to the genetic susceptibility to bipolar disorder and
schizophrenia (71,72). To date, one study has examined hSKCa3 polymorphisms
in AN (60). Using a family-based association method, a positive association was
found, with longer alleles from the hSKCa3 gene found at higher rates in the AN
sample compared to controls.
340 GRICE
Norepinephrine Transporter
Norepinephrine transporter (NET) expression is regulated by norepinephrine;
thus, it has an important role in noradrenergic transmission. This system
regulates many neuroendocrine systems implicated in the pathophysiology of
eating disorders as well as anxiety disorders. In clinical research, reduced
noradrenergic activity is found in AN patients following normal-weight
restoration (73,74) although the mechanism underlying these findings are
unknown. Using a family-based association method, a polymorphic region in the
promoter of the NET gene was studied in the restricting subtype of AN (62). A
significant association was found between an insertion in this region and the
restricting AN subtype, suggesting that this NET gene variant, or a variant in
linkage disequilibrium with it, increases risk for restricting AN.
LINKAGE ANALYSIS IN EATING DISORDERS

Three genetic linkage studies of eating disorders are in the litera-
ture (31,75,76). The two linkage studies of AN are based on the same sample of
about 200 affected families. These families contained at least two family
members affected with an eating disorder, all ascertained through a proband
with AN. Both studies used an allele-sharing linkage strategy to identify genetic
loci that contribute to AN. In the first linkage study of AN, the linkage analysis
of the entire sample did not yield any statistically significant results (31).
However, when the phenotype was narrowed to include only those families
ascertained through a proband with the restricting subtype of AN, a significant
linkage signal of 3.45 was found on chromosome 1p. Several interesting
candidate genes related to eating phenotypes are found in this region, including
an orexin receptor gene (Hcrtrl), the delta opioid receptor, and the 5-HT1D
receptor. Further studies will determine if there are specific associations
between these candidate genes, or other genes in this region that have not yet
been identified, and restricting AN. The second linkage study of AN used two
clinically based covariates, drive for thinness and obsessionality, which were
shown to discriminate subsets in the clinical sample (75). When these covariates
were incorporated into the linkage analysis, the genome region with greatest
statistical significance showed linkage on chromosome Iq (LOD = 3.46), a
different region from that seen in the first AN linkage study, and two weaker
signals on chromosome 2 and 13.
The BN linkage study examined about 300 families affected with eating
disorders, all ascertained through a proband with the purging subtype of BN (76).
Purging methods must have at least included regular vomiting, and a lifetime or
current history of AN was acceptable. Using this phenotype, the highest linkage
signal of 2.92 was found on chromosome 10p. A second peak in this same
region of 10p approached significance and a smaller suggestive linkage signal
was found on chromosome 14. The sample was then narrowed by selecting only
those families in which at last one other affected relative had regular vomiting
behavior (by design, the proband had already met this criteria). When this subset
of 133 families was studied, the initial linkage signal at chromosome 10p
increased to 3.39. This chromosomal region has been implicated in other
disorders, such as obesity, alcoholism, schizophrenia, and bipolar disorder,
perhaps reflecting that this region harbors a gene(s) that generically increases
susceptibility to a range of psychiatric and related disorders (as summarized
in (76)). It is also possible that a gene(s) in this region may specifically alter
susceptibility to BN and obesity, two disorders that may have overlapping
vulnerabilities and/or manifestations.
OBESITY
Compared to the genetic studies of DSM-IV-defined eating disorders, there are
a plethora of studies of human obesity. This intensity of research has identified
putative regions affecting obesity phenotypes on all chromosomes except
chromosome Y and with over 250 significant regions of interest in the genome
under study, a summary of which is beyond the scope of this chapter (for review,
see 77). Well over 150 studies covering approximately 60 candidate genes have
reported significant associations with obesity. Genome-wide linkage analyses of
obesity-related phenotypes have uncovered similar numbers of loci linked to
obesity indicators. As in eating disorders, a variety of phenotypes related to
obesity have been defined, including BMI, body fat mass, percentage of body
fat, fat-free mass, skinfolds, resting metabolic rates, and plasma leptin levels—
again indicating the importance of phenotypical definition beyond absolute
clinical categories. Relevant to this review of the genetics of eating disorders,
many of the genes discussed in this chapter also have been examined in obesity.
Significant findings in obesity genetics have been associated with serotonergic
genes (5-HTlB, 5-HT2C, 5HT2A), melanocortinergic genes (MC4-r, POMC,
NPY, AGRP), ER-α, UCP-2/UCP-3, and tumor necrosis factor-α. As research
continues to delve into the genetic basis of eating disorders we should expect to
see greater understanding of which pathways (e.g., those related to feeding
behavior, satiety, and/or energy metabolism) are affected in both obesity
disorders and eating disorders. Delineation of the specific genetic variations
associated with relevant phenotypes will allow us to discover if abnormal eating
behaviors can be seen as a spectrum of disorders ranging from obesity to
anorexia or if the genetic bases for these disorders are in fact distinct.
342 GRICE
SUMMARY
Although research in the genetics of obesity syndromes far outdistances research
in the genetics of eating disorders at this time, our knowledge and research in
AN and BN far exceeds that for other eating disorder syndromes or diagnoses.
It is apparent that AN and BN are heritable eating disorders that result from
complex interactions of genetic liabilities and environmental factors. The
positive case-control studies associated with several genes of interest (e.g., the
5-HT2A receptor, melanocortinergic system genes) have piqued interest in the
molecular genetic basis of eating disorders. Recent findings of significant linkage
peaks on chromosome 1p and 1q (related to an AN subtype) and on chromosome
10p (related to a BN subtype) have spurred the field forward with new
enthusiasm. Thus, many promising leads are emerging from genetic linkage
studies and candidate gene analyses; however, we must await further molecular
studies to confirm the specific genes that confer heritability to eating disorders.
Impediments to progress are due in part to the low population prevalence of eating
disorders and perhaps also the low rates of clinical identification of eating
disorder phenotypes, both of which make it very difficult to ascertain samples
sufficiently powerful to detect genes of less than major effect. In addition, it is
only in the past few years that molecular geneticists have had the means to examine
the human genome with the level of specificity and detail required for human
genetic studies. As a result of these technological and methodological advances,
the field of molecular genetics in eating disorders is growing rapidly. The large-
scale linkage studies of AN and BN and the larger TDT and association studies
are evidence of the more methodologically robust and controlled analyses that
have recently been undertaken. Collaborations such as these, combined with
ongoing developments in molecular genetic and genomic analytical tools, will
further strengthen genetically based studies and bring us closer still to the
identification of specific genes that confer susceptibility to eating disorders.
REFERENCES
Disorders. 4th ed. Washington, DC, 1994:544–545.
3. Hoek HW, van Harten PN, van Hoeken D, Susser E. Lack of relation between
culture and anorexia nervosa-results of an incidence study on Curacao. N Engl J
Med 1998;338(17):1231–1232.
4. Timmerman MG, Wells LA, Chen SP. Bulimia nervosa and associated alcohol
abuse among secondary school students. J Am Acad Child Adol Psychiatry 1990; 29
(1):118–122.
5. Kendler KS, MacLean C, Neale M, Kessler R, Heath A, Eaves L. The genetic

epidemiology of bulimia nervosa. Am J Psychiatry 1992; 48:1627–1637.
6. Sullivan PF. Mortality in anorexia nervosa. Am J Psychiatry 1995; 152:1073–1074.
7. Hsu LKG, Crisp AH, Harding B. Outcome of anorexia nervosa. Lancet 1979; 1
(8107):61–65.
8. Caspar RC, Eckert ED, Halmi KA, et al. Bulimia: its incidence in patients with
anorexia nervosa. Arch Gen Psychiatry 1980; 37:1030–1035.
9. Strober M, Salkin B, Burroughs J, Morrell W. Validity of the bulimia-restricter
distinction in anorexia nervosa. J Nerv Ment Dis 1982; 170:345–351.
10. Gull WW. Anorexia nervosa (apepsia hysterica, anorexia hysterica). Trans Clin Soc
London 1874; 7:22–28.
11. LasègueE-C. On hysterical anorexia. Med Times Gazette, 1873:265–266:367–369.
12. Gerson ES, Cloninger CR. Genetic approaches to mental disorders. Washington,
DC: American Psychiatric Press, 1994.
13. Spielman RS, McGinnis RE, Ewens WJ. Transmission test for linkage
disequilibrium: the insulin gene region and insulin-dependent diabetes mellitus
(IDDM). Am J Hum Genet 1993; 52:506–516.
14. Lander E, Kruglyak L. Genetic dissection of complex traits: guidelines for
interpreting and reporting linkage results. Nat Genet 1995; 11(3):241–247.
15. Risch N, Merikangas K. The future of genetic studies of complex human diseases.
Science 1996; 273:1516–1517.
16. Risch N. Searching for genetic determinants in the new millennium. Nature 2000;
405:847–856.
17. Hoyer D, Hannon JP, Martin GR. Molecular, pharmacological and functional
diversity of 5-HT receptors. Pharmacol Biochem Behav 2002; 71(4):533–554.
18. Collier D A, Arranz M J, Li T Mupita D, Brown N, Treasure J. Association
between 5-HT2A gene promoter polymorphism and anorexia nervosa. Lancet
1997; 350: 412.
19. Sorbi S, Nacmias B, Tedde A, Ricca V, Mezzani B, Rotella CM. 5-HT2A promoter
polymorphism in anorexia nervosa. Lancet 1998; 351:1785.
5-HT2A promoter polymorphism -1438G/A, anorexia nervosa, and obsessive
compulsive disorder. Lancet 1998; 351:1785–1786.
gene polymorphisms in anorexia nervosa and bulimia nervosa. Neurosci Lett.
1999; 277(2): 134–136.
22. Ricca V, Nacmias B, Cellini E, Di Bernardo M, Rotella CM, Sorbi S. 5-HT2A
receptor gene polymorphism and eating disorders. Neurosci Lett 2002; 323(2):
105–108.
23. Hinney A, Ziegler A, Nother MM, Remschmidt H, Hebebrand J. 5-HT2A receptor
gene polymorphisms, anorexia, and obesity. Lancet 1997; 350:1324–1325.
24. Campbell DA, Sundaramurthy D, Markham AF, Pieri LF. Lack of association
between 5-HT2A gene promoter polymorphism and susceptibility to anorexia
nervosa. Lancet 1998; 351:499.
344 GRICE
25. Ziegler A, Hebebrand J, Gorg T, Rosenkranz K, Fichter M, Herpertz-Dahlmann B,

Remschmidt H, Hinney A. Further lack of association between the 5-HT2A gene
promoter polymorphism and susceptibility to eating disorders and a metaanalysis
pertaining to anorexia nervosa. Mol Psychiatry 1999; 4(5):410–412.
26. Ando T, Komaki G, Karibe M, Kawamura N, Hara S, Takii M, Naruo T,
Kurokawa N, Takei M, Tatsuta N, Ohba M, Nozoe S, Kubo C, Ishikawa T. 5HT2A
promoter polymorphism is not associated with anorexia nervosa in Japanese
patients. Psychiatr Genet 2001; 11 (3): 157–160.
27. Gorwood P, Ades J, Bellodi L, Cellini E, Collier DA, Di Bella D, Di Bernardo M,
Estivill X, Fernandez-Aranda F, Gratacos M, Hebebrand J, Hinney A, Hu
X, Karwautz A, Kipman A, Mouren-Simeoni MC, Nacmias B, Ribases M,
Remschmidt H, Ricca V, Rotella CM, Sorbi S, Treasure J. The 5-HT(2A)—
1438G/A polymorphism in anorexia nervosa: a combined analysis of 316 trios from
six European centres. Mol Psychiatry 2002; 7(l):90–94.
28. Nishiguchi N, Matsushita S, Suzuki K, Murayama M, Shirakawa O, Higuchi S.
eating disorders in Japanese patients. Biol Psychiatry 2001; 50(2): 123–128.
metabolism between nonbulimic and bulimic patients with anorexia nervosa. Am J
Psychiatry 1984; 141(12):1598–1601.
30. Herzog DB, Field AE, Keller MB, West JC, Robbins WM, Staley J, Colditz GA.
Subtyping eating disorders: is it justified? J Am Acad Child Adol Psychiatry 1996;
35(7):928–936.
31. Grice DE, Halmi KA, Fichter MM, Strober M, Woodside DB, Treasure JT, Kaplan
AS, Magistretti PJ, Goldman D, Bulik CM, Kaye WH, Berrettini WH. Evidence
for a susceptibility gene for anorexia nervosa on chromosome 1. Am J Hum Genet
2002; 70(3):787–792.
32. Kipman A, Bruins-Slot L, Boni C, Hanoun N, Ades J, Blot P, Hamon M, Mouren-
Simeoni M, Gorwood P. 5-HT(2A) gene promoter polymorphism as a modifying
rather than a vulnerability factor in anorexia nervosa. Eur Psychiatry 2002; 17(4):
227–279.
33. Simansky KJ. Serotonergic control of the organization of feeding and satiety. Behav
Brain Res 1996; 73(l-2):37–42.
34. Tecott LH, Sun LM, Akana SF, Strack AM, Lowenstein DH, Dallman MF, Julius D.
Eating disorder and epilepsy in mice lacking 5-HT2c serotonin receptors. Nature
1995; 374(6522):542–546.
35. Burnet PW, Smith KA, Cowen PJ, Fairburn CG, Harrison PJ. Allelic variation of
the 5-HT2c receptor (HTR2C) in bulimia nervosa and binge eating disorder.
Psychiatr Genet 1999; 9(2):101–104.
36. Westberg L, Bah J, Rastam M, Gillberg C, Wentz E, Melke J, Hellstrand M,
Eriksson E. Association between a polymorphism of the 5-HT2c receptor and
weight loss in teenage girls. Neuropsychopharmacology 2002; 26(6):789–793.
37. Pauwels PJ. 5-HT IB/D receptor antagonists. Gen Pharmacol 1997; 29(3):
293–303.
38. Hinney A, Herrmann H, Lohr T, Rosenkranz K, Ziegler A, Lehmkuhl G, Poustka

F, Schmidt MH, Mayer H, Siegfried W, Remschmidt H, Hebebrand J. No evidence
for an involvement of alleles of polymorphisms in the serotonin1Dbeta and 7 receptor
genes in obesity, underweight or anorexia nervosa. Int J Obes Relat Metab Disord
1999; 23(7):760–763.
39. Levitan RD, Kaplan AS, Masellis M, Basile VS, Walker ML, Lipson N, Siegel GI,
Woodside DB, Macciardi FM, Kennedy SH, Kennedy JL. Polymorphism of the
serotonin 5-HT1β receptor gene (HTRIB) associated with minimum lifetime body
mass index in women with bulimia nervosa. Biol Psychiatry 2001; 50(8): 640–643.
40. Bernardis LL, Bellinger LL. The lateral hypothalamic area revisited: neuro anatomy,
body weight regulation, neuroendocrinology and metabolism. Neurosci Biobehav
Rev 1993; 17(2): 141–193.
41. Roth BL, Craigo SC, Choudhary MS, Uluer A, Monsma FJ Jr, Shen Y, Meltzer
HY, Sibley DR. Binding of typical and atypical antipsychotic agents to
5hydroxytryptamine-6 and 5-hydroxytryptamine-7 receptors. J Pharmacol Exp
Ther 1994; 268(3):1403–1410.
42. Lesch KP, Bengel D, Heils A, Sabol SZ, Greenberg BD, Petri S, Benjamin J,
Muller CR, Hamer DH, Murphy DL. Association of anxiety-related traits with a
polymorphism in the serotonin transporter gene regulatory region. Science 1996;
274(5292): 1527–1531.
43. Fumeron F, Betoulle D, Aubert R, Herbeth B, Siest G, Rigaud D. Association of a
functional 5-HT transporter gene polymorphism with anorexia nervosa and food
intake. Mol Psychiatry 2001; 6(1):9–10.
44. Hinney A, Barth N, Ziegler A, von Prittwitz S, Hamann A, Hennighausen K, Pirke
KM, Heils A, Rosenkranz K, Roth H, Coners H, Mayer H, Herzog W, Siegfried
A, Lehmkuhl G, Poustka F, Schmidt MH, Schafer H, Grzeschik KH, Lesch KP,
Lentes KU, Remschmidt H, Hebebrand J. Serotonin transporter gene-linked
polymorphic region: allele distributions in relationship to body weight and in
anorexia nervosa. Life Sci 1997; 61:PL 295–303.
45. Sundaramurthy D, Pieri LF, Gape H, Markham AF, Campbell DA. Analysis of the
serotonin transporter gene linked polymorphism (5-HTTLPR) in anorexia nervosa.
Am J Med Genet 2000; 96(l):53–55.
46. Di Bella DD, Catalano M, Cavallini MC, Riboldi C, Bellodi L. Serotonin
transporter linked polymorphic region in anorexia nervosa and bulimia nervosa.
Mol Psychiatry 2000; 5(3):233–234.
47. Matsushita S, Nakamura T, Nishiguchi N, Higuchi S. Association of serotonin
transporter regulatory region polymorphism and abnormal eating behaviors. Mol
Psychiatry 2002; 7:538–540.
48. Han L, Nielsen DA, Rosenthal NE, Jefferson K, Kaye W, Murphy D, Altemus M,
Humphries J, Cassano G, Rotondo A, Virkkunen M, Linnoila M, Goldman D. No
coding variant of the tryptophan hydroxylase gene detected in seasonal affective
disorder, obsessive-compulsive disorder, anorexia nervosa, and alcoholism. Biol
Psychiatry 1999; 45(5):615–619.
346 GRICE
49. Hinney A, Lentes K-U, Rosenkranz K, Barth N, Roth H, Ziegler A, Hennighausen

K, Coners H, Wurmser H, Jacob K, Romer G, Winnikes U, Mayer H, Herzog W,
Lehmkuhl G, Poustka F, Schmidt MH, Blum WF, Pirke KM, Schafer H, Grzeschik
KH, Remschmidt H, Hebebrand J. Beta3 adrenergic receptor allele distributions in
children, adolescents, and young adults with obesity, underweight, and anorexia
nervosa. Int J Obesity 1997; 21:224–230.
50. Hinney A, Becker I, Heibult O, Nottebom K, Schmidt A, Ziegler A, Mayer H,
Siegfried W, Blum WF, Remschmidt H, Hebebrand J. Systematic mutation
screening of the pro-opiomelanocortin gene: identification of several genetic
variants including three different insertions, one nonsense and two missense point
mutations in probands of different weight extremes. J Clin Endocrinol Metab
1998; 83(10):3737–3741.
51. Bruins-Slot L, Gorwood P, Bouvard M, Blot P, Ades J, Feingold J, Schwartz JC,
Mouren-Simeoni MC. Lack of association between anorexia nervosa and D3
dopamine receptor gene. Biol Psychiatry 1998; 43:76–78.
52. Rosenkranz K, Hinney A, Ziegler A, Hermann H, Fichter M, Mayer H, Siegfried
W, Young JK, Remschmidt H, Hebebrand J. Systematic mutation screening of the
estrogen receptor beta gene in probands of different weight extremes:
identification of several genetic variants. J Clin Endocrinol Metab 1998; 83(12):
4524–4527.
53. Rosenkranz K, Hinney A, Ziegler A, von Prittwitz S, Barth N, Roth H, Mayer H,
Siegfried W, Lehmkuhl G, Poustka F, Schmidt M, Schafer H, Remschmidt H,
Hebebrand J. Screening for mutations in the neuropeptide Y Y5 receptor gene in
cohorts belonging to different weight extremes. Int J Obesity Relat Metab Disord
1998; 22(2): 157–163.
54. Hinney A, Bornscheuer A, Depenbusch M, Mierke B, Tolle A, Middeke K, Ziegler
A, Roth H, Gerber G, Zamzow K, Ballauff A, Hamann A, Mayer H, Siegfried W,
Lehmkuhl G, Poustka F, Schmidt MH, Hermann H, HerpertzDahlmann BM,
Fichter M, Remschmidt H, Hebebrand J. No evidence for involvement of the
leptin gene in anorexia nervosa, bulimia nervosa, underweight or early onset
extreme obesity: identification of two novel mutations in the coding sequence and
a novel polymorphism in the leptin gene linked upstream region. Mol Psychiatry
1998; 3(6):539–543.
55. Hinney A, Schneider J, Ziegler A, Lehmkuhl G, Poustka F, Schmidt MH, Mayer
H, Siegfried W, Remschmidt H, Hebebrand J. No evidence for involvement of
polymorphisms of the dopamine D4 receptor gene in anorexia nervosa,
underweight, and obesity. Am J Med Genet 1999; 88(6):594–597.
56. Campbell DA, Sundaramurthy D, Gordon D, Markham AF, Pieri LF. Association
between a marker in the UCP-2/UCP-3 gene cluster and genetic susceptibility to
anorexia nervosa. Mol Psychiatry 1999; 4(1):68–70.
57. Ando T, Ishikawa T, Kawamura N, Karibe M, Oba M, Tatsuta N, Hara S, Takii M,
Naruo T, Takei M, Kurokawa N, Nozoe S, Kubo C, Komaki G. Analysis of tumor
necrosis factor-alpha gene promoter polymorphisms in anorexia nervosa. Psychiatr
Genet 200; 11(3):161–164.
58. Frisch A, Laufer N, Danziger Y, Michaelovsky E, Leor S, Carel C, Stein D, Fenig

S, Mimouni M, Apter A, Weizman A. Association of anorexia nervosa with the
high activity allele of the COMT gene: a family-based study in Israeli patients. Mol
Psychiatry 2001; 6(2):243–245.
59. Vink T, Hinney A, van Elburg AA, van Goozen SH, Sandkuijl LA, Sinke RJ,
Herpertz-Dahlmann BM, Hebebrand J, Remschmidt H, van Engeland H, Adan RA.
Association between an agouti-related protein gene polymorphism and anorexia
nervosa. Mol Psychiatry 2001; 6(3):325–328.
60. Koronyo-Hamaoui M, Danziger Y, Frisch A, Stein D, Leor S, Laufer N, Carel C,
Fennig S, Minoumi M, Apter A, Goldman B, Barkai G, Weizman A, Gak E.
Association between anorexia nervosa and the hsKCa3 gene: a family-based and
case control study. Mol Psychiatry 2002; 7(l):82–85.
61. Eastwood H, Brown KM, Markovic D, Pieri LF. Variation in the ESRl and ESR2
genes and genetic susceptibility to anorexia nervosa. Mol Psychiatry 2002; 7(l):
86–89.
62. Urwin RE, Bennetts B, Wilcken B, Lampropoulos B, Beumont P, Clarke S, Russell
J, Tanner S, Nunn KP. Anorexia nervosa (restrictive subtype) is associated with a
polymorphism in the novel norepinephrine transporter gene promoter
polymorphic region. Mol Psychiatry 2002; 7(6):652–657.
63. Hebebrand J, Fichter M, Gerber G, Gorg G, Hermann H, Geller F, Schafer H,
Remschmidt H, Hinney A. Genetic predisposition to obesity in bulimia nervosa: a
mutation screen of the melanocortin-4 receptor gene. Mol Psychiatry 2002; 7:
647–651.
64. Pettersson K, Gustafsson JA. Role of estrogen receptor beta in estrogen action.
Annu Rev Physiol 2001; 63:165–192.
65. Wade GN, Gray JM. Gonadal effects on food intake and adiposity. A metabolic
hypothesis. Physiol Behav 1979; 22:593.
66. Dagnault A, Richard D. Involvement of the medial preoptic area in the anorectic
action of estrogens. Am J Physiol 1997; 272:R311-R317.
67. Boss O, Samec S, Paoloni-Giacobino A, Rossier C, Dulloo A, Seydoux J, Muzzin
P, Giacobino JP. Uncoupling protein—3: a new member of the mitochondrial
carrier family with tissue-specific expression. FEBS Lett 1997; 408:39–42.
68. Del Mar Gonzalez-Barroso M, Ricquier D, Cassard-Doulcier AM. The human
uncoupling protein-1 gene (UCPl): present status and perspectives in obesity
research. Obes Rev 2000; l(2):61–72.
69. Fleury C, Neverova M, Collins S, Raimbault S, Champigny O, Levi-Meyrueis C,
Bouillaud F, Seldin MF, Surwit RS, Ricquier D, Warden CH. Uncoupling
protein-2: a novel gene linked to obesity and hyperinsulinemia. Nat Genet 1997;
15:269–272.
70. Lu XY. Role of central melanocortin signaling in eating disorders. Psychopharmacol
Bull 2001; 35(4):45–65.
71. Chandy KG, Fantino E, Wittekindt O, Kalman K, Tong L-L, Ho T-H, et al.
Isolation of a novel potassium channel gene kSKCa3 containing a polymorphic CAG
348 GRICE
repeat: a candidate for schizophrenia and bipolar disorder? Mol Psychiatry 1998; 3:
32–37.
72. Dror V, Shamir E, Ghanshani S, Kimhi R, Swartz M, Barak Y, et al. hSKca3/
KCNN3 potassium channel gene: association of longer CAG repeats with
schizophrenia in Israeli Ashkenazai Jews, expression in human tissues and
localization to chromosome Iq21. Mol Psychiatry 1999; 4:254–260.
73. Kaye WH, Jimeson DC, Lake CR, Ebert MH. Altered norepinephrine metabolism
following long-term weight recovery in patients with anorexia nervosa. Psychiatry
Res 1985; 14:333–342.
74. Pirke KM, Kellner M, Phillip E, Laessle R, Krieg JC, Fichter MM. Plasma
norepinephrine after a standardized test meal in acute and remitted patients with
anorexia nervosa and healthy controls. Biol Psychiatry 1992; 31:1074–1077.
75. Devlin B, Bacanu SA, Klump KL, Bulik CM, Fichter MM, Halmi KA, Kaplan AS,
Strober M, Treasure J, Woodside DB, Berrettini WH, Kaye WH. Linkage analysis
of anorexia nervosa incorporating behavioral covariates. Hum Mol Genet 2002; 11
(6):689–696.
76. Bulik CM, Devlin B, Bacanu SA, Thornton L, Klump KL, Fichter MM, Halmi KA,
Kaplan AS, Strober M, Woodside DB, Bergen AW, Ganjei JK, Crow S, Mitchell J,
Rotondo A, Mauri M, Cassano G, Kellp P, Berrettini WH, Kaye WH. Significant
linkage on chromosome 10p in families with bulimia nervosa. Am J Hum Genet
2003; 72:200–207.
77. Rankinen T, Perusse L, Weisnagel SJ, Snyder EE, Chagnon YC, Bouchard C. The
human obesity gene map: the 2001 update. Obesity Res 2002; 10 (3): 196–243.
15
Management of Eating Disorders: Inpatient
and Partial Hospital Programs
Wayne A.Bowers, Arnold E.Andersen, and Kay Evans
University of Iowa Hospital & Clinics
Iowa City, Iowa, U.S.A.
It was the best of times; it was the worst of times. One might use that phrase
when examining the current state in the management of eating disorders,
especially when management occurs in an inpatient or partial hospitalization
setting. Inpatient treatments for eating disorders have evolved considerably,
with increasing acceptance of a multidisciplinary approach, but still remain
overly variable even with the current amount of evidence-based information.
Earlier hospital-based treatments included one or several of the following: high
doses of antipsychotic medications, tube feeding, nursing management, bed rest
with a high-calorie diet, strict behavioral contingency management,
hyperalimentation, and a variety of medical regimes (1–5). The literature also
describes a variety of variably successful psychological interventions to restore
weight among persons with an eating disorder. Several controlled trials of
behavior therapy have been reported (6–8). Strict behavioral interventions have
recently been criticized as being narrow minded, leading to short-term
compliance but frequent relapse. Long-term outcome studies suggested that
weight restoration alone by strictly medical or behavioral methods was a
temporary phenomenon, which at times produced negative effects (9).
A multidimensional perspective to the management of eating disorders was
developed (10–13) that advocated that weight gain achieved by compre hensive
management of eating disorders in the context of identifying, challenging, and
changing distorted cognitions was more enduring. Behavior therapy was best
seen as one aspect of a total treatment program that included additionally
required cognitive, family, and medical interventions (14,15). The complex
nature of eating disorders suggested the need for a coordinated,
multidisciplinary approach to treatment, focusing on the combined biological,
social, behavioral, and psychological needs of the patient. Successful treatment
was described as a skillful blend of weight restoration, psychotherapy
(individual, group, family), psychoeducational interventions, medical
management, and, at times, pharmacotherapy (10,14). Pharmacotherapy was
seen as an adjunct to the other therapies. Low-dose antipsychotics and
350 BOWERS ET AL.
antianxiety agents are at times useful to reduce anxiety associated with fear
of loss of control surrounding meals and weight gain, but are not used to
increase hunger. While antidepressant medication was suggested to treat
concomitant depressive symptoms or to reduce obsessive-compulsive symptoms
sometimes seen in these patients, such agents are relatively ineffective in starved
patients (16,17). Some classes of antidepressants have been found to be useful in
maintenance of weight after restoration (18).
Over the past 15 years the most progress has been made in creating effective
outpatient treatments for bulimia nervosa (19,20). The use of cognitive-
behavioral therapy and interpersonal therapy has been shown to be effective in
both short- and long-term treatment (21,22) and have reduced the need for
hospitalization for many individuals with bulimia nervosa. In a similar manner,
the use of psychopharmacological agents, often layered on top of manual-based
proven psychotherapies, have been shown to be effective in the management of
bulimia nervosa (18,23,24). With a greater understanding of treatment options
for bulimia nervosa the need for hospital-based intervention has decreased
except where outpatient treatments have failed or patients have severe
comorbidity.
In the same time period, less progress has been made in the effective
management of anorexia nervosa. Although the use of family therapy has been
shown to work effectively with adolescents and this approach has been
manualized (25), it is still not widely used in practice. However, it may be very
effective in a limited number of younger patients. Although many patients are
successfully treated in an outpatient setting, hospital-based care by a skilled team
remains the intervention of choice for those individuals who fail outpatient
treatment settings or are too ill. The literature has shown that inpatient care is
an effective method to manage both the physical and psychological aspects of
anorexia nervosa (14). However, the change in health care influenced by health
maintenance organizations (HMOs) has altered the face of inpatient treatment.
The pressure to reduce length of stay (LOS) for hospitalized patients and
reduced insurance coverage for the management of eating disorders has limited
treatment. Bezold et al. (26) reported that the average length of stay in a
psychiatric hospital decreased 25% between 1988 and 1992. The pressure to
reduce cost is translated into reductions in LOS for individuals with an eating
disorder. The same type of decrease in LOS has been reported regarding
hospitalization for eating disorders. Over a 15-year period from 1984 to 1998,
the length of stay for a hospitalized eating disorder patient went from 149.5
days in 1984 to 22.7 days in 1998 (27). The reduction in LOS affects the health
of the patient, adversely influencing long-term care, and contributing to high
readmission rates (28).
INPATIENT AND PARTIAL HOSPITAL PROGRAMS 351
One positive result of managed care creating pressure to reduce LOS has
been the development of day treatment or partial hospitalization programs
(PHPs) (29). Although these programs are influenced by their own health care
climates, there are similarities among PHPs (30). Among the common elements
are use of a multidisciplinary staff, group treatment as a primary method of
treatment, and the use of the PHP as part of a continuum of care for some
programs. The design of the program often reflects the area in which the
program works, such as a freestanding program or as part of a hospitalbased
treatment model. Most programs run from 3 to 5 days a week and integrate
some form of cognitive-behavioral therapy, readiness, or motivational therapy
into the treatment (31). Empirical data about the effectiveness of PHPs is
sparse. However, there is an indication that these programs can be effective in
continuing gains from an inpatient treatment program as well as assist reduction
in symptoms when patients “step up” from outpatient status (28,32). One
study (28) documented empirical decision rules for optimizing successful
transition from inpatient to PHP.
The American Psychiatric Association guidelines for the management of eating
disorders (33) give both specific and broad recommendations for evidence-
based, current best-practice treatment of eating disorders resulting from
research studies and clinical consensus (34). The revised guidelines have detailed
the necessity of integrating nutritional rehabilitation, psychosocial treatments,
medical procedures, and psychopharmacological interventions, and summarize
levels of care for a treatment continuum. This continuum includes outpatient,
intensive outpatient, partial hospitalization and full-day programming, and
residential and inpatient care. The remainder of the chapter presents a
prototype for inpatient and partial hospitalization based on APA guidelines in
the management of eating disorders.
INPATIENT TREATMENT
The comprehensive treatment of anorexia nervosa often requires inpatient care
as part of a continuum of care to restore healthy mental, physical, and social
functioning. This type of program must achieve safe, prompt, and effective
short-term hospital-based improvement. It must also prepare patients for
transition to a less intense, step-down, partial hospitalization treatment,
followed by long-term continued outpatient care emphasizing relapse
prevention and health promotion. The conceptual model most appropriate for
guiding the management of anorexia nervosa is that of a multifactorial etiologic
process. Eating disorders have never been adequately explained by single
etiologic factors but require a more complex, multifactorial understanding of
both origin and treatment. Since treatments logically grow out of assumptions of
352 BOWERS ET AL.
the nature of the disorder, the clearest possible description of known

contributing factors is important for guiding effective management of eating
disorders.
Admission to hospital for management of anorexia nervosa or bulimia

nervosa remains a clinical decision based on multiple factors, summarized in
Table 1. Many of these factors interact with or potentiate each other. For
example, a very rapid weight loss of 25 pounds may be medically more
dangerous than a slower weight loss of 40 pounds. Hypokalemia with an
irregular but nonbradycardic heartbeat may be more medically serious than a
very slow regular heartbeat of 40, gradually attained.
The broad goals of inpatient care are weight restoration and beginning
treatment for the psychological and environmental factors that contribute to
maintenance of the disorder, especially overvalued beliefs and cognitive
distortions, and dysfunctional family systems. Table 2 summarizes significant
but achievable goals for the inpatient care of persons with anorexia nervosa.
Weight restoration (a vital but not exclusive goal) means restoration of a fully
healthy body weight, with rebuilding of body and organ tissue as well as organ
functioning, not excessive fluid weight, as may occur with hyperalimentation.
Restoration to a healthy body weight is a means, not an end, to comprehensive
treatment. The conclusive work of treatment involves a fundamental and
enduring change in distorted thinking concerning weight, shape, size, and
appearance. Treatment is focused on decreasing the overinvestment in thinness
as a means of dealing with crucial central issues in life, such as mood regulation,
personal identity, or family stability.
The initial goal of inpatient care is medical stabilization. Medical stabilization
is intended to differentiate between slowly produced symptoms of starvation
that are part of the body’s adaptive response to decreased energy intake and
which will generally respond to simple nutritional rehabilitation versus those
medical signs and symptoms that are life threatening or atypical. This distinction
requires the clinician to thoroughly understand the adaptive responses of the
body to starvation. Many of the social behaviors and psychological symptoms
attributed to anorexia nervosa are, in fact, due to starvation and will normalize
by restoration to a healthy body weight. The rapidity of weight loss, methods of
weight loss, physical examination results, and laboratory tests are some of the
factors that need to be understood in the acute medical stabilization of the
patient. Some medical symptoms will improve with weight restoration alone
(stable bradycardia in most cases) whereas others require acute intervention
(prolonged QT interval). The key is to distinguish between adaptive responses
TABLE 1 Criteria for Admission to Inpatient Care for Eating Disorders
TABLE 2 Goals of Treatment
to starvation, which will improve without specific intervention versus

potentially life-threatening signs and symptoms.
The significance of nutritional rehabilitation cannot be stressed enough in any
treatment setting. Patients are not able to perform the detailed psychological
work of challenging and changing beliefs without also allowing their bodies,
especially brain functioning, to recover physically as well. Placing the majority
of the recovery and healing work in the beginning of treatment on
354 BOWERS ET AL.
psychotherapy alone without concurrently working on weight restoration is a

questionable use of time and money.
In an inpatient or partial hospitalization setting, food is the patient’s primary
medicine. Sometimes it is the only medicine. At other times, it is one of several
medications. The use of a treatment protocol that deals with all of the specifics
related to management of the patient’s weight restoration creates consistency.
Nurses and nursing assistants initially remain with patients for 24-hour support
and supervision, until a normal eating pattern is established and comprehensive
assessment of the patient’s psychological and physical state has been obtained.
Nurses sit with patients at all meals and encourage them to eat. We emphasize
psychological support and use the milieu for group encouragement while
empathizing with the patient’s fear of fatness. No discussion of weight or
calories is permitted; rather, we emphasize self-understanding of the patient’s
feelings and thoughts. An empathic nursingsupervised weight restoration
program using normal food in a milieu setting with group support results in
patients’ beginning to eat three meals a day with only moderate anxiety within
24 hours. Occasionally, very anxious patients receive a small amount of
antianxiety agent (lorazepam, 0.5 mg 1 h before meals) for a week or two. Rarely,
small doses of antipsychotic medications may be used. Using our approach a
nasogastric tube is rare and in treating more than 900 patients,
hyperalimentation has not been necessary in any case.
There are several approaches to setting the goal for desired weight gain. In
general, we strive for a healthy normal weight. The three standards generally
used are (a) the Metropolitan Life tables (35) for patients 18 and over; (b) the
nomograms devised by Frisch and McArthur (36) for achieving the weight
necessary for return of periods in females and for adolescent girls; or (c) a body
mass index appropriate for age. A reasonable goal is the midrange of the weight
on the Metropolitan Life chart for a given height (with appropriate age
correction and, occasionally, frame correction) or a body mass index (BMI)
between 20 and 25. For female patients younger than 18, with secondary
amenorrhea, we use the weight identified by Frisch and McArthur (36) for a
50% chance of return of menstrual cycles. It should be noted that the weight for
return of periods is about 10 pounds higher than the weight required to begin
menstrual cycles during normal development. For patients below 14, 100%
weight for height for age is used as the definition for 100% of expected
weight in children and young adolescents, as available on the internet at
http://www.cdc.gov/growthcharts/.
However, picking a number from a chart is not the whole answer. Some
attention should be given to the weight at which the patient functioned well if
she or he had a time of stable weight and height before the onset of illness. The
average anorexic patient often begins dieting at 5–10% above the matched
population ideal weight at the onset of dieting. There is a rationale for setting the
goal weight of these patients at 5–15% above the “ideal” weight. Since many of
these patients may be biologically normal only when above the ideal in weight.
However, few patients accept this reasoning and few clinicians practice
individualization of weight goals within the normal range.
Where practical considerations dictate a short inpatient treatment period,
moderate weight gain to 85–90% of normal may have to be accepted. In this
case, close follow-up is required in a partial hospitalization program or an
outpatient clinic. A goal weight range, rather than a single point, should be set
so that patients can fluctuate comfortably within a 4- to 5-pound (1.4 kg) range.
The weight goal is not firmly set when the patient comes to the hospital but
only after treatment has been underway for several weeks. The weight range is
made as a decision among the team members if variation from protocol is
needed. Not telling patients their goal weight range until they are in the middle
of it allows changes to be made by staff without incurring the wrath or fear of
patients perceiving that a promise is being broken about a given weight range.
The initial food prescribed begins with 1200–1500 calories per day,
according to the patient’s admission weight, low in fat, salt, and lactose. No
diet foods of any kind are allowed. The dietitian has an essential role in relating
to patients, families, and staff. They take a complete nutritional history from the
patient upon admission. However, they do not discuss treatment directly with
the patient until weight is in the maintenance range. Every day the dietitian is
present at staff rounds to help make decisions about changes in dietary
programs. If the dietitian and the patients interact directly during nutritional
rehabilitation, patients may demand endless changes in menu. Patients name
three specific foods to delete from their menu, but other than these three
specific choices (e.g., artichoke, pork chops, scrambled eggs), they do not
determine the foods prescribed. Vegetarianism is permitted only if part of an
established religious or philosophical practice (for example, Seventh-Day
Adventist) preceding the eating disorder. Most adolescent vegetarianism in
eating-disordered patients represents an early phase of their eating disorder.
Calories are increased by 500 every 4–5 days until a maximum of 3500–4500
calories per day is achieved. The exact number will depend on the individual
rate of weight gain, the height of the patient, and the presence of gastrointestinal
discomfort. Once nutritional rehabilitation has been underway for several
weeks, most calories can be prescribed in fairly dense form, including a
moderate amount of fats and sweets. A safe continuing weight restoration
averaging 3 pounds a week in females and 4 pounds a week in males can be
achieved without significant medical symptoms, except for occasional pedal
edema (easily managed without diuretics by feet elevation, limitation of salt,
and psychoeducation).
356 BOWERS ET AL.
Although the initial emphasis of inpatient care is weight restoration and/or

disrupting the chaotic binge-purge cycle, additional interventions, especially
psychoeducation and cognitive-behavioral therapy (CBT) are soon added. These
interventions include various theoretical models such as psychodynamic,
systems, and cognitive-behavioral therapies using individual, group, family,
occupational, and recreational therapy formats (15). Specific practice guidelines
in the management of anorexia nervosa emphasize the use of individual and
family therapy among other approaches during inpatient treatment (33). These
guidelines suggest the use of cognitive therapy as a model for psychotherapy.
CBT has been shown to be the most effective psychological method in the
management of bulimia nervosa (37). CBT can be effectively applied to
management of anorexia nervosa, although definitive demonstration of its
effectiveness in anorexia nervosa has not been proven. Shared goals between the
two disorders include decreased illness-driven eating and social patterns, and
their replacement by healthy behaviors. Because anorexia nervosa and bulimia
nervosa share symptoms (overemphasis on body shape and weight as sources of
self-esteem and identity, relentless drive for thinness, phobic fear of normal
weight, rigid dietary habits), CBT seems well suited for the management of a
heterogeneous mix of eating disorders during inpatient treatment (37–39).
What is suggested as a model for inpatient treatment is a hospital unit with
all psychological interventions based on the principles of CBT. CBT
conceptualizes eating disorders in a developmental framework with
primacy on cognition mediating distressed emotion and resulting abnormal
behavior (40,41). However, this theory also accommodates factors from
psychodynamic and biological paradigms. Cognitive theory in general, and
especially theory of eating disorders, views biology as an important part of the
disorder, similar to other disorders such as depression. The cognitive model also
blends with more dynamic approaches, as it encompasses the idea that an
individual’s life does not occur in a vacuum. The model will acknowledge many
interpersonal and intrapersonal factors that contribute to the etiologic
progression and maintenance of the disorder. The cognitive model places a high
value on developing alternate ways of seeing the world and coping with day-to-
day events. Cognitive theory and therapy work with dynamic concepts by
changing developmental templates, schemas, and core beliefs (42,43).
The cognitive model views an eating disorder as a final common pathway of
multiple events or experiences (40). Through various life experiences
interacting with personality features, specific distorted ideas regarding the self,
the world, and the future are learned (schemas), which then create vulnerability
to an eating disorder. Vulnerable individuals, who are often introverted,
sensitive, persevering, fearful of the challenges of development, and socially
isolated, develop the idea that weight loss will somehow alleviate psychological
distress and dysphoria (40,44). Dieting, weight loss, and attaining thinness
become factors these individuals manipulate in an attempt to exercise control
over their internal and external environments (44,45). Continued weight loss,
which has become a sign of control, leads to social praise at first; then, shortly
after, social criticism that may be threatening to these patients’ sense of internal
and external control. This perceived criticism leads to increased social isolation
that reinforces distorted cognitions and maladaptive behaviors of an eating
disorder.
We use a design called the comprehensive model of cognitive therapy (46) to
organize an inpatient unit. Unit leaders, primary therapists, and adjunctive
therapists are all trained in cognitive therapy when using a comprehensive model.
Each intervention is designed to blend with the other psychotherapies. The
milieu (a) fully accommodates cognitive therapy; (b) supports individual,
family, and group therapies with cognitive therapy interventions; and
(c) integrates psychoeducational programs to complement and reinforce the
learning of cognitive therapy principles. Beck’s (47) original model,
modifications of this model for inpatient settings (39,48), and adaptations by
Garner et al. (44) for outpatient management of anorexia nervosa along with
Fairburn’s (49) approach to managing bulimia nervosa are the foundation of unit
structure.
The cognitive therapy milieu operates on the premise that patients are
affected by their environment and in turn influence that environment. The
milieu is a microcosm of the patient’s world providing the treatment staff and
patient with an opportunity to understand their problems. Recreation of the
patient’s world assists the treatment team in identifying cognitive distortions,
basic assumptions, schemas, and core beliefs as they occur, with recognition of
transference relationships as conveying vital information about family and peer
relationships. Through recognition, acknowledgment, and acceptance of their
problems, patients can gain new insight that allows them to address and modify
their distorted thinking patterns and beliefs.
PROGRAM SCHEDULING
Patients are involved in some form of treatment much of the day. A detailed
weekly schedule is displayed in Table 3, which includes several types of group
treatments. Typically they begin with a psychoeducational group, in which they
learn about basic cognitive therapy principles, the effects of starvation, and
principles of healthy social and psychological functioning. Patients participate in
an activities therapy group three times per week with content focusing on
building leisure time skills. An occupational therapy group meets twice a week.
The first session focuses on meal planning and purchasing. The second session
358 BOWERS ET AL.
focuses on meal preparation and is followed by a structured group discussion of

attitudes and emotions after the meal. Twice a week the patient participates in
an additional occupational therapy group that focuses on coping skills and
emphasizes role playing. Patients participate in a cognitive therapy group three
times a week and that group alternates with a body perception group. Snacks
are provided in the groups with a structured discussion time following the snack.
Dinner is served with a structured observation and discussion time following the
meal. Structured observation is divided into various levels with the most intense
level being 24-h continuous observation by the unit staff. As a patient progresses
through treatment, the amount of time under staff observation is changed
(Table 4). Study time, if appropriate, is scheduled before the evening meal,
except on the weekends. A final social or recreational activity occurs from
7:00 p.m. until 8:30 p.m.
The interdisciplinary approach provides the patient with a comprehensive
treatment program whereby each member of the team has a clear understanding
of the unique and shared treatment goals of each discipline. Team members
work closely together, meeting at least twice a week on these issues to provide
the patients with the structure they need in treatment. This team focuses on the
goal of changing illness behavior and thinking, not only in the protected
environment of the inpatient unit but on an enduring basis after discharge. The
treatment team consists of a staff psychiatrist, psychiatry resident, psychologist,
advanced registered nurse practitioner (ARNP), primary nurse, social worker,
occupational therapist, dietitian, and activities therapist. Some patients will also
need a vocational rehabilitation therapist or, if the patient is a student, an
educational consultant. This team works with the patient to understand his or
her individual life history, including the cognitive conceptualization behind the
eating disorder. The issues the team will work on in collaboration with the
patient are multifaceted. These issues include self-esteem, body image, possible
sexual abuse, addictions, family issues, marital relationships, parent-child
relationships, and interpersonal dynamics. There is also a focus on coping skills
for stress management, compulsive exercise, leisure time activities, nutrition
and cooking, assertiveness, relaxation, mood, and perfectionism.
TABLE 3 Daily Patient Schedule

360 BOWERS ET AL.
TABLE 4 Inpatient Eating Disorder Observation Levels (EDO)
A comprehensive psychiatric evaluation is obtained by a detailed psychiatric

interview. This time-honored process is supplemented by a standardized written
assessment tool, such as the Eating Disorders Evaluation 50). This
comprehensive, semistructured interview is designed to assess broad clinical
features of an eating disorder, especially attitudes and behaviors related to the
illness over the preceding 4 weeks. A part of the psychiatric case organization
includes all features of the case, including a summary of pertinent features,
diagnosis and differential diagnosis, etiological factors, treatment plans, and
prognosis. Brief self-report measures such as the Eating Disorders Inventory
(EDI; 51) and Eating Attitudes Test (EAT; 52) identify specific aspects of the
disorder, including desire for thinness, binge-purge behavior, and restraint in
eating. Because the EAT and EDI are valid, reliable, and easily administrated
instruments, repeat administration of these tests can document severity of
illness at admission as well as record improvements during treatment.
Often treatment of comorbid conditions is as challenging as the primary
eating disorder itself. Personality functioning and the presence of axis I
comorbid disorders such as depression and anxiety disorders are important to
assess. Affective disorders are the most common associated axis I comorbidity,
present in 50–80% of eating disorders with bulimic symptomology, and a
substantial number of anorexic patients, with many having secondary but severe
362 BOWERS ET AL.
depressions. Intellectual functioning warrants assessment for a number of

reasons. Many patients have expectations for academic or vocational
performance beyond their ability based more on persevering traits than inherent
giftedness. Evidence of actual level of intellectual functioning may be
sympathetically used to change these expectations and therefore modify some of
the cognitive distortions arising from such misperceptions. In addition,
significant neuropsychological deficits are present in about 35% of starved
patients.
Although supportive psychotherapy and psychoeducation are often used
during the early phases of inpatient care, we are more aggressive in beginning a
cognitive-behavioral psychotherapeutic approach from the earliest part of
treatment. Individual and group therapy use Beck’s model (46) with
modifications specifically for an inpatient eating disorders unit (39,53). It must
be recognized that this approach may be only partially effective until there is
adequate weight restoration, but the basic concepts are familiar when patients
are more effective in cognitive functioning. However, in our experience even
patients who have difficulty with the more abstract concepts of cognitive
therapy can benefit from the early behavioral components such as mood
monitoring, problem solving, and graded task assignments. We also believe that
early use of behavioral CBT methods establishes the groundwork for more
cognitive interventions. With ongoing weight restoration through nutritional
interventions early in therapy, there is a synergetic effect with the cognitive
therapy. This work will be increasingly successful as weight restores a patient to
85–90% of a healthy goal weight and then to fully normal weight for age and
height.
Individual cognitive therapy also places a high value on the articulate
identification, expression, and understanding of emotions. Many times, “I feel
fat” is the global and generic phrase for any dysphoria prior to cognitive therapy.
Increased awareness, understanding, and expression of emotion are gradually
achieved through the therapist’s observation of inconsistencies, incongruities,
and inappropriate emotional reactions from the patient’s everyday events.
Confirmation and reinforcement of emotions that are a genuine part of the
patient’s past and present experience are essential. The patient is encouraged to
express all emotions, especially “unacceptable” emotions. With the therapist
serving as a model for expression of emotion, the patient can learn that open
expression of emotions does not lead to rejection (40) or out-of-control
behavior. There is a consistent emphasis on separating “I feel” from “I think”
statements, e.g., changing “I feel fat” to “I think I am fat and those thoughts
make me anxious and depressed.”
Cognitive therapy attempts to help the patient recognize and change the rigid
standards employed to determine self-worth. The message communi cated is
that positive self-evaluation develops from success through mastery in small,

stepwise, increasingly challenging personal activities. Exceptional or perfect
performances are not necessary. Competence by way of reasonable standards
(emphasizing adequacy, not perfection), as well as learning and accepting “in-
betweens,” is very important. Self-acceptance, despite personal shortcomings
based on unrealistic standards, is a fundamental goal for the psychotherapist
working with a patient with an eating disorder. Cognitive therapy can also
address the lack of trust in and the fear of feelings or expression of emotions (40).
To accomplish this, the therapist must confirm genuine expressions of inner
feelings while labeling misconceptions and errors in the patient’s thinking.
Denial or absence of seemingly appropriate affect should be explored in greater
detail. It is critical to progress slowly and let the patient learn to identify his or
her emotions. Another crucial component of CBT is assisting patients in
identifying their affective states and in promoting acceptance of these feelings as
real and wholesome, with “no bad feelings, but responsible actions.”
Group Therapy
Psychoeducational Group
All groups are ongoing and a patient will enter group immediately after
admission to the unit. The psychoeducational group explores various aspects of
cognitive therapy (i.e., cognitive distortions, automatic thoughts) in a didactic
fashion and its relationship to an eating disorder. The purpose of this group is to
teach patients the basic concepts of cognitive therapy and information about the
effects of the disorder. After each group, assignments using these concepts are
given to the patients to complete before the next group. Some of the objectives
of this group are that the patient will:
Define the basic cognitive therapy principles (i.e., cognitive triad,

assumptions, automatic thoughts, cognitive distortions, schemas, core
beliefs).
Demonstrate the ability to rate his or her mood.
Identify automatic thoughts that he or she experiences.
Demonstrate application of the reframing technique for changing negative
automatic thoughts.
Demonstrate an ability to use the thought records.
Define and then apply the problem solving to his or her own life
experiences.
364 BOWERS ET AL.
Define or demonstrate at least one cognitive therapy intervention that is

personally applicable (e.g., identifying automatic thoughts, chal lenging
automatic thoughts, reattribution techniques, advantages/disadvantages
and homework).
The instructional methods used by the group leaders are didactic, role plays,
discussions, practice sessions, homework, peer feedback, and creative
expression exercises (artwork, etc.). The psychoeducational groups run daily
for 1 hour. Outcomes of the group are measured by satisfactory completion of
homework assignments, active participation in the group discussions, and
activities that reflect an understanding of the content being taught.
The materials and information given to each patient have been adapted from
various authors in the area of cognitive therapy. Ancillary areas covered in this
group include exploring self-identity and self-esteem, values clarification,
feelings identification, and problem solving. Other psychoeducational material
has been integrated directly relating to eating disorders. Included is information
on the effects of starvation, how the disorder functions psychologically, and the
social and media views of an individual’s appearance (54).
Cognitive Group Therapy

Group cognitive therapy combines Beck’s theory and interventions in a process-
oriented framework (15,55). Group therapy in the management of eating
disorders is increasingly recognized as an important, effective, and economical
psychotherapeutic tool (15). Group therapy may not only be economical; it
contains factors unique to effective groups, namely, peer challenges, feedback,
support, and empathic paralleling. As a basic approach, a blend of process
orientation (56) and cognitive-behavioral principles (15,55) appears to be most
effective. Blending of these two models gives the group latitude to deal with
personal and interpersonal issues. Using the curative factors of group (56) and
cognitive therapy principles creates a focus on cognitive and developmental
factors involved with eating disorders. The group can also influence the
perceptions of the patients and permit patients to assist in each other’s recovery
from disorder through self-disclosure and confrontation of symptomatic
behavior, distorted ideas, and negative attitudes. An inpatient group may be
more diagnostically heterogeneous (including anorexia nervosa and bulimia
nervosa). However, the common themes of both eating disorders blend the
otherwise diverse population into a psychologically homogeneous group by
emphasizing the common features.
In the groups that are run on the unit, such concepts as cognitive distortions,
automatic thoughts, schemas, and core beliefs are discussed in a setting that uses
situations that arose during the day or between groups. In addition, material
that comes from the group itself is available to show how the ideas of cognitive
therapy are ever present with patients. It is also an opportunity for more skilled
members of the group to assist newer group members in first identifying
automatic thoughts, schemas, and core beliefs.
A cognitive therapy group challenges the weight- and shape-centered view of
the world that patients have established through their distorted cognitions. In
our experience, much of the group work lends itself toward helping the patients
understand how their cognitions affect their mood and consequent behaviors.
Another healing factor is the ability of group members to easily identify in
others the ramifications of their own eating disorder. As patients help others
identify and change negative cognitions, they also improve their own, often
coming to resolution of their issues as reflected in others.
Body Perception Group

Our body perception group focuses on helping patients understand their body
distortions and how these distortions affect their lives and sustain the eating
disorder. We have found this area to be one of the most difficult aspects of the
illness. The group has been extremely valuable for adjusting distorted body
images. The patients are supportive yet confrontive, helping each other face
these distortions and bring their cognitions into the realm of reality. What we
often see is that the distortion usually does not completely resolve, but through
this work the patients learn to diminish the impact of these thoughts and begin
to put them into proper perspective. The cognitive therapy model is effective in
helping patients challenge their belief systems and realize their lack of validity,
as well as establishing alternatives. When patients are challenged to provide a
rationale to support these distortions, they are frequently surprised to find that
they cannot offer any validation.
This group operates from two basic premises: (a) all patients in the group
have the diagnosis of an eating disorder and (b) the diagnosis of an eating
disorder usually includes a distorted body image. Therefore, the patients cannot
trust their perceptions of how they appear to themselves and others. The format
of the group is discussion based, with the patients sharing in each session how
they are feeling about the changes their bodies are undergoing in the process of
restoration. It is an accepted fact in the group that the patients’ body sizes are
changing and that is a frightening experience for them. The discussions within
the group are twofold. First, patients are encouraged to openly and honestly
express feelings and thoughts concerning body perceptions. Second, we move
to fact-versus-fiction discussions regarding distorted body image. Patients need
to discover that to feel positive about their bodies they must learn to rely on
366 BOWERS ET AL.
facts rather than distorted perceptions. By identifying, challenging, and

reframing automatic thoughts, patients begin to feel different about their
bodies.
Family Therapy
Family therapy maintains a cognitive framework designed to work with family
communications and schemas (57). Family therapy is important to treatment
outcome with this patient population and is an ongoing process in our program
of care. Parents and siblings frequently exhibit a sense of hopelessness regarding
the recovery of their family member, often intermixed with anger and anxiety,
with fathers more often showing the former and mothers the latter. Many of the
families we have worked with have been told that their family member will
probably not survive. Family therapy sessions are held on a regular basis with
the patient, family members, and/or a significant other. The content is
determined by the patient’s issues that have been pinpointed in the day-to-day
treatment process. These families often have difficulty identifying feelings and
expressing emotions, and not infrequently there are underlying family secrets
and unspoken issues that impair open communication. The hopelessness and
anger these individuals are feeling must also be addressed and assisted to
resolution. The focus of the sessions is on the interactions between the patient
and the family and resolution of the eating disorder. Our premise is that the
eating disorder maybe a symptom of other unresolved underlying problems.
However, families are assumed to be blameworthy for any aspect of the illness,
and often need to be relieved of the belief that they could have prevented the
illness, and that in fact this is an illness rather than a personal or family failing.
This work is always done within the CBT framework. The families are taught
through the therapy sessions to identify feelings and automatic thoughts, to
challenge automatic thoughts, and to reframe and examine the changes in their
feelings before and after the challenge. Some of the parents have marital issues
that need to be resolved, and such couples are referred to a therapist in their
locale or within the facility of treatment where they can pursue these issues
more intensely. Such issues may include triangulating, untreated mood
disorders, waning marital bonding, and so forth.
Another form of family therapy is increasingly important while working with
inpatients and partial hospitalization programs. A manual for psychotherapeutic
interventions with families for adolescent AN has recently been developed (25)
incorporating elements of the Maudsley treatment program that have been
found effective for adolescent AN patients (58). This protocol underscores the
central role of parents as a resource in the treatment of adolescent patients with
AN. Unlike more traditional family therapy models in which the patient is seen
as having developed a problem in response to external or internal factors (e.g.,

genetic, physiological, familial, or sociocultural), the Maudsley approach
focuses on how the family can effectively promote healthy eating behavior per
se. This treatment emphasizes the parents’ ability to help the adolescent
overcome the “intrusion” of AN in his or her normal development. The main
focus of treatment is empowerment of the parents so that they can succeed in
restoring their starving child. It is only after the eating disorder has been
successfully addressed that the parents will hand control over eating back to the
adolescent. It is at this point that the family will begin to discuss other issues.
The theoretical underpinning of the Maudsley approach is the view that the
adolescent is imbedded in the family and that the parents are critical in the ultimate
success of treatment. The eating disorder is seen to be interfering with regular
adolescent development. Therefore, the parents should take an active role in
their offspring’s treatment while at the same time showing respect for the
adolescent. This treatment pays close attention to adolescent development and
guides the parents to assist their adolescent with developmental tasks once the
eating disorder has been removed. In doing so, any meaningful work on other
family conflicts or disagreements must be deferred until the eating disorder is
out of the way. In some cases, effective parental guidance may prevent the
hospitalization of even very starved adolescent patients.
Role of Medications in Treatment

The use of medicines to stimulate appetite is generally unhelpful and
counterproductive for two reasons. First, appetite mechanisms in brain are
normal even in starved individuals. Second, without psychotherapeutic
improvement, increased appetite leads to increased anxiety and attempts to vomit
or exercise. Antidepressants have been advocated for both anorexia and bulimia
nervosa and may well have a role in management of comorbid depression in
both (17). However, a suggested practice has been to prescribe antidepressants
only after patients’ body weights are normal, their eating patterns are normal,
and after they have had experience with intensive psychotherapy.
Antidepressants have been demonstrated to be generally ineffective in starved
patients (16). After these three goals have been achieved, if the patient still
meets criteria for major depressive illness, then antidepressants are prescribed.
The current reports in the literature are not yet sufficiently convincing to
suggest that antidepressants have a routine role during the low-weight phase of
illness. Fluoxetine is approved for management of bulimia nervosa. We
selectively, rather than generally, use selective serotonin reuptake inhibitors for
management of the binge-purge subtype of anorexia nervosa since CBT alone is
often effective. Obsessive-compulsive symptoms may also be improved with
368 BOWERS ET AL.
weight restoration and should be reevaluated after weight restoration before anti-
OCD medications are used. Exceptions may of course occur. Occasionally
patients benefit from prokinetic agents (Reglan) to decrease bloating, or H2
blockers or proton pump inhibitors (Zantac or Prilosec) where reflux
esophagitis is present.
THE DISCHARGE PROCESS

Aftercare planning begins as close as possible to the time of admission. Eating
disorders severe enough to require inpatient treatment will require experienced
long-term follow-up, preferably partial hospitalization followed by experienced
outpatient treatment usually for one to several years. The data from follow-up
studies encouragingly demonstrate that 76% of eating disorder symptoms in
adolescents requiring inpatient treatment may be completely ameliorated after
several years of relapse prevention (59). The characteristics of satisfactory
aftercare include pre-discharge decision making concerning step-down to partial
hospitalization versus outpatient treatment alone. We involve the aftercare team
in the discharge process, transmitting information about the course of
treatment, sharing both the philosophy and practice of treatment. Patients
returning to rural areas or to areas without experienced professionals may soon
experience worsening of their symptoms. This is particularly true in view of the
increasingly common problem of managed care-driven discharge of patients
before they have had a chance to establish adequate weight, healthy patterns of
behavior, (eating and exercise), and decreased cognitive distortions. Data do
not support the frequently practiced discharge of patients at very low weight,
immediately after medical danger or self-harm danger has passed but well
before a healthy body weight has been achieved.
Readmission to hospital for management of relapse will generally be more
effective and shorter if it occurs sooner rather than later. Prompt readmission
should occur when the patient falls below 85% of target weight but a higher
threshold may be appropriate. In addition, maintenance of body weight in the
metastable weight range of 85–90% of target without improvement after
6 months should also warrant readmission. Other proven reasons for admission
to a hospital include return of severe depressive illness or serious medical
complications of the eating disorder. Eating disorders are spectrum disorders
with a variable course and severity. A subgroup of patients may require repeated
admissions. There is a tendency to blame patients for return to illness, as if the
eating disorder was entirely voluntary, and for some health professionals or
families to take a negative or punitive view toward eating disorder patients
requiring readmission. For those patients, a minority who do have a chronic,
severe, and relapsing eating disorder, readmission is necessary whenever
indicated by clinical symptomatology and should not be a source of stigma or

rejection. Clinicians recognize the need for readmissions in other disorders,
such as diabetes or cardiac disease, with comparable stigma.
PARTIAL HOSPITALIZATION/DAY TREATMENT

The goals of a partial/day hospitalization program in the management of
anorexia nervosa are (a) restore patients weight to 95–100% of their target
range; (b) develop effective personal application of cognitive therapy to defeat
the core psychopathology of body image distortion, drive for thinness, and fear
of fatness; (c) increase development of leisure skills—not a cultural nicety but
an essential for patients turning to their eating disorder during unstructured
time in place of driven activity or feeling “stuck” or bored; (d) develop healthy
coping skills to deal with developmental and everyday challenges; (e) replace
body image distortion, drive for thinness, and fear of fatness, with healthy,
accurate beliefs and attitudes; (f) develop healthy family relationship;
(g) develop and demonstrate accurate beliefs about nutrition, especially the role
of a moderate amount of heart-healthy lipids in place of a fat-phobia; (h) stop all
bingeing, purging, and restriction of caloric intake; and (i) develop healthy self-
esteem and mood stabilization, especially through internal selfregulation.
For those patients who are having difficulty with weight restoration at the
outpatient level, a partial/day hospitalization can also be of great value as a
“step-up” in intensity. The structured program that monitors meals and
activities can assist in positive weight restoration. Patients’ nutritional intake can
be systematically increased until they are taking in an adequate number of
calories (approx. 3500/day) to sustain a weight increase of 2 pounds per week.
Along with the monitoring of their weight restoration, a structured partial
hospitalization program can also address the medical and psychological aspects of
starvation that interfere with response to outpatient care. Often individuals with
anorexia have difficulty with weight restoration despite the best of intention and
desire. When this is the problem the use of a partial hospitalization program is
beneficial. As mentioned earlier, without adequate restoration of weight much
of what happens in therapy is hindered. The intensive structure in a partial
program is similar to that in an inpatient program, with development of a milieu
to enhance cognitive-behavioral principles.
The structure is different from inpatient care only in that part of the day is
spent in this intensive milieu. The structure of the program ideally covers from
8 a.m. until 6 p.m. in order to accommodate three full meals and at least two
snacks. One change during meals is that patients may have more access to the
selection and preparation of their own meals. Also, patients may not have the
370 BOWERS ET AL.
constant observation during meals or just after meals, as these are gradually
diminished.
Decision to admit a patient to a partial program is usually based on failure to
make progress during outpatient treatment or as a step-down from inpatient
care. Often progress is measured by weight restoration or weight stabilization,
as well as by ability to challenge cognitive distortions, and improvement in
comorbid symptoms, such as depression, obsessive-compulsive disorder, etc.
Third-party payers when working with inpatients with anorexia often request
movement to partial hospitalization. A partial program as part of a continuum
of treatment allows patients to move to less restrictive care. It would be prudent
to move from inpatient to partial hospitalization when a patient is at least
80–85% of target weight range, but individuals may need more time in
inpatient treatment based on empirical studies. Weight is not the only criteria
for movement to a less restrictive level of treatment. The ability to engage in
psychotherapy must also be taken into consideration. Also, there needs to be
criteria for when to bring a patient back onto an inpatient treatment program if
partial hospitalization is ineffective.
Treatment in a partial/day hospitalization program is primary handled in a
group format, both for cost and effectiveness reasons. Group treatment can be
varied for those patients who are more or less advanced in the use of cognitive
therapy. Group approaches also offer more socialization in the context of the
cognitive model, as well as more chances to practice the concepts during the
day. A psychoeducation group functions much as in inpatient treatment, but it
can be split into more or less advanced tracks. If a patient has begun the partial/
day program after failure in outpatient treatment the psychoeducational group
would focus on more basic principles of cognitive therapy. Groups would build
on the principles of cognitive therapy with an emphasis on the use of mood
ratings, mastery and pleasure, graded task assignments, identification of
automatic thoughts, cognitive distortions, and use of thought records. In
addition, information regarding the effects of starvation on the body and
psychological processing would be introduced with an emphasis on how these
affect the interaction of thoughts, feelings, and behavior.
Patients who are more advanced in the principles of cognitive therapy would
have a different track for the psychoeducational group with a greater focus on
the ideas of schemas and core beliefs. This group would also focus on the
societal aspects of eating disorders and discuss how these unrealistic standards
influence and perhaps contribute to their own cognitive distortions about
weight, shape, and appearance. The main influence of this group would be on
the identification of their own schemas in order to understand the
developmental aspects of their disorder. Patients would be shown methods to
spot their core distorted beliefs by monitoring their emotional shifts and
situations in which they find themselves using their disorder as a method for coping
with intense emotions. This group would also use other existing material to
assist in changing schemas and core beliefs.
Cognitive group therapy is the main psychotherapeutic vehicle in the partial/
day hospitalization program. This type of group (depending on availability of
therapists) can be run daily but must occur at least twice a week. The length of
the session also is dependent on the frequency of the meetings, ranging from
four to five times each week to longer sessions twice a week. This group can
have two possible formats. One format is based on the individual cognitive
therapy session, with definite structure. At the beginning of the group an agenda
is developed from topics presented by the group members. All members are
encouraged to be a part of the group, but more often time is spent with an
individual topic on an individual member. Other members of the group and the
group leaders give feedback to the member(s) involved. In this format group
leaders are more interactive and lead the discussion among the group members.
The alternative format is to develop a more process-oriented group in the partial/
day hospitalization program. This cognitive group therapy is similar to that
discussed in the chapter on inpatient treatment, blending of process
orientation (56), and cognitive-behavioral principles (15,55). These two models
allows more group latitude to deal with various personal and interpersonal issues.
The curative factors of group (56) and cognitive therapy principles create a
focus on cognitive and developmental factors involved with eating disorders. This
group format influences the perceptions of the patients and assists patients in
recovering from their disorder through self-disclosure and confrontation of
symptomatic behavior, distorted ideas, and schematic material. It also provides
an environment that validates emotions and assists patients in tolerating
uncomfortable feelings while facilitating self-understanding.
The body image group also parallels the group run in an inpatient setting with
a focus on helping patients understand their body distortions and how these
distortions affect their lives and sustain the eating disorder. Patients are
supportive yet confrontive with one another, helping each other face their
distortions and bring their cognitions into basis in fact.
Unlike inpatients, the patients in partial hospitalization spend part of the day
outside of a formal treatment setting. This time may be very anxiety producing
but represents an essential practical format for gaining mastery in an everyday
unstructured setting. Urges to restrict, exercise, or purge can be strong. In
order to prepare patients for these urges and for efficient use of free time, each
day is ended with a planning group emphasizing a plan for healthy thinking and
behavior when not in treatment. Format is structured around alternative healthy
plans that each member needs to practice as an alternative to not engaging in the
disorder. This may include alternatives to the urge to restrict, exercise/purge,
372 BOWERS ET AL.
or binge. Specific plans may include relaxation, journaling, use of thought

records, distraction, contacting on-call staff, and using each other as supports
during times of intense urges to engage in their disorder. Weekends also are
very difficult for some patients in a partial/day hospitalization program. Again
emphasis is placed on the patients using their own resources to not engage in the
eating disorder and to prepare plans that offer a greater sense of control over
the disorder, as well as increased pleasure in self-regulation and normal
developmental social and recreational interactions. Occupational therapy can
prepare patients for daily living experi ences and work with them in meal
preparation including choice of foods, purchasing food, and cooking meals.
Groups of patients are exposed to meal planning, shopping, meal preparation,
and the actual eating of the meal. This gives patients a chance to support each
other in what may have been a very difficult and solitary task, i.e., grocery
shopping. Food is purchased twice a week, and members of the group prepare
and eat the meal together. Again, with the help of group support and leader
assistance, the planning, preparation, and consumption of a meal is approached
and the problems arising from this task are exposed and challenged. In addition,
patients learn that food and meals can be a strong social “glue” that helps to
create the social support that often has been lacking in an anorexic patient’s life.
Activities therapy has an important role in assisting the inpatient to reduce
potentially dangerous exercise patterns, especially important in patients with
osteoporosis, and to increase healthy social, recreational, and leisure activities.
Patients need to understand what “normal” exercise looks like. Patients will
begin an exercise program while in the program, including basic aerobic
exercise and weight training under supervision of the activities therapist. The
group focuses on leisure activities and learns how to use time wisely and in
healthy, relaxing ways. This is very important for patients who have engaged in
overexercise. An exercise prescription is given in place of an all-or-none pattern
of drivenness versus sedentary patterns.
Involuntary Patients
One subject that has important bearing on inpatient treatment and, to a lesser
degree, on partial hospitalization is the use of involuntary treatment. Although a
controversial subject, involuntary commitment to treatment, through a
judiciously applied legal process, is at times a life-saving way to assist individuals
who have a life-threatening disorder but refuse hospitalization. There is a very
modest literature that looks at the usefulness of legal commitment, and
suggestions have been made that in some cases eating disorder patients who
desire not to be treated should be respected even if it may mean their
death (60). We strongly disagree. The literature that exists is generally positive
about the judicious use of involuntary treatment. Invohmtary treatment clearly

has benefit in the short term (61,62), with most patients expressing at the time
of discharge recognition of their peril at the time of admission. Also civil
commitment is a valuable treatment tool when lifethreatening situations emerge
in an individual with an eating disorder who lacks the competence to make
treatment decisions (63). It is noted that involuntary treatment does not require
more extreme measures, such as nasogastric tube feeding. In addition, there is
little support that involuntary treatment damages the therapeutic relationship
with the patient (62). The small literature that exists suggests that 10–15% of
severely ill patients who refused treatment and were involuntarily committed
have similar treatment results as those who were treated voluntarily (64). In
addition, the majority of those treated involuntarily agreed retrospectively with
the necessity of treatment and showed good will toward the treatment process.
In summary, involuntary treatment must be considered at times, especially
when the patient is in a life-threatening situation and in denial about the
necessity of treatment.
REFERENCES
1. Dally P, Sargent W. Treatment and outcome of anorexia nervosa. Br Med J 1960;
1:1770–1773.
2. Silverman J. Anorexia nervosa: clinical observations in a successful treatment
program. J Pediatrics 1974; 84:68–73.
3. Russell GFM. The present status of anorexia nervosa. Psychol Med 1977; 7:
353–367.
4. Maloney MJ, Farrell MK. Treatment of severe weight loss in anorexia nervosa with
hyperalimentation and psychotherapy. Am J Psychiatry 1980; 137:310–314.
5. Vandereycken W. The use of neuroleptics in the treatment of anorexia nervosa
patients. In: Garfinkel PE, Garner DM, eds. The Role of Drug Treatment in Eating
Disorders. New York: Brunner/Mazel, 1987.
6. Agras WS, Schneider JA, Arnow B, Rathburn SD, Telch CF. Cognitivebehavioral
treatment with and without exposure plus response prevention in the treatment of
bulimia nervosa: a reply to Leitenberg and Rosen. J Consult Clin Psychol 1989; 57:
778–779.
7. Eckert ED, Goldberg SC, Halmi KA, Casper RC, Davis JM. Behavior therapy and
8. Touyz SW, Beumont PJV, Glaun D, Philips T, Cowie I. A comparison of lenient
and strict operant conditioning programs in refeeding patients with anorexia
9. Vandereycken W. The place of behavior therapy in the inpatient treatment of
anorexia nervosa. Br Rev Bulimia Anorexia Nervosa 1989; 3:55–60.
374 BOWERS ET AL.
10. Andersen AE, Morse C, Santmyer K. Inpatient treatment of anorexia Nervosa. In:
Garner DM, Garfinkel PE, eds. Handbook of Psychotherapy for Anorexia Nervosa
and Bulimia. Guilford Press, 1985:311–343.
11. Andersen, A.E. New York:Inpatient and outpatient treatment of anorexia nervosa.
In: Brownell KD, Foreyt J.P, eds., Handbook of Eating Disorders: Physiology,
Psychology, and Treatment of Obesity, Anorexia and Bulimia. New York:Basic
Books, 1986.
12. Garner DM, Garfinkel PE, Irvine MJ. Integration and sequencing of treatment
approaches for eating disorders. Psychother Psychosom 1986; 67–75.
13. Vandereycken W. Inpatient treatment of anorexia nervosa: some researchguided
changes. J Psychiatr Res 1985; 19:413–422.
14. Andersen AE, Bowers WA, Evans KK. Inpatient treatment of anorexia nervosa. In:
Garner DM, Garfinkel PE, eds. Handbook of Treatment for Eating Disorders.
2d ed. New York: Guilford Press, 1997:327–353.
15. Bowers WA, Andersen AE. Inpatient treatment of anorexia nervosa: review and
recommendations. Harvard Rev Psychiatry, 1994.
nervosa: an historical case-control study. J Child Adolesc Psychopharmacol 1999;
9:195–201.
17. Mitchell JE, Peterson CB, Myers T, Wonderlich S. Combining pharmacotherapy
and psychotherapy in the treatment of patients with eating disorders. Psychiatr Clin
North Am 24:315–324.
18. Kaye WH, Weltzin TE, Bulik CM. An open trial of fluoxetine in patients with
anorexia nervosa. J Clin Psychiatry 1991; 52:464–471.
19. Keel PK, Mitchell JE, Davis TL, Crow SJ. Long-term impact of treatment in
women diagnosed with bulimia nervosa. Int J Eating Disord 2002; 31:151–158.
20. Rosenblum J, Forman S. Evidenced-based treatment of eating disorders. Curr Opin
Pediatrics 2002; 14:379–383.
21. Bowers WA, Ansher LS. Cognitions in anorexia nervosa: changes at discharge from
a cognitive therapy milieu inpatient program. J Cognitive Psychother 2000; 14:
393–401.
22. Wilson GT, Fairburn CC, Agras WS, Walsh BT, Kraemer H. Cognitivebehavioral
therapy for bulimia nervosa: time course and mechanisms of change. J Consult Clin
Psychol 2002; 70:267–274.
23. Nakash-Eisikovits O, Dierberger A, Westen D. A multidimensional metaanalysis of
pharmacotherapy for bulimia nervosa: summarizing the range of outcome in
controlled clinical trials. Harvard Rev Psychiatry 2002; 10:193–211.
24. Zhu AJ, Walsh BT. Pharmacologic treatment of eating disorders. Can J Psychiatry
2002; 47:227–234.
25. Lock J, Le Grange D, Agras WS, Dare C. Treatment Manual for Anorexia Nervosa:
A Family-Based Approach. New York: Guilford Press, 2001.
26. Bezold H, MacDowell M, Kunkel R. Predicting psychiatric length of stay. Admin
Policy Men Health 1996; 23:407–423.
27. Wiseman CV, Sunday SR, Klapper F, Harris WA, Halmi KA. Changing patterns of
hospitalization in eating disorder patients. Int J Eating Disord 2001; 30:69–74.
28. Howard WT, Evan KK, Quintero-Howard CV, Bowers WA, Andersen AE.
Predictors of success or failure of transition to day hospital treatment for inpatients
with anorexia nervosa. Am J Psychiatry 1999; 156:1697–1702.
29. Anzai N, Lindsey-Dudley K, Bidwell RJ. Inpatient and partial hospital treatment
for adolescent eating disorders. Child Adol Psychiatr Clin North Am 2002; 11:
279–309.
30. Zipfel S, Reas DL, Thorton C, Olmsted MP, Williamson DA, Gerlinghoff M,
Herzog W, Beumont PJ. Day hospitalization programs for eating disorders: a
systematic review of the literature. Int J Eating Disord 2002; 31:105–107.
31. Thorton C, Beumont P, Touyz S. The Australian experience of day programs for
patients with eating disorders. Int J Eating Disord 2002; 1–10.
32. Kaplan AS, Omlsted MP. Partial Hospitalization. In: Garner DM, Garfinkel PE,
eds. Handbook of Treatment for Eating Disorders. 2nd ed. New York: Guilford
Press, 1997:354–360.
33. American Psychiatric Association. Practice guideline for eating disorders (revised).
Am J Psychiatry 157(suppl):l-39.
34. Yager J. Implementing the Revised American Psychiatric Association Practice
Guidelines for the treatment of patients with eating disorders. Psychiatr Clin North
Am 2001; 24:185–199.
35. Metropolitan Life Insurance Company. Metropolitan height and weight tables.
NewYork, 1983.
36. Frisch RE, McArthur JW. Menstrual cycles: fatness as a determinant of minimum
weight for height necessary for their maintenance or onset. Science 1974; 185:
949–951.
37. Fairburn CG. Eating disorders. In: Clark DM, Fairburn CG, eds. Science and
Practice of Cognitive Behavior Therapy. New York: Oxford University Press,
1997:209–242.
38. Eckert ED, Mitchell JE. An overview of the treatment of anorexia nervosa.
Psychiatr Med 1989; 7:293–315.
39. Bowers WA. Cognitive therapy for eating disorders. In: Wright JH, Thase ME,
Beck AT, Ludgate JW, eds. Cognitive Therapy with Inpatients: Developing a
Cognitive Therapy Milieu. New York: Guilford Press, 1993.
40. Garfinkel PE, Garner DM. Anorexia Nervosa: A Multidimensional Perspective.
New York: Brunner/Mazel, 1982.
41. Garner DM. Individual psychotherapy for anorexia nervosa. J Psychiatr Res 1985;
19:423–433.
42. Beck JS. Cognitive Therapy: Basics and Beyond. New York: Guilford Press, 1995.
43. Freeman A. A psychosocial approach to conceptualizing schematic development for
cognitive therapy. In: Kuehlwein KT, Rosen H, eds. Cognitive Therapies in
Action. New York: Jossey-Bass, 1993:54–87.
376 BOWERS ET AL.
44. Garner DM, Vitousek KM, Pike KM. Cognitive-behavioral therapy for Anorexia
Nervosa. In: Garner DM, Garfinkel PE, eds. Handbook of Psychotherapy for
Anorexia Nervosa and Bulimia. New York: Guilford Press, 1997:94–144.
45. Bowers WA. Cognitive model of eating disorders. J Cognitive Psychother An Int
Quart2001; 15:331–340.
46. Wright JH, Thase ME, Beck AT, Ludgate JW. Cognitive Therapy with Inpatients:
Developing a Cognitive Therapy Milieu. New York: Guilford Press, 1993.
47. Beck AT, Rush J, Shaw BF, Emery G. Cognitive Therapy of Depression. New
York: Guilford Press, 1979.
48. Bowers WA. Basic principles for applying cognitive-behavioral therapy to anorexia
nervosa. Psychiatr Clin North Am 2001; 24:293–304.
49. Wilson GT, Fairburn CC, Agras WS. Cognitive-behavioral therapy for
bulimia nervosa. In: Garner DM, Garfinkel PE, eds. Handbook of Treatment for
Eating Disorders. 2d ed. New York: Guilford Press, 1997:67–93.
50. Fairburn CG, Cooper PJ. The Eating Disorders Examination. 12th ed. In: Fairburn
51. Garner DM, Olmsted MP. Manual for the Eating Attitudes Test (EDI). Odessa,
FL: Psychological Assessment Resources, Inc., 1993.
52. Garner DM, Garfinkel PE. The eating attitudes test. An index of the symptoms of
53. Garner DM, Garfinkel PE. Handbook of Treatment for Eating Disorders. 2d ed.
New York: Guilford Press, 1997:145–177.
54. Garner DM. Psychoeducational principles in treatment. In: Garner DM, Garfinkel
PE, eds. Handbook of Treatment for Eating Disorders. 2d ed. New York: Guilford
Press, 1997:145–177.
55. Bowers WA. Eating disorders. In: White JR, Freeman AS, eds.
CognitiveBehavioral group therapy for specific problems and populations.
Washington, DC: American Psychological Association, 2000:127–148.
56. Yalom I. The Theory and Practice of Group Psychotherapy. 4th ed. New York: Basic
Books, 1995.
57. Dattilio FM. Families in crisis. In: Dattilio FM, Freeman A, eds.
CognitiveBehavioral Strategies in Crisis Intervention. New York: Guilford Press,
1994: 2789–3010.
58. Dare C, Eisler I. Family therapy for anorexia nervosa. In: Garner DM, Garfinkel
PE, eds. Handbook of Treatment for Eating Disorders. 2d ed. New York: Guilford
Press, 1997:307–326.
59. Strober M, Freeman R, Morrell W. The long-term course of severe anorexia
nervosa in adolescents: survival analysis of recovery, relapse, and outcome
predictors over 10–15 years in a prospective study. Int J Eating Disord 1997; 22:
339–360.
60. Draper H. Anorexia nervosa and respecting a refusal of life-prolonging therapy:
limited justification. Bioethics 2000; 14:120–133.
61. Ramsey R, Ward A, Treasure J, Russell GF. Compulsory treatment in anorexia

nervosa. Short-term benefits and long-term mortality. Br J Psychiatry 1999; 175:
147–153.
62. Russell GFM. Involuntary treatment in anorexia nervosa. Psychiat Clin North Am
2001; 24:337–350.
63. Appelbaum PS, Rumpf T. Civil commitment of the anorexic patient. Gen Hosp
Psychiatry 1998; 20:225–230.
64. Watson TL, Bowers WA, Andersen AE. Involuntary treatment of eating disorders.
Am J Psychiatry 2000; 157:1806–1810.
378 BOWERS ET AL.
16
Nutrition Counseling for Anorexia Nervosa,
Jillian K.Croll
Eating Disorders Institute
St. Louis Park, Minnesota, U.S.A.
Dianne Neumark-Sztainer
University of Minnesota
Minneapolis, Minnesota, U.S.A.
Comprehensive care of patients with eating disorders involves a

multidisciplinary treatment team with nutritional rehabilitation and stabilization
as a significant cornerstone of treatment (1,2). Nutritional rehabilitation and
counseling requires a skilled registered dietitian versed in eating disorder
treatment approaches. The guiding principle in nutritional treatment of eating
disorders is to help the patient become reacquainted with normal eating. Eating
disorders rob sufferers of the practice of normal eating and frequently mask the
body’s natural hunger and satiety cues. This theft makes relearning how to honor
the body’s hunger and satiety signals, how to discern physiological urges to eat
from emotional triggers, and how to view food and eating as a joyful, nourishing
part of life rather than a frightening, threatening enemy, or a difficult and
complex journey. Nutritional counseling involves guiding the patient through this
process and helping him or her reestablish a healthy relationship with food.
NUTRITIONAL COUNSELING APPROACHES

The first step in nutritional management of an eating disorder, regardless of
diagnosis, involves a thorough nutritional assessment (see Fig. 1 for a
description of typical components of the initial eating disorder nutrition
assessment). Information gathered from this assessment, along with pertinent
medical, psychological, and psychiatric data, is used to determine appropriate
level of treatment (i.e., inpatient, day, outpatient, residential, etc.) and to
develop a nutritional treatment plan highlighting areas of focus for nutrition
education, behavior modification plans, and expectations for weight restoration
or stabilization, as appropriate. Nutritional approaches can vary across
treatment settings and diagnoses, but typically involve nutritional rehabilitation,
380 CROLL AND NEUMARK-SZTAINER
FIGURE 1 Components of an Initial Eating Disorders Nutritional Assessment.

Gather information from patient and family regarding:
• Detailed weight history, including high and low weight and ages at which these
occurred
• Change in eating patterns with onset of the eating disorder
• Current typical eating behaviors
• 24 hour recall, either of previous day or typical day
• Current nutritional intake
• Past and present symptom use
• Food and eating “rules” dictated by the eating disorder
• Supplement Use
• Assessment of motivation to change eating behaviors
• Exercise patterns
• Treatment history
• Familial factors (familial dieting, comments or teasing about weight, use of diet
products, etc.)
Review pertinent data:
• Growth charts (for children and adolescents)
• Electrolyte and hydration status, if available
• Weight, height, and vital signs
• Psychological and medical evaluation, if available
development of meal planning skills, normalization of eating patterns, and

participation in experiential activities that allow for opportunities to practice
new skills and stimulus control (2,3). Regular weight monitoring should be
done by the dietitian or another medical provider.
Anorexia Nervosa
Nutritional Assessment
The initial nutritional assessment of a person suffering from anorexia nervosa
should include a detailed history of the onset of the eating disorder, a thorough
recall of typical eating patterns and eating disorder symptom use, and questions
regarding potential precipitating events (e.g., weight-related teasing, bodily
changes associated with puberty, comments by family or friends, etc.). A
detailed description of a typical day should include times of eating occasions,
type and amount of food eaten, beverages consumed, and if and when bingeing
and purging occurs. Research comparing reported intake to observed intake has
NUTRITION COUNSELING 381
shown that diet history given by patients with anorexia nervosa provides a quite
reliable picture of nutritional intake (4). It is also important to assess the food
rules dictated by anorexia. Typical food-related rules commonly seen with
anorexia nervosa include the following: no eating after a certain time in the
evening, no eating of specific foods (such as meat, sweets, fats, snacks), eating
foods in a specific order, following certain caloric or fat gram daily limits, trying
to go all day without eating, and isolating self from others while eating. Foods
are often categorized as “good” or “bad” or “safe” or “scary,” and these
categorizations often determine eating behaviors (5). While these behaviors and
rules are largely the result of semistarvation, they can perpetuate the eating
disorder. Most tend to significantly decrease or abate with nutritional
rehabilitation (6–8). A detailed exercise history, including type of exercise,
duration, intensity, and compulsivity, is a necessary part of the evaluation to
determine energy expended on physical activity. As with eating patterns,
particular attention should be given to assessing exercise patterns in all settings:
at home, at school, at health clubs, as part of clubs or teams, or in other
settings. The physical activity level is often high in patients with anorexia
nervosa and contributes significantly to energy balance (9,10). Thorough
assessment is necessary, as anorexia imposes many rules and regulations.
Clinicians may need to ask many questions, as patients may not readily
volunteer information regarding eating disorder-related eating and exercise
patterns and may have altered perceptions of these behaviors. Family members
or friends, if present at the assessment, can often provide additional valuable
historical data regarding the patient’s eating habits and change in eating (e.g.,
onset of dieting; whether certain eating behaviors, like vegetarianism, predated
the eating disorder or co-occurred with the eating disorder; etc.).
Nutritional Rehabilitation
In addition to weight restoration, nutritional interventions seek to correct any
nutritional deficiencies resulting from the illness. Research has shown that
patients with anorexia nervosa may have abnormal vitamin status, specifically
indicators of low riboflavin and vitamin B6, deficiencies in essential fatty acids,
low serum albumin and other nutrients (11,12). Osteopenia is common among
patients with anorexia nervosa (13–15), making dietary calcium intake a critical
part of nutritional rehabilitation. Research suggests that weight restoration in
conjunction with adequate dietary calcium and vitamin D is associated with
improvement in bone mineralization (16,17).
Adequate nutrition rehabilitation can be a significant challenge. Many
patients experience and are frustrated by delayed gastric emptying resulting
from ongoing food restriction and complain of feeling excessively full in the
early stages of treatment (18). Delayed gastric emptying and nutrient

deficiencies typically resolve with ongoing treatment and normalization of
dietary intake. A low-dose multivitamin, multimineral supplement is viewed as
a lowrisk component of nutritional rehabilitation (3). For patients who have had
very minimal intake prior to initiation of treatment and need to restore
considerable weight, edema frequently develops during the nutritional
rehabilitation and refeeding process (19). The rapid weight gain associated with
edema can be quite anxiety producing for patients, and they need ongoing
reassurance that the rapid gain is primarily due to fluid accumulation and will
resolve with continued treatment. Close medical monitoring, particularly in the
beginning of inpatient treatment, is necessary to reduce the risk of refeeding
syndrome, characterized by a sudden precipitous drop in phosphorus,
magnesium, and potassium, and cardiac arrhythmias, particularly a prolonged QT
interval (20–22). Slow progression of caloric intake and careful monitoring
helps to prevent refeeding syndrome.
Weight Restoration Regimens

Weight restoration plans differ based on treatment setting and degree of weight
restoration needed. In an inpatient setting, given the typically low caloric intake
of patients prior to treatment, patients are generally started at a relatively low
caloric level, approximately 1200–1500 kcal/day. Kilocalories are then
increased incrementally by 200–300 per day over the course of 1–2 weeks to a
caloric level that supports weight restoration of approximately 2–3 pounds
(0.9–1.4 kg) per week (1,2). Similar expectations are common among
residential treatment programs. In a day treatment setting, patients are typically
started at a higher caloric level, approximately 1800 kcal/day, and increased
incrementally 200–300 kcal/day to reach a level that supports adequate weight
restoration. Weight restoration expectations are slightly lower in a day
treatment setting, typically 1.5–2.0 pounds (0.68–0.90 kg) per week, since
patients typically eat only one to two meals and one to two snacks during
treatment hours and consume the rest of their meal plan outside of treatment.
Because the majority of meals and snacks are eaten outside of treatment in an
outpatient setting, caloric levels vary, depending on the severity of the
weight loss and the progress the patient is able to make. Most outpatient
settings have weight restoration guidelines of 1.0–2.0 pounds (0.45–0.90 kg)
per week (1–3).
Energy needs increase as patients restore weight, due to both the restored
body weight and increased metabolic rate (23–26), and become quite high (up
to 3000–5000 kcal/day). Projected energy requirements for weight gain can be
determined by adding estimated resting energy expenditure, requirements for
FIGURE 2 Calculation of energy requirements. (From Harris JA, Benedict FG. (1919)
Biometric studies of basal metabolism in man. Carnegie Institute of Washington:
Publication No. 279.)
activity, and kilocalories needed for weight restoration (see Fig. 2 for details
regarding these calculations). It is important to note that this provides only an
estimate, and patients will respond differently at different times in the process,
as resting energy expenditure has been shown to increase and peak during
refeeding (23–26). Kilocalorie levels may need frequent adjustment until the
desired rate of weight restoration is achieved and weight restoration is
adequate (1–3).
At high caloric levels, consuming adequate food may present a challenge to
patients. Calorically dense foods, more frequent meals and snacks, or oral
liquid supplements can be utilized to meet the high-energy needs of these patients.
Some inpatient, partial, and residential programs use nasogastric tube feedings
or total parenteral nutrition (TPN) to increase caloric consumption, but both
require close medical monitoring, may increase the risk of refeeding syndrome,
and may be unnecessarily overwhelming for some patients (1–3). Use of food
rather than supplemental feedings or nutritional support approaches for weight

restoration more effectively supports longterm recovery (2). Once patients
restore weight and are able to remain in the clinician-established goal weight
range, caloric intake may be gradually decreased to a maintenance range,
typically 1800–2800 kcal/day (24).
Though it varies greatly by treatment program, in the early stages of
inpatient and residential treatment patients typically have little choice in meal
and snack foods and are asked to view the food they are being asked to eat as
medicine. As patients are able to finish meals without refusing to eat or
requiring any liquid supplements to replace uneaten foods, they progress to
selecting meals and snacks based on their individualized meal plan. As patients
progress in treatment, they gain independence with the meal plan and can
participate in decisions regarding necessary meal plan changes. In day treatment
and outpatient settings, patients have more flexibility in food choices, but
initially may need to view food in the same way, as a medicinal type of treatment.
In order to create an environment conducive to the changing of eating disorder
behaviors, a number of rules must be enforced at mealtimes. Patients are
typically allowed 30–60 min per meal and 15 min per snack; gum and mints,
often used by patients to assuage denied hunger, are not allowed; caffeine intake
is limited; and food-related behaviors, such as excessive cutting of food, taking
tiny bites, hiding food, scraping off fat, and excessive use of condiments, are
monitored and redirected. Across all settings, an individualized meal plan is
created for each patient based on caloric and weight restoration needs, eating
patterns, schedule, and special dietary needs (e.g., diabetes, lactose
intolerance).
Meal Plans
The individualized meal plan serves many purposes. On a broad level, it can
serve as the client’s personalized road map to successful, normal eating. On a
more detailed level, it provides the patient with a concrete presentation of what
she or he needs to eat on a daily basis to progress in treatment. The meal plan is
typically created with the client, taking into account true food likes and dislikes
and the current prescribed caloric level. The meal plan encourages patients to
challenge the eating disorder and themselves to reintroduce foods forbidden by
the eating disorder. Meal plans, while calculated based on caloric needs, are
usually presented in terms of servings from each food group, similar to the food
guide pyramid recommendations (27), rather than calories. See Table 1 for
examples of typical meal plans at different caloric levels and Table 2 for
recommended portion sizes of foods in each food group.
TABLE 1 Sample Meal Plans Across Increasing Kilocalorie Levels
a Number of servings per groups varies; based on individualized meal plan created for
patient.
b Liquid supplements may be used at higher calorie levels, if desired, to meet caloric
needs. If liquid supplements are used, the number of grain and dessert servings may be
reduced accordingly to meet required caloric level.
Detailed explanations of portion sizes, with food models, real food, or

pictures, can help relieve some anxiety about forbidden foods and provide
invaluable nutrition education for clients. Clear descriptions of portion sizes for
each food group and typical examples of foods in each food group, with
examples relevant to the client’s culture, schedule, and habits, all help patients
to challenge the eating disorder. Explanation of overall nutritional needs and
discussion of specific nutrient needs and function help patients to understand the
rationale behind the meal plan. The meal plan should include all the food
groups, particularly challenging foods or food groups such as fats and desserts.
Anorexia takes away a person’s freedom with food, and the meal plan can be a
powerful tool for allowing a person to reclaim the freedom of normal eating.
If patients are not able to eat challenging foods in the safe environment of
treatment or with the support of their treatment team, it is unlikely that they
will readily reincorporate those foods outside of treatment or when they are no
longer engaged in treatment. Staff-supported experiential nutrition activities,
such as eating in different types of restaurants, at food courts in shopping malls,
and in other social settings, are often part of group treatment programs.
Practice following the meal plan in a variety of settings in the context of
treatment programming can help patients gain the necessary skills to navigate
challenging and difficult situations they will likely face outside of the treatment
setting. Given the exaggerated societal messages regarding limiting fat, sweets,
TABLE 2 Sample Portion Sizes for Each Food Group
Source: Adapted from Food Guide Pyramid http://www.nal.usda.gov/fnic/Epyr/

pyramid.html) and Methodist Hospital Eating Disorders Institute meal planning
materials.
sugar, and other foods, patients face a difficult task in overcoming those
messages and incorporating challenging foods in their meal plan. Education
regarding the concepts of normal eating is critical, including learning to eat
when hungry, eating until full, including a wide variety of foods, not avoiding
or fearing specific foods, eating with others, eating in a variety of settings,
eating from all the food groups, not being obsessed with thoughts of food, not
purging or exercising to counteract food eaten, etc. Normal eating includes
having all kinds of foods, not categorizing foods as positive or negative, and
incorporating variety and moderation as mainstays (28).
Ongoing Treatment
While many patients with anorexia nervosa experience inpatient or partial
hospitalization treatment, most of the time spent in treatment is in an
outpatient setting with a multidisciplinary team. Regular follow-up with the
dietitian is necessary and generally occurs on a weekly basis, typically until
weight restoration is complete and has been maintained for at least 2 weeks.
Once the patient is within their goal weight range, visits with the dietitian may
decrease in frequency to bimonthly or monthly. Visits often include taking the
patient’s weight. Weights should be taken with the patient wearing only a gown
and undergarments, preferably with them facing away from the numbers on the
scale. Weight changes are used as a guideline for meal plan changes and
strategies and are an integral piece of medical monitoring. Because of the
intense focus on weight associated with anorexia nervosa, discussions with the
patient regarding current weight, weight change, and goal weights should focus
on progress and strategies for change rather than be overly focused on numbers.
As adequate weight restoration is achieved, energy needs decrease and the meal
plan can be changed accordingly. Weight maintenance energy levels vary by
individual based on age, gender, physical activity, and weight, but are typically
in the range of 2000–2800 kcal/day. Figure 2 provides an equation for
calculating weight restoration and weight maintenance needs.
Nutritional care in an outpatient setting incorporates a number of techniques
to aid patients in the process of incorporating variety into their meal plan and
regaining foods lost to the eating disorder. Usually clients will be asked to eat at
structured, regular intervals and participate in selfmonitoring, cognitive-
behavioral therapy techniques (29,30) involving planning meals and snacks
ahead of time, eating at set times, and keeping daily food records. Eating
specific amounts and types of foods at appropriate set times throughout the day,
even if not hungry, allows for relearning of hunger and satiety cues and normal
eating patterns.
Food records include information about food and beverage intake, including
time of meal or snack, location, whether alone or with others, what and how
much was eaten, rating of hunger and satiety levels, as well as any thoughts or
feelings associated with eating. If patients also struggle with bingeing and/or
purging, urges or occurrences of the behaviors can be tracked on these forms as
well. Review of the patient’s food records provides an opportunity to discuss
progress made with the meal plan and challenges faced with specific foods or
situations. Self-monitoring can be challenging for some patients for a variety of
reasons. It may be difficult for patients to write down what they are eating. It may
seem overwhelming and anxiety provoking to see it all down on paper, or it
may be difficult to record symptoms. For others, it may be another avenue for
the perfectionism of anorexia to take hold, driving them to have perfectly
written and detailed food records each week. Initially, patients often struggle
with rating hunger and satiety levels because the eating disorder has resulted in
loss of connection between hunger signals and eating behaviors. Reconnection
to innate hunger and satiety signs may take considerable time. Clinicians should
encourage patients to consider food records as another tool in the fight against
anorexia and a way for them to remind themselves of their meal plan and their
goals, rather than something to obsess about or feel like they have to do
perfectly (see Fig. 3 for an example of a comprehensive food record form).
Family Involvement
With adolescent patients, parents need to be included in initial education
sessions regarding the meal plan, portion sizes, expectations for the patient
while in treatment, and ways family members can support the patient. While
parents are often not present during ongoing follow-up sessions, they generally
need to be regularly apprised of weight changes, progress with the meal plan,
and strategies or goals for the week, as well as given the opportunity to ask
questions regarding the meal plan. Parents also can provide insight regarding the
patient’s eating behaviors, successes, and difficulties at home that patients may
not fully reveal to the clinician. For younger patients, there is evidence to
suggest that the patient’s family can be integrally involved in the
reestablishment of normal eating patterns (31). This approach aids parents in
supporting the recovery of their child by giving them primary jurisdiction over
implementation of the meal plan and the patient’s eating. Such approaches must
be carefully implemented so as not to set up a power struggle between the
patient and the parents, but rather should help parents to support their child in a
constructive and productive manner. In this type of approach, parents would be
more involved in ongoing nutrition sessions.
Regardless of their level of involvement in the meal plan, families often need
assistance in knowing what to say to help patients have success with their meal
plan. No one approach works well with all families, and families should be
encouraged to ask the patient what they should and should not do or say to be
supportive when it seems like the patient is struggling with the eating disorder
during a meal or snack. To be most effective and productive, these conversations
should take place outside of meal and snack times.
FIGURE 3 Example of a Food Record Form.
Overall, as part of a multifaceted approach to treatment of the whole person,

carefully monitored nutritional rehabilitation is critical to recovery. Ongoing
nutritional counseling helps patients with anorexia nervosa determine
reasonable, achievable goals and to celebrate the successful accomplishment of
those goals over time, and provides strategies for overcoming obstacles along
the way.
Bulimia Nervosa
The initial nutritional assessment of a person suffering from bulimia nervosa is
similar to that of anorexia nervosa (see Fig. 1 for typical assessment questions).
A detailed description of a typical day should include all eating occasions,
including binge and purge occurrences as well as nonbinge meals and snacks. A
careful exploration of type and amount of food eaten and beverages consumed
through the day can reveal an accurate picture of foods commonly eaten during
a binge, triggers for binge and purge episodes, and those foods the patient is
comfortable eating without feeling the need to purge. Typically, foods that are
high in fat or sugar, such as fast food, bakery items, salty snack foods, candy,
and ice cream, are foods considered “bad foods” or binge foods and are avoided
unless purging is possible or planned (32). When purging is not planned or
possible, patients with bulimia nervosa will usually restrict food intake or limit
intake to foods considered to be healthy, such as fruit, vegetables, and other low-
fat foods. Research has shown that when not engaging in binge eating, persons
with bulimia nervosa exhibit significant dietary restraint, eat fewer meals and
snacks, and take in significantly less energy and fat than controls (33,34). These
findings, along with studies examining energy retained from binge eating after
purging [approximately 1200 kcal/binge (35,36)], explain why many patients
with bulimia nervosa have a relatively normal body weight despite characteristic
restrictive eating alternating with binge eating and purging.
Nutritional Rehabilitation
Nutritional rehabilitation focuses on helping patients set goals around reducing
binge-purge and restricting behaviors, increasing variety in food choices,
practicing eating foods that are considered binge foods without purging,
planning approaches for eating in challenging situations (e.g., restaurants,
special occasions), and encouraging healthy, moderate exercise (1–3). Because
individuals with bulimia nervosa typically maintain a normal or near-normal
weight, weight restoration is not typically the focus of nutritional treatment.

However, many patients desire a body weight that, while within a normal
range, is unrealistically low for them given their genetic weight predisposition or
set range (1,3). These patients need nutritional and therapeutic interventions
and education to be able to accept a body shape and weight that may be
different from than they desire, but ultimately where they are able to be
emotionally and cognitively healthy.
The main focus of nutritional intervention and education for patients with
bulimia nervosa is to establish a pattern of normal eating, with three meals and
one to three snacks per day, in an effort to prevent the restrictbinge-purge cycle
and allow the patient to become reacquainted with their hunger and satiety
signals. Relearning normal eating patterns can be supported via a meal plan that
provides a guide for the recommended meals and snacks and food records for
self-monitoring. Meal plan structure is similar to those used with patients with
anorexia nervosa, providing recommended servings from all food groups based
on nutritional needs (refer to Tables 1 and 2 for details). Energy needs are
dependent on current weight status, physical activity level, age, and gender.
Typically, energy needs for people with bulimia nervosa are lower than those
for people with anorexia nervosa needing to restore weight and as such remain
more stable across treatment (3). While energy needs vary by individual,
weight maintenance meal plans are typically in the range of 1800–2600
kilocalories (refer to Fig. 2 for energy requirement calculation equations).
Self-monitoring is a key element in the nutritional management of bulimia
nervosa. Food records are also similar to those used with patients with anorexia
nervosa, but often have a stronger emphasis on hunger and satiety levels, which
can be useful in helping patients identify how periods of restriction result in
binge episodes. In addition to periodic hunger and satiety ratings, food records
should include food and beverage intake, time of meal or snack, location,
whether alone or with others, what and how much was eaten, binge and purge
urges or episodes, as well as any thoughts or feelings associated with eating.
Patients may find it difficult and shameful to record binge and purge episodes or
the amount of food eaten during binges, but regular documentation can help
identify triggers and reduce binges (36). Patients may need to be encouraged to
view food records and self-monitoring as objective data-gathering tools from
which they can draw conclusions and identify solutions. Patients usually find
that the food records help keep them on track with their meal plan and aid in
planning eating on a day-to-day basis and for special events or occasions. Food
records often provide direction to goal setting for the week. Weekly goals may
include reincorporating one or more challenging foods into the diet (e.g.,
sharing a candy bar with a friend); eating in a challenging situation such as a
restaurant, school cafeteria, or a friend’s house (e.g., going to a fast-food
restaurant); trying a difficult foodrelated task (e.g., grocery shopping); or

participating in physical activity (e.g., walking for 30 minutes and stretching for
30 minutes 3 days that week).
Cognitive-Behavioral and Nutritional Therapies

It is well established that cognitive-behavioral therapy (CBT) is an effective
intervention for the management of bulimia (29,37–41) and is a primary
psychotheraputic approach. Inherent to CBT are a number of nutritional
components, namely, nutrition education, meal planning, establishment of
regular eating patterns, and discouragement of dieting (2,39). Studies have
sought to examine the nutritional therapy components of CBT separately from
the cognitive therapy components (38–40,42) and have found the nutritional
counseling component to be an effective part of CBT. Hsu and colleagues support
an approach including both components of CBT in which there exists overlap
between the two approaches, but each component retains a primary distinctive
focus, either cognitive or nutritional, during work with the patients to reduce
symptom use and reestablish normal eating (39). Sundgot-Borgen and
colleagues also support the integration of physical activity protocols into the
treatment of these patients as a method of improving body image, promoting
physical fitness, and reducing symptom use (38). The dietitian can have a key
role in the nutritional components of CBT and physical activity
recommendations and monitoring.
Nutrition Education and Counseling

The primary goals of nutritional intervention and education are to encourage
development of normalized eating habits, improve nutrition knowledge, and
support the development of skills related to grocery shopping, meal
preparation, and situational stressors (e.g., restaurants, special occasions), so
that the patients can follow the meal plan and prevent symptom re-
occurrence (1,2,39,40). Education regarding principles of normal eating,
psychological and physiological effects of starvation, nutritional requirements,
and metabolism is critical (43). Nutrition education includes discussion of the
benefits of the patient’s self-imposed forbidden foods, such as fat or sweets, and
the nutritional reasons for consuming adequate quantities of less forbidden but
still challenging foods, such as dairy products for adequate calcium or meats for
adequate protein intake (2). Patients may benefit from understanding the
physiological effects of dieting, particularly the effects of diminished seroto nin
levels that occur with dieting (44). Restrictive food intake can result in
decreased plasma tryptophan, the precursor of serotonin, and has been
shown to be associated with alterations in mood, increased irritability,

depressive symptoms, and weight and shape concerns (44–46). Skipping
meals and restrictive eating can lead to increased use of binge-and-purge
behaviors (33,34). Oftentimes patients need substantial education regarding the
futility and riskiness of fad diets and need to be equipped with media literacy
skills to sift through the innumerable messages regarding nutrition that are so
readily available from the media. Once patients understand the concepts of
metabolism as well as the effects of food restriction, weight maintenance, and
normal eating, they are better equipped to focus on symptom reduction.
Patients may also need significant education regarding side effects of stopping
binge-purge behavior. Cessation of laxative use or purging can result in fluid
retention, weight shifts, and slow gastrointestinal function (2,3,47). Patients
need to be informed that these uncomfortable changes are transitory and will
resolve, but only if they continue to abstain from bingeing and purging.
Adequate fluid intake, adequate dietary fiber from whole grains, fruits, and
vegetables, and moderate exercise will aid in the reestablishment of body
homeostasis (2,3). Some patients need high levels of support to manage the
anxiety related to these side effects and may need an increased level of
programming, such as a day treatment program, in order to stop binge-purge
behaviors and effectively manage the side effects.
Ideas for Support of the Meal Plan

Patients typically have the most difficulty reintroducing forbidden foods into
their regular eating patterns, for fear of weight gain or body composition
changes, and need support and strategies to effectively reintegrate these
foods (2,3). To facilitate reintroduction of challenging foods, patients may find
that using single-serving food items or repackaging foods into smaller, portioned
containers can help prevent the escalation of a binge when eating a challenging
food (41).
Often patients find it easier to initially consume challenging foods in the
company of family or friends, where they can find comfort in the distraction of
social interaction and reassurance when others eat the same food. By asking
support people to eat with them and engage in an activity after the meal,
patients can receive support during the meal, as well as after the meal to help
prevent them from being overly focused on the food that was just eaten. For
patients who have experienced increased isolation or withdrawal from loved
ones as a result of the eating disorder, asking support people to eat with them
increases their social interaction and supports successful meal or snack
completion. Families can help their loved ones by supporting the meal plan
through planning, preparing, shopping, and eating with the patient, going to
a variety of restaurants and social situations with the patient and supporting the
patient’s efforts to follow the meal plan in these challenging situations.
Nutritional intervention often focuses on helping patients to manage
overwhelming situations that involve food by focusing on the component parts
of the situation and making a plan for success using the meal plan as a guide. For
example, if a patient has a great deal of difficulty going grocery shopping without
it resulting in a binge-purge episode, the clinician can help the patient break
down the task into manageable parts. First, together they can make a grocery
list and determine what area of the store each food is in. Next they can
determine what type and quantity of each food the patient needs to buy to meet
his or her meal plan. At this point, it is important to explore what challenges
may exist for particular foods and how those challenges can be overcome. For
example, the meal plan discourages use of fat-free food items, but the patient
feels compelled by the eating disorder to buy only fat-free cream cheese. So the
patient takes a friend shopping and asks the friend to put regular cream cheese in
the cart. Patients may need education regarding the nutritional benefits of
challenging foods to understand that such foods generally provide more
nutrients and satiety than fat-free or lowfat versions. After making the list and
determining what needs to be purchased and where it is located, the dietitian
may have the patient practice by going to the grocery store with the list, confirm
the location of all the items, but not buy anything on this occasion. It is probably
a good idea for the patient to take no money alone so as to help prevent a binge-
purge episode. After this practice run, the patient may feel more assured of
success because he or she is familiar with the location of items in the store. Then
the patient would be asked to go to the grocery store, with a support person if
necessary, to actually purchase the foods on the list in the specified quantities
and report back to the clinician regarding success and challenges.
In addition to these techniques used in an outpatient setting, experiential
nutritional activities are often part of partial day treatment or intensive
outpatient treatment. Such activities typically include supported exposure to
difficult foods or situations, such as meals out at restaurants, grocery shopping,
or meal preparation for guests as a group. These experiential activities are
commonly supervised by a dietitian and a psychologist or other clinician, and
include a process group cofacilitated by both clinicians for the most effective
utilization of the experience. Patients are often able to overcome challenging
foods or situations in these types of experiential activities and recognize that
they can complete food-related tasks that they thought were overwhelming.
Overall, as with anorexia nervosa, nutritional intervention and education is a
critical part of a multifaceted approach to management of bulimia nervosa and
seems most effective for patients when paired with cognitivebehavioral work
with a psychologist or other clinician. Ongoing nutrition counseling helps
patients with bulimia nervosa establish regular eating patterns and decrease
symptom use, develop skills for use when reintroducing challenging foods, and
have better understanding of nutritional needs and physiological effects of
symptom use.
Given that patients with binge eating disorder often seek treatment for obesity
and that nutritional counseling is generally a service incorporated into obesity
treatment, a nutrition professional may be the primary clinician discussing binge
eating with the patient. Binge eating disorder research is in the relatively early
stages and as such numerous assessment and treatment recommendations and
protocols are under evaluation, many of which have nutritional therapy or
nutrition education components. The initial nutritional assessment for binge
eating disorder should include a detailed account of the onset of binge eating, a
weight and dieting history, and any historical events the patient believes may be
related to the binge eating (e.g., weight-related teasing; comments by parents,
friends, significant other; etc.). In general, people experiencing binge eating are
more likely to have had a history of being overweight since childhood, have a
significant history of large weight fluctuations (20 pounds or more), and
experience an increase in binge eating as weight increases (48–51). Because
binge eating may develop during periods of dietary restriction, such as a very-
low-calorie diet (52), a careful history of diet attempts and outcomes should be
taken. Assessment should include a thorough recall of typical eating patterns and
binge eating occurrence and duration, along with an assessment of the out-of-
control nature of the reported binge eating episodes.
A detailed description of a typical day, including all eating occasions, can help
identify patterns in eating and triggers for binge eating. As with bulimia
nervosa, a careful exploration of type and amount of food eaten and beverages
consumed through the day can reveal an accurate picture of foods commonly
eaten during a binge, as well as those foods the patient eats outside of binge
episodes. Typically, foods that are high in fat or carbohydrates and lower in
protein, such as candy, baked goods, and ice cream, are foods included in binge
episodes (49,52). Caloric contribution from binge eating to total daily intake
varies greatly depending on the episode, ranging from a minor contribution of
fewer than 100 kilocalories to a large contribution of several thousand
kilocalories (49,53). Studies have shown an average caloric intake difference of
approximately 1000 kilocalories between nonbinge eating days and binge

days (54,55).
Nutrition Intervention and Education

Treatment for binge eating disorder to date has included CBT or interpersonal
therapy with a mental health professional, often with the addition of a focus on
weight regulation or loss with a nutrition professional. CBT treatment adapted
from bulimia nervosa, focusing on normalizing eating habits, identifying
triggers, and changing distorted attitudes regarding eating and weight in an
effort to reduce or eliminate binge eating behaviors, is followed by a focus on
realistic weight control. Research has shown that patients who are able to cease
binge eating behaviors lose weight (49,56). Ongoing research involves
examination of the most effective timing of the focus on weight control—either
concurrently with CBT treatment or after the patient is able to demonstrate
consistent, binge-free eating (36,57,58). When weight loss is discussed, it is
critical that patients establish reasonable weight loss goals, such as a more
modest 10–20% weight loss, rather than try to reach an “ideal weight.”
Moderate weight loss can result in improvements in blood pressure, lipid
profile, and blood glucose control (59). Even if cessation of binge eating results
in weight maintenance only, the typical ongoing weight gain associated with
binge eating is avoided (60). In addition, regular physical activity can contribute
to weight loss and at the same time help reduce binge eating. Research indicates
the addition of exercise to CBT treatment results in significantly greater weight
loss and abstinence from binge eating than CBT alone (61).
Nutritional counseling should include establishment of an individualized meal
plan that incorporates three meals and one or more structured snacks per day,
includes enjoyable foods, and encourages regular physical activity. The meal
plan, based on a nondieting approach to eating, should focus on food groups and
portions, not calories or calorie counting (62). (Refer to Tables 1 and 2 for
sample meal plans and portion size guides.) The primary focus of the meal plan
is on eating regular meals and snacks. While energy needs of patients with binge
eating disorder may vary significantly based on weight, age, gender, and activity
level, typical daily energy needs likely range from 1600 to 2200 kilocalories.
Eating regularly scheduled meals and snacks throughout the day helps prevent
physiological urges to binge resulting from overwhelming hunger (63). While
patients are not required to eat particular foods at each meal and snack, the
meal plan is designed to provide variety throughout the day and incorporate all
food groups. When hunger identification or management is difficult,
prepackaged single-serving food items or repackaging of foods into smaller,
portioned containers can help provide guidance regarding serving size and
prevent the escalation of a binge (63). The nondieting approach to weight

management emphasizes that all foods can be included and has a strong focus on
moderation. Rather than prescribing eating according to rigid diet rules, the
nondieting approach encourages variety, moderation, and acceptance. It
supports the establishment of normal eating and helps reduce binge eating,
depression, anxiety, and body dissatisfaction (62,64). This approach may be
challenging for a person who is used to implementing rigid diet rules in an effort
to lose weight and/or decrease binge eating, but is ultimately liberating for
patients.
Self-monitoring of eating habits via food records is an instrumental part of
nutritional treatment. Food records should include food and beverage intake,
time of meal or snack, location, whether alone or with others, what and how
much was eaten, binge eating episodes, as well as any thoughts or feelings
associated with eating (see Fig. 3 for a sample food record). It can be also be
beneficial for patients to document physical activity on food records in order to
track activity patterns in relation to eating habits and support behavior change
efforts. Regular review of food records with the patient can identify patterns of
eating and relationships between eating and situation, which provides direction
for goal setting. While it may be difficult and shameful for patients to record the
amount of food eaten during binges or their typical eating habits, the
process can support healthful food choices, help identify triggers, and reduce
binges (36). Attention must be given to the inundation of nutrition information
and fad diets through the media and education provided regarding the
misleading information associated with fad diets. Because enhanced self-
acceptance seems to support recovery from binge eating disorder (65), patients
must be equipped with appropriate media literacy skills to ward off conflicting
societal messages regarding food, weight, and appearance and maintain positive
attitudes about self-acceptance.
Patients also need nutrition education regarding appropriate portion sizes,
menu planning, grocery shopping, dining out, variety and moderation with food
choices to combat restrictive eating followed by binges, incorporating healthy
eating into a busy lifestyle, and making positive food choices in stressful or
triggering situations. Many patients have firmly held but unsubstantiated beliefs
regarding nutrition or food and need sound nutrition information to dispel these
myths. As patients learn more about normal eating and how to incorporate
realistic principles of healthy eating into their lifestyle, they are better able to
discern between sound nutrition information in the media and information with
no factual basis.
On the whole, nutritional management of binge eating disorder focuses on
cessation of binge eating and establishment of normal eating behaviors and
regular physical activity, and is an integral part the overall therapeutic treatment
for binge eating disorder. If necessary, focus on healthy, moderate weight loss
or lifelong weight control can be incorporated as patients are able to reduce or
cease binge eating behaviors.
SUMMARY OF NUTRITIONAL TREATMENT

GOALS FOR ANOREXIA NERVOSA, BULIMIA
NERVOSA, AND BINGE EATING DISORDER
Normalization of nutritional status and dietary habits is a primary goal in the
management of eating disorders. Nutritional rehabilitation, education, and
counseling are integral components of eating disorder treatment across all
treatment settings and should be conducted by qualified nutritional personnel,
typically a registered dietitian.
In summary, the main components and goals of nutritional treatment of
anorexia nervosa include the following:
Adequate weight restoration through increased energy intake via food rather than
enteral or parenteral nutrition products
Meal plan education and guidance focused on food groups, portion sizes, and
variety to promote normalization of eating patterns
Supported reintroduction of foods eliminated from the diet to dispel fears related
to these foods
Adequate nutrient intake to support nutritional recovery, possibly including a
low-dose multivitamin and multimineral supplement
Nutrition education regarding normal, healthy eating; and normalization of
dieting, restrictive eating, and excessive energy expenditure behaviors.
The guiding principles of nutritional management of bulimia nervosa include the

following:
Normalization of eating patterns through establishment of regular meals and

snacks
Adequate energy and nutrient intake to support weight maintenance or weight
restoration, as necessary
Meal plan education and guidance emphasizing the importance of avoiding long
periods without eating with a focus on hunger and satiety signals
Nutrition education regarding normal, healthy eating
Goal setting to achieve cessation of dieting or food restriction behaviors,
excessive exercise, and binge-purge behaviors
Supported reintroduction of binge foods into the diet via stimulus control
planning and experiential activities
Binge eating disorder nutritional treatment goals and components include the
following, with particular attention to strategies for managing eating in high-risk
or triggering situations and establishment of regular meals and snacks.
Normalization of eating patterns through establishment of regular meals and

snacks
Meal plan education and guidance emphasizing the importance of avoiding long
periods without eating, with a focus on hunger and satiety signals
Nutrition education regarding realistic healthy eating principles Goal setting to
achieve cessation of binge eating Supported reintroduction of binge foods
into the diet via stimulus control
When appropriate, introduction of a nondiet approach to weight control to
facilitate lifelong weight management
Nutritional components of eating disorder treatment help patients to

successfully overcome the eating disorder and maintain lifelong normal, healthy
eating habits, through education, planning, skill development, and practice.
REFERENCES
1. American Psychiatric Association. Practice guideline for the treatment of patients
with eating disorders. Am J Psychiatry 2000; 157(1): 1–39.
2. American Dietetic Association. Nutrition intervention in the treatment of anorexia
nervosa, bulimia nervosa, and eating disorder not otherwise specified (EDNOS).
J Am Diet Assoc 2001; 101(7):810–819.
3. Rock CL, Curran-Celentano J. Nutritional management of eating disorders.
Psychiat Clin North Am 1996; 19(4):701–713.
4. Hadigan CM, Anderson EJ, Miller KK, Hubbard JL, Herzog DB, Klibanski A,
Grinspoon SK. Assessment of macronutrient and micronutrient intake in women
with anorexia nervosa. Int J Eating Disord 2000; 28(3):284–292.
5. Sunday SR, Einhorn A, Halmi KA. Relationship of perceived macronutrient and
caloric content to affective cognitions about food in eating-disordered, restrained,
and unrestrained subjects. Am J Clin Nutr 1992; 55(2):362–371.
6. Garner DM. Pathogenesis of anorexia nervosa. Lancet 1993; 341:1631–1635.
8. Rock CL, Curran-Celentano J. Nutritional disorder of anorexia nervosa: a review.
Int J Eating Disord 1994; 15(2): 187–203.
9. Casper RC, Schoeller DA, Kushner R, Hnilicka J, Gold ST. Total daily
energy expenditure and activity level in anorexia nervosa. Am J Clin Nutr 1991;
53(5): 1143–1150.
10. Pirke KM, Trimborn P, Platte P, Fichter M. Average total energy expenditure in
anorexia nervosa, bulimia nervosa, and healthy young women. Biol Psychiatry
1991; 30(7):711–718.
11. Rock CL, Vasantharajan S. Vitamin status of eating disorder patients: relationship
to clinical indices and effect of treatment. Int J Eating Disord 1995; 18(3):257–262.
12. Holman RT, Adams CE, Nelson RA, Grater SJ, Jaskiewicz JA, Johnson SB,
Erdman JW Jr. Patients with anorexia nervosa demonstrate deficiencies of selected
essential fatty acids, compensatory changes in nonessential fatty acids and decreased
fluidity of plasma lipids. J Nutr 1995; 125(4):901–907.
13. Bachrach LK, Katzman DK, Litt IF, Guido D, Marcus R. Recovery from
osteopenia in adolescent girls with anorexia nervosa. J Clin Endocrinol Metab 1991;
72(3):602–606.
14. Rigotti NA, Neer RM, Skates SJ, Herzog DB, Nussbaum SR. The clinical course of
osteoporosis in anorexia nervosa. A longitudinal study of cortical bone mass. JAMA
1991; 265(9): 1133–1138.
15. Rigotti NA, Nussbaum SR, Herzog DB, Neer RM. Osteoporosis in women with
anorexia nervosa. N Engl J Med 1984; 311(25):1601–1606.
16. Abrams SA, Silber TJ, Esteban NV, Vieira NE, Stuff JE, Meyers R, Majd M,
Yergey AL. Mineral balance and bone turnover in adolescents with anorexia
nervosa. J Pediatrics 1993; 123(2):326–331.
17. Hay PJ, Delahunt JW, Hall A, Mitchell AW, Harper G, Salmond C. Predictors of
osteopenia in premenopausal women with anorexia nervosa. Calc Tissue Int 1992;
50(6):498–501.
18. Waldholtz BD, Andersen AE. Gastrointestinal symptoms in anorexia nervosa. A
prospective study. Gastroenterology 1990; 98(6):1415–1419.
19. Vaisman N, Corey M, Rossi MF, Goldberg E, Pencharz P. Changes in body
composition during refeeding of patients with anorexia nervosa. J Pediatrics 1988;
113(5):925–929.
20. Solomon SM, Kirby DF. The refeeding syndrome: a review. J Parent Ent Nutr
1990; 14:90–97.
21. Rock CL. Nutritional and medical assessment and management of eating disorders.
Nutr Clin Care 1999; 2:332–343.
22. Fisher M, Simpser E, Schneider M. Hypophosphatemia secondary to oral refeeding
in anorexia nervosa. Int J Eating Disord 2000; 28(2): 181–187.
23. Obarzanek E, Lesem MD, Jimerson DC. Resting metabolic rate of anorexia
nervosa patients during weight gain. Am J Clin Nutr 1994; 60(5):666- 675.
24. Krahn DD, Rock C, Dechert RE, Nairn KK, Hasse SA. Changes in resting energy
expenditure and body composition in anorexia nervosa patients during refeeding.
J Am Diet Assoc 1993; 93(4):434–438.
25. Salisbury JJ, Levine AS, Crow SJ, Mitchell JE. Refeeding, metabolic rate, and
weight gain in anorexia nervosa: a review. Int J Eating Disord 1995; 17(4):
337–345.
26. Platte P, Pirke KM, Trimborn P, Pietsch K, Krieg JC, Fichter MM. Resting
metabolic rate and total energy expenditure in acute and weight recovered patients
with anorexia nervosa and in healthy young women. Int J Eating Disord 1994; 16
(l):45–52.
27. Department of Health and Human Services and the United States Department of
Agriculture. The food guide pyramid. 1996. http://www.nal.usda.gov/fnic/Fpyr/
pyramid.html.
28. Satter E. How to Get Your Kid to Eat…But Not Too Much. Palo Alto, CA: Bull
Publishing, 1987.
29. Wilson GT. Cognitive behavior therapy for eating disorders: progress and
problems. Behav Res Ther 1991; 37(suppl 1):S79-S95.
30. Garner DM, Vitousek KM, Pike KM. Cognitive-behavioral therapy for anorexia
nervosa. In: Garner DMG ed. Handbook of Treatment of Eating Disorders. New
31. Lock J, LeGrange D, Agras WS, Dare C. Treatment Manual for Anorexia Nervosa:
A Family-Based Approach. New York: Guilford Press, 2001.
32. Hetherington MM, Altemus M, Nelson ML, Bernat AS, Gold PW. Eating behavior
in bulimia nervosa: multiple meal analyses. Am J Clin Nutr 1994; 60(6):864–873.
33. Hadigan CM, Kissileff HR, Walsh BT. Patterns of food selection during meals in
women with bulimia. Am J Clin Nutr 1989; 50(4):566–759.
34. Weltzin TE, Hsu LK, Pollice C, Kaye WH. Feeding patterns in bulimia nervosa.
Biol Psychiatry 1991; 30(11):1093–1110.
35. Kaye WH, Weltzin TE, Hsu LK, McConaha CW, Bolton B. Amount of calories
retained after binge eating and vomiting. Am J Psychiatry 1993; 150(6):969–971.
36. Goldfein JA, Devlin MJ, Spitzer RL. Cognitive behavioral therapy for the
treatment of binge eating disorder: what constitutes success? Am J Psychiatry
2000; 157(7): 1051–1056.
37. Miller WR, Rollnick S. Motivational Interviewing: Preparing People to Change
Addictive Behavior. New York: Guilford Press, 1991.
38. Sundgot-Borgen J, Rosenvinge JH, Bahr R, Schneider LS. The effect of exercise,
cognitive therapy, and nutritional counseling in treating bulimia nervosa. Med Sci
Sports Exercise 2002; 34(2): 190–195.
39. Hsu LK, Rand W, Sullivan S, Liu DW, Mulliken B, McDonagh B, Kaye WH.
Cognitive therapy, nutritional therapy and their combination in the treatment of
bulimia nervosa. Psychol Med 2001; 31(5):871–879.
40. Hsu LK, Holben B, West S. Nutritional counseling in bulimia nervosa. Int J Eating
Disord 1992; 11:55–62.
41. Fairburn CG, Agras WS, Wilson GT. The research on the treatment of bulimia
nervosa: Practical and theoretical implications. In: Anderson GH, Kennedy SH,
eds. The Biology of Feast and Famine: Relevance to Eating Disorders. San Diego:
Academic Press, 1992:318–340.
42. Hsu LK, Santhouse R, Chesler BE. Individual cognitive behavioral therapy for
bulimia nervosa: the description of a program. Int J Eating Disord 1991; 10:
273–283.
43. Story M. Nutrition management and dietary treatment of bulimia. J Am Diet Assoc
1986; 86:517.
44. Cowen PJ, Smith KA. Serotonin, dieting, and bulimia nervosa. Adv Exp Med Biol
1999; 467:101–104.
2000; 47(2):151–157.
47. Mitchell JE, Specker SM, DeZwann M. Comorbidity and medical complications of
bulimia nervosa. J Clin Psychiatry 1991; 52:13.
48. Telch CF, Agras WS, Rossiter EM. Binge eating increases with increasing
adiposity. Int J Eating Disord 1992; 7:115–119.
49. Marcus MD. Binge eating in obesity. In: Fairburn CG, Wilson GT, eds. Binge
Eating: Nature, Assessment, and Treatment. New York: Guilford Press, 1993:
77–96.
50. Wilfley DE, Grillo GM. Eating disorders: a women’s health problem in primary
care. Nurs Prac Forum 1993; 5:34–45.
51. Telch CF, Agras WS. The effects of a very low calorie diet on binge eating. Behav
Ther 1993; 24:177–193.
52. Yanovski SZ, Leet M, Yanovski JA, Flood M, Gold PW, Kissileff HR, Walsh BT.
Food selection and intake of obese women with binge-eating disorder. Am J Clin
Nutr 1992; 56(6):975–980.
53. Rossiter EM, Agras WS, Telch CF, Bmce B. The eating patterns of non-purging
bulimic subjects. Int J Eating Disord 1992; 11:111–120.
54. Timmerman GM. Caloric intake patterns of nonpurge binge-eating women. West J
Nurs Res 1000; 20:103–118.
55. Yanovski SZ, Sebring NG. Recorded food intake of obese women with binge eating
disorder before and after weight loss. Int J Eating Disord 1994; 15:135–150.
56. Wilson GT. Assessment of binge eating. In: Fairburn CG, Wilson GT, eds. Binge
Eating: Nature, Assessment, and Treatment. New York: Guilford Press, 1999:
3227–3249.
57. Grillo GM. The assessment and treatment of binge eating disorder. J Pract
Psychiatry Behav Health 1998; 4:191–200.
58. Williamson DA, Martin CK. Binge eating disorder: a review of the literature after
publication of the DSM-IV. Eating Weight Disord: Stud Anorexia. Bulimia, and
Obesity 1999; 4(3):103–114.
59. National Heart, Lung, and Blood Institute. Clinical guidelines on the identification,
evaluation, and treatment of overweight and obesity in adults—The evidence
report. Obesity Res 1998; 6(suppl 2):51S-209S.
60. Fairburn CG, Cooper Z, Doll HA, Norman P, O’Connor M. The natural course of
bulimia nervosa and binge eating disorder in young women. Arch Gen Psychiatry
2000; 57(7):659–665.
61. Pendleton VR, Goodrick GK, Poston WS, Reeves RS, Foreyt JP.
Exercise augments the effects of cognitive-behavioral therapy in the treatment of
binge eating. Int J Eating Disord 2002; 31(2): 172–184.
62. Goodrick GK, Poston WS II, Kimball KT, Reeves RS, Foreyt JP. Nondieting
versus dieting treatment for overweight binge-eating women. J Consult Clin
Psychol 1998;66(2):363–368.
63. Fariburn CG, Marcus MD, Wilson GT. Cognitive-behavioral treatment of binge
eating and bulimia nervosa. In: Fairburn CG, Wilson GT, eds. Binge Eating:
Nature, Assessment, and Treatment. New York: Guilford Press, 1993:361–404.
64. Tanco S, Linden W, Earle T. Well-being and morbid obesity in women: a
controlled therapy evaluation. Int J Eating Disord 1998; 23(3):325–339.
65. Devlin MJ. Binge-eating disorder and obesity. A combined treatment approach.
Psychiatr Clin North Am 2001; 24(2):325–335.
17
An Overview of Cognitive-Behavioral
Approaches to Eating Disorders
Stephen A.Wonderlich, James E.Mitchell, and Lorraine
Swan-Kremier
University of North Dakota School of Medicine and Health
Sciences and the Neuropsychiatric Research Institute
Fargo, North Dakota, U.S.A.
Carol B.Peterson and Scott J.Crow
University of Minnesota
Minneapolis, Minneapolis, U.S.A.
Cognitive-behavioral therapy (CBT) is the most thoroughly studied form of

treatment for the eating disorders. Consistent with other behavior therapy
interventions, CBT generally avoids the modification of underlying
psychological conflicts in favor of the direct change of cognitions and behavior
that comprise the eating disorders. Furthermore, CBT attempts to modify those
cognitive and behavioral processes, which are thought to maintain the disorder,
or be proximal causal variables, while not addressing more distal factors that
may have causal significance (e.g., early relationships). Like the CBT techniques
used to manage mood and anxiety disorders (1,2), CBT for bulimia nervosa is
generally a short-term treatment in which the clinician is encouraged to actively
focus and direct the process of change.
Most of the empirical studies of CBT for eating disorders have examined the
efficacy of this technique in the management of bulimia nervosa (BN), although
its extrapolation to binge eating disorder (BED) has gained increased attention
in recent years. While descriptions of treatments for anorexia nervosa (AN)
often include key elements of CBT, controlled treatment trials for this disorder
have been limited.
COGNITIVE BEHAVIORAL THERAPY IN THE

MANAGEMENT OF BULIMIA NERVOSA
CBT for BN is designed to reduce the frequency of binge eating and purging
behaviors while simultaneously reducing restrictive eating patterns (3). CBT for
406 WONDERLICH ET AL.
BN generally targets the modification of several key bulimic behaviors. First,

there is an explicit emphasis on decreasing restrictive dieting and increasing the
consumption of calories through eating meals and snacks. Second, dysfunctional
thoughts about shape and weight are directly confronted and alternative thought
patterns encouraged. Third, bulimic individuals are encouraged to identify the
triggers or precipitants of bulimic behavior and develop alternatives to high-risk
binge eating situations. Finally, through the moderation of extreme thought
processes and enhancement of behavioral control, CBT is posited to indirectly
enhance self-esteem. These goals are pursued in the context of a collaborative
treatment relationship through the implementation of several clinical
techniques (4,5): (a) daily self-monitoring of food consumption patterns,
maladaptive cognitions, and binge/purge behaviors; (b) learning to identify the
presence of problematic thoughts and questioning the logic associated with such
thoughts; (c) the development of active and systematic approaches to identifying
problems in living, constructing and testing solutions to these problems; and
(d) identifying healthy lifestyle patterns that facilitate the prevention of relapse
of bulimic symptoms.

TREATMENT OF BULIMIA NERVOSA
Table 1 delineates more than 20 controlled trials of CBT for BN (7,35). As can
be seen, whether provided in an individual or group therapy format, CBT has
been shown to be superior to minimal or waiting-list control intervention in
terms of reducing the frequency of binge eating and purging behaviors. These
findings have been discussed in a number of detailed reviews of studies of
psychotherapy outcomes in BN (3,6), but it is important to note that a
significant fraction of bulimic individuals remain symptomatic at the end of
these short-term structured treatments or relapse after completing treatment.
Furthermore, Table 1 reveals that CBT has generally been superior to other
interventions including behavior therapy, psychodynamically oriented
psychotherapy, supportive psychotherapy, and pharmacotherapy. Interestingly,
however, studies comparing CBT to interpersonal therapy (IPT), an approach
originally developed for the management of depression with no explicit focus on
bulimic symptoms, have revealed essentially no difference in efficacy at long-
term follow-up (21,31). These findings have resulted in debate about what
factors are most significant for the etiology and maintenance of BN, as well as
the most appropriate treatment for BN (36). However, the wealth of evidence
documenting the efficacy and rapid action of CBT for BN leads most to continue
to consider CBT as the treatment of choice for this condition (37).
COGNITIVE-BEHAVIORAL APPROACHES 407

TREATMENT OF BINGE EATING DISORDER
Many of the techniques that have been used in the management of BN have been
adapted and slightly modified for the management of BED. Generally speaking,
CBT has been effective in reducing the frequency of binge eating episodes in
subjects with BED (38,39). Interestingly, and similar to the BN treatment
literature, interpersonal therapy (IPT) has been shown to be as effective as CBT
in binge reduction in subjects with BED (39). Furthermore, follow-up
assessments of BED subjects with both types of treatment reveal slight increases
in binge eating over time, but a general persistence of the treatment effect (40).
A common finding in the management of BED is that currently available
psychological treatments are reasonably effective in reducing binge frequency
but seem to have little effect on short-term weight loss. This is a significant
finding for the participants, who oftentimes are overweight or obese and desire
a more potent effect for their weight. A recent large-scale follow-up study
suggested that treated BED individuals lost approximately 5% or more of their
body weight and furthermore that those who were abstinent from binge eating
were most likely to maintain these losses (40). This is significant in that BED
subjects often follow a trajectory of continued weight gain over time, and the
ability to preempt such weight gain and even obtain small degrees of weight loss
may be clinically significant. Furthermore, even a 5% weight loss for this
population may confer significant health benefits and should not be considered
trivial.

MANAGEMENT OF ANOREXIA NERVOSA
CBT for AN shares numerous features with CBT for BN. For example, CBT for
both conditions emphasizes psychoeducation about medical and dietary factors,
cognitive restructuring approaches, and behavioral interventions surrounding
increased food consumption through meals and snacks (41). However, there are
also differences between CBT for AN and BN. Some of the differences may be
attributable to symptomatic differences between AN and BN, particularly the
need for weight gain in AN, as well as the greater degree of symptom
egosyntonicity in AN. Furthermore, the motivation for treatment in AN may be
different than in BN, which can influence the style and timing of the
interventions (42).
TABLE 1 Controlled Trials for Cognitive-Behavioral Therapy for Bulimia Nervosa
TABLE 1 Continued
CBT, cognitive behavioral therapy; BT, behavioral therapy; ERP, exposure and response preventions; ss, single site; ms, multiple sites; IPT,
interpersonal psychotherapy; SPT, supportive therapy; Imip, imipramine; Desip, desipramine; NT, nutritional counseling.
There is limited evidence supporting the efficacy of CBT in treating patients

with AN. One study found CBT to be as effective as behavior therapy (43), and
a more recent investigation suggested that CBT was useful in preventing relapse
among weight-recovered anorexic individuals (44). However, there is a clear
need for a rigorous empirical trial of CBT for AN. Difficulties associated with
all treatment trials for AN (e.g., low base rate disorder, marked treatment
resistance, high dropout rates) make such a study difficult. Large multicenter
research strategies are likely to be needed to successfully complete such a study.
Currently, CBT for AN is best conceptualized as one component of a
multifaceted treatment plan that includes other modalities, such as family
therapy, drug therapy, nutritional rehabilitation, and the use of hospital-based
treatments. Such a multidisciplinary approach is consistent with recent practice
guidelines for AN (45).
GUIDELINES FOR COGNITIVE BEHAVIORAL

TREATMENT OF THE EATING DISORDERS
As noted previously, CBT is intended to be a symptom-focused, short-term
treatment that attempts to directly modify eating disorder symptoms and
associated behaviors. An emphasis is placed on proximal causal or maintaining
factors for the disorders, and distal, historical, and causal factors are
deemphasized. Furthermore, interventions are generally oriented around active
patient participation, self-monitoring of ongoing behavior, completion of
homework, active efforts to confront and modify thinking, and other
behaviorally oriented approaches, while more psychodynamic interventions,
focusing on treatment resistance, transference, countertransference, and
unconscious cognitive thought processes, are not utilized.
CBT is also a structured treatment, which generally follows a predetermined
sequence of strategies. For example, Table 2 outlines a general sequencing of
treatment phases from CBT for BN, which may be modified somewhat for
application with AN or BED. Phase I is quite behavioral in nature with an
emphasis on teaching the patient essential medical and dietary concepts as well
as to identify when problem behaviors are occurring and environmental stimuli
that precipitate or prevent eating disorder symptoms. Furthermore, the patient
is directly encouraged to diversify their meal plan with a particular emphasis on
inclusion of feared foods. Phase II takes a decidedly more cognitive approach
with an emphasis on identifying maladaptive cognitions and subjecting them to
careful examination and empirical testing to determine their accuracy and
objectivity. Also, patients are encouraged to monitor sequences of their behavior
in the form of eliciting cues and mediating behaviors and cognitions and,
ultimately, eating disorder behaviors. Finally, in phase III patients who have
TABLE 2 Phases of Cognitive-Behavioral Therapy
made substantial behavioral change are encouraged to develop plans to prevent

relapse and maintain a healthier lifestyle.
Initiating CBT
The initial sessions of CBT must take into account the patient’s likely
ambivalence about making a change and emphasize the development of
therapeutic rapport. In particular, eating-disordered individuals with highly
obsessional, paranoid, or avoidant cognitive styles may enter treatment with
considerable hesitation. Several authors (37,40) have discussed approaches to
managing such ambivalence that are consistent with CBT. These interventions
typically include recognizing the adaptive functions of the disorder for the
patient, accepting the patient’s thought processes as genuine, and a
demonstration by the clinician of a clear knowledge of both general
psychopathology and also eating disorders.
Self-Monitoring
Early in CBT, the patient is provided with some type of self-monitoring form,
which may include information about time of day, environmental situations,
food and liquid consumed, presence of eating disorder symptoms, and
asso ciated maladaptive thoughts and feelings. Such a self-monitoring approach
is a clinically useful way to gather a considerable amount of information about
the patient’s day-to-day “ecological niche” surrounding their eating disorder
symptoms. Patients should be instructed about the importance of
selfmonitoring, which is designed to provide them with a greater sense of
control over and knowledge about their eating disorder symptoms.
Early in treatment, patients may be encouraged to record the consumption of
meals, snacks, and bulimic symptoms (binges) in their selfmonitoring records.
This allows the clinician to see the presence or absence of well-structured
schedules of eating that are reasonably dispersed throughout the day.
Obviously, anorexic individuals’ food records will reveal the absence of healthy
and diversified eating, while the record of a bulimic individual may reveal a
more chaotic and unpredictable pattern of eating. Self-monitoring may
ultimately include an identification of environmental cues for particular types of
eating behaviors, including binge eating, skipping meals, and purging.
Eventually, the patient may also include data on particular thoughts and feelings
in the face of specific environmental events or cues. Thus, the selfmonitoring
may help the clinician and patient to identify a pattern of particular
environmental stimuli, associated intrapsychic activity in the form of cognitions
and feelings, and overt behavioral activity such exercising, binge eating, or
purging.
Cues and Consequences

One important aspect of self-monitoring is to help the patient identify stimuli
that elicit problematic eating behaviors. For example, self-monitoring may
reveal that periods of significant food restriction precede binge eating or that
interpersonal conflict precedes periods of significant starvation. It is important
for the individual to begin to identify those stimuli—whether cognitive,
behavioral, or environmental—that increase the likelihood of specific eating
disorder behaviors. Once these stimuli are identified, patients need to learn to
respond to these stimuli in an alternative fashion. For example, if driving by a
grocery store is a trigger to stop, buy binge food, and binge-eat, the patient may
simply attempt to identify an alternative route that does not include exposure to
the grocery store. Another useful strategy for dealing with triggers for
problematic behaviors is distraction. For example, the experience of seeing ones
body shape in a mirror may be a trigger for purging behavior or excessive

exercise. However, if the patient can learn to create a “pause” between the
stimulus and the problematical behavior, there may be time to develop
alternatives and modify thought processes. Also, it may be useful for patients to
learn to use certain stimuli to elicit positive healthy behaviors, which are part of
their recovery. For example, patients may be encouraged to eat all of their
meals at scheduled times and at a particular place in their home. These changes
help to link positive eating behaviors to particular stimuli and avoid the
association of eating and random or unpredictable stimuli.
Treatment should also focus on the consequences of eating behavior. First,
patients need to be educated about the potential negative consequences of eating
disorder symptomatology. Typically, these include medical, social, and financial
consequences of their eating disorder behavior. However, patients should also be
encouraged to identify the positive consequences of their eating disorder
behavior. For example, for the extremely shy and inhibited individual, the
solitary lifestyle of AN with extreme eating behaviors and individual exercise
regimes may assist them in managing social anxiety. Alternatively, binge eating
and purging behaviors may help the bulimic individual to avoid negative
emotions associated with critical self-evaluation and negative social comparison.
It is often useful to discuss with patients how the eating disorder is helpful to
them and begin to identify alternative means of meeting similar needs.
Recently, some clinicians have utilized strategies in which patients identify the
positive and negative aspects of their eating disorder behaviors, and are also
asked to identify what the consequences of their eating disorder behaviors will be
5 or 10 years from now (41,44). Furthermore, they might be asked how their
lives might change if they continue to engage in their eating disorder as opposed
to achieving recovery. Such strategies help the patient to identify more clearly
the potential destructive consequences of their symptoms in the future but also
fully acknowledge the benefits they obtained from these symptoms currently.
Cognitive Restructuring
Figure 1 depicts a basic cognitive model underlying CBT. This model
emphasizes the close relationship between thoughts, feelings, and behaviors. In
CBT, cognitions or thoughts are a central operational construct. It is not
stressful environmental stimuli that are ultimately thought to result in eating
disorder symptoms but the cognitive appraisal of those stimuli. For example, an
eating-disordered individual may see her body in the mirror (stimulus) and have
the thought “Oh no, I can’t believe how fat I’ve become.” Such a thought is
associated with a negative emotional response, and in combination this thought
and feeling may increase the likelihood of some selfdestructive behavior
FIGURE 1 Model underlying cognitive-behavioral therapy.
(i.e., dieting, purging). Thus, eating-disordered individuals are encouraged to

examine their cognitions carefully, identify instances of maladaptive thinking,
and attempt to identify a more adaptive response to the cue.
Typically, patients are able to identify problematic thinking patterns through
self-monitoring. At this point in the treatment, self-monitoring logs may begin
to include areas in which cognitions and feelings can be recorded. Patients are
encouraged to simply describe their thoughts or feelings without censorship as
they occur. This may be difficult at first and patients must be encouraged to
practice identifying their thoughts. Often, eating-disordered individuals, like
persons with other forms of psychopathology, display specific types of
maladaptive thinking. Beck et al. (2) catalogued many of these cognitions in
depressed individuals and found recurring themes and categories that have been
described in eating disorder patients as well (37). These include dichotomous
thinking (e.g., if I am not beautiful, I am ugly), overgeneralization (e.g., I
wasn’t able to follow a meal plan today so I will never be able to follow one),
emotional reasoning (e.g., I know I am fat because I feel fat), catastrophizing
(e.g., I have gained 3 pounds in 2 weeks, I know that I am becoming obese), and
others. Patients should be taught about various forms of maladaptive thinking so
that they can begin to monitor for the presence of such thinking in their daily
activities. Once the patient has identified such negative thinking patterns, they
can begin to work toward modification. Two general strategies are frequently
employed in CBT for the eating disorders: (a) behavioral experimentation and
(b) testing the validity of the thought-through questions.
Behavioral experimentation refers to identifying a negative thought and the
logical or predicted consequence or outcome associated with that thought.
Then, the individual intentionally creates a behavioral situation, that should result
in the predicted outcome if the thought is accurate. Typically, because of the
high degree of distortion in the maladaptive thought, the behavioral prediction
is not confirmed through the experiment. For example, if an individual has the
thought “eating three meals a day will make me fat,” the patient and clinician can
devise an experiment to test this thought. For example, the patient may agree
that if she was able to eat three meals a day for at least 5 days, this would
provide a good test of whether or not she would gain weight, as oppose to
staying the same or perhaps even losing weight. If the patient were willing to try
this on a trial basis for 5 days, she could actually weigh herself before starting
and again 5 days later. Of course, the patient would have to agree to engage in
the experiment and be willing to report on the outcome. Most likely, at least in
BN, there would be little change in weight because calories may be ingested
through meals rather than traditional binge eating patterns. The patient should
be encouraged to identify the inference that can be drawn from the experiment
and whether or not it conflicts with the prediction associated with the eating
disorder.
The second means of restructuring thought patterns is to assist the patient in
questioning the actual thought. For example, if a patient had the thought “I will
be happy when I am below 100 pounds,” the therapist may ask a series of
questions to test the validity of this idea. For example, the patient may be asked
if she had ever been below 100 pounds previously and whether or not she was
happy then. If not, the therapist may ask what it means that she weighed less
than 100 pounds and was not happy. If she states that she was happy, the
therapist may ask if she was happy all the time. It would be expected that she
was not happy all the time, and that could lead to a discussion of what other
factors must have influenced her mood besides weight, given that she was under
a 100 pounds but still had periods of unhappiness. Also, the patient may be
asked if she knew any other people who weighed less than a 100 pounds and
whether or not they were happy. In general, the therapist should take the
approach of asking the patient what evidence there is to support the thought
that weighing less than 100 pounds will make her happy and, alternatively, what
evidence there is that fails to support the thought. Ideally, the therapist and
patient collaboratively examine the evidence for the thought, examine its
validity, and begin to develop more adaptive and accurate alternative thought
patterns.
Early in treatment, cognitive restructuring tends to focus on issues

surrounding food, weight, and shape. However, as the individual gains greater
control of symptoms, cognitive restructuring may move to other associated
problems. Issues regarding interpersonal relationships, intimacy, academic or
vocational success, and broader issues associated with personality functioning
may also be targeted. However, these issues are unlikely to be a central theme
for the therapy, especially in short-term treatment.
Recently, cognitive approaches to the management of AN have placed a
greater emphasis on interpersonal functioning and deeper cognitive structures,
particularly self-schemas and interpersonal schemas (41). As the therapist listens
to descriptions of maladaptive or automatic thoughts, recurrent themes of
cognition about self or others may be identified. Particularly with anorexic
individuals, working with these more generalized and stable attitudes about the
self and the behavior of others may be useful. For example, the individual who
believes that she must be loved by everyone, and that other people tend to be
extremely critical and rejecting, displays schemas about self and others that are
highly maladaptive and may promote further entrenchment in eating disorder
symptoms as a means of avoiding frightening interpersonal scenarios. Helping
such patients to recognize the thoughts about self and others and possibly the
origins of such thinking is important, but the fundamental approach to cognitive
restructuring remains intact. Individuals must subject such schemas to empirical
tests to evaluate their accuracy. However, it is important to remember that
attempts to modify these types of deeper cognitive structures may be arduous
and time consuming, and will probably not be included in short-term
treatments.
Problem Solving and Social Skills Training

Other strategies may be included in CBT for eating disorders. Often, patients
will suggest that their eating disorder symptoms have become coping
strategies for a wide variety of problems. In this situation, it is important that
the therapy focus on the more productive problem-solving strategies. In some
fashion, the following problem-solving steps should be introduced to the
patient: (a) recognize that problems cannot be avoided and are a normal part of
living; (b) specify in a detailed fashion the nature of the problem; (c) once the
problem is clearly understood, brainstorm solutions to the problem without
worrying about the adequacy of the solutions; (d) after carefully considering the
pros and cons of each option, choose one of the solutions and try it out; (e) after
trying one of the solutions, evaluate it to see if the problem has been resolved.
This series of steps, originally outlined by D’zurilla and Goldfried (46), may
with practice help the individual to identify solutions to problems that are
outside the realm of eating disorder symptoms.
One specific type of problem solving lies in the area of interpersonal
relationships. Many problems in living involve relationships to other people.
When this is the situation, patients should be encouraged to carefully identify
the nature of the problem they are having with the other person. Assertiveness
training is a helpful approach to several types of interpersonal problems,
including difficulty in expressing negative emotions, problems setting limits,
and problems expressing positive feelings. In therapy, role playing may
greatly facilitate the patient’s ability to effectively deal with problematic
interpersonal situations. Seeing the clinician model appropriate social responses
in a role play with the patient may facilitate imitation. The old teaching axiom
of “see one, do one, teach one,” is an appropriate way to think about improving
social skills.
Case Study
Cindy was a 28-year-old, married mother of a 6-year-old son and 3-year-old
daughter at the time of presentation to the outpatient eating disorders clinic.
She had recently relocated to the area after her husband took a managerial
position with a local manufacturing company. The impetus to her seeking
treatment was the distress and shame she’d been experiencing since her daughter
walked into the bathroom and observed her mother vomiting.
Cindy was seen for an initial evaluation, which included a detailed review of
the onset of eating disorder symptoms, current eating patterns and
compensatory behaviors, thoughts and feelings associated with weight and shape,
weight and menstrual history, as well as assessment of comorbid
psychopathology, chemical use, medical status, and psychosocial history.
Cindy reported the onset of eating disordered behavior at the age of 15 when
she entered a new high school following her family’s relocated. She recalls her
volleyball teammates ridiculing others’ appearance and feared that she would
also be the subject of ridicule. Cindy began skipping breakfast and lunch (a
practice common to her peer group) and consuming only low-calorie and low-
fat foods at the evening meal with her family. She was motivated to continue
dieting by the attention she received from peers. Cindy recalled first self-
inducing vomiting after she ate pizza and sweets at a postgame celebration.
Vomiting behavior quickly escalated, and Cindy also began to binge-eat. She
indicated that because she could vomit she would “make it worth my while” and
consume “forbidden foods” she enjoyed. Binge eating and vomiting continued at
a rate of two or three times a week throughout high school and her early college
years. Although she continued to be extremely concerned about her weight and
shape, Cindy maintained a 6-year abstinence from binge eating and vomiting
after she left college, married, had a child, and breast-fed.
Cindy described a gradual relapse in eating disorder symptoms in the year
before her relocation. Anticipating loneliness in a new town, frustrated that she
had not met her goal of completing her degree, and increasingly disparaging of
her body and abilities, Cindy attempted to lose weight through dietary
restriction believing she would feel more confident and better about herself in
general. Binge eating and vomiting quickly returned and escalated. At the time
of presentation to the outpatient eating disorder clinic, Cindy weighed
145 pounds at a height of 5 foot 4 inches (body mass index = 24.9). She
reported extreme dissatisfaction with her shape, feeling extremely fat and
unworthy as a result of her weight. Cindy reported her ideal weight as
110 pounds, believing she would be more confident, a better mother, and more
acceptable to others.
In attempts to achieve her ideal weight, Cindy had a daily plan to avoid
dietary fat, restrict calories to 1000 kcal/day, and exercise for 60 min. Cindy
described “losing the battle” to follow her strict plan and admitted to binge
eating and purging daily. Each day began with attempting to avoid eating, only
to be preoccupied with food, often distracting her from tasks at hand and time
with her children. She typically nibbled on food she prepared for her daughter’s
snack and midday meal. Chastising herself, concluding she had already “blown
it” for the day, and vowing to begin anew the next day, Cindy reported binge
eating on her children’s uneaten meals, 12 ounces of chips, and 6 cups of ice
cream in the early afternoon during the children’s nap time.
Cindy denied use of laxatives, enemas, diuretics, syrup of ipecac, chewing
and spitting, or rumination of food. She denied exercising in an excessive
manner, in fact berated herself for not achieving her plan of daily exercise.
Cindy had been on an antidepressant for 15 months; however, she continued to
struggle with dysphoric mood, amotivation, low energy, poor concentration,
and low self-esteem. Based on evaluation results, it was recommended that
Cindy participate in individual CBT for bulimia. She was referred to her
primary physician for review of medications and serum electrolytes.
Cindy’s treatment began with a detailed description of the cognitive model
of the maintenance of bulimia with particular emphasis on the cyclic relationship
between strict dieting (avoidance of eating, restriction of intake, and food
avoidance), binge eating, and purging. Cindy agreed that insecurity and low self-
esteem precipitated and exacerbated concerns about her weight and shape, and
she believed that her eating disorder had become a way to cope with these
negative mood states. Self-monitoring was presented as the initial treatment
assignment. Cindy was instructed to record the time and place of eating
episodes; type and amount of food eaten; whether the episode was considered a
meal, snack, or binge; whether vomiting occurred; and thoughts, feelings, and
circumstances surrounding the episode of eating (Fig. 2). She was informed of
the importance of accurate self-monitoring and its part in providing vital
information on triggers for eating disorder behaviors as well as guiding
treatment sessions. Despite anticipating embarrassment about the therapist
being aware of the intimate details of her eating disorder, Cindy left the session
stating an understanding of and commitment to the CBT plan.
Despite her stated commitment, Cindy arrived at her second scheduled
appointment having not self-monitored. Time was taken to explore with Cindy
FIGURE 2 Food and liquid daily monitoring form.

what had interfered with her completing this crucial first assignment, and the
discussion gave way to identifying the advantages and disadvantages of
addressing her eating disorder. Cindy expressed significant ambivalence
regarding treatment, embarrassment and shame about her symptoms, and fear of
becoming fat were she to change her behavior. Cindy was provided with
reassurance that most normal-weight patients do not gain weight as a result of
CBT.
Cindy presented to her third session having completed self-monitoring,
which was reviewed in great detail with particular emphasis on potential cues of
binge eating and vomiting. A pattern of dietary restriction, avoiding foods
thought to be “fattening,” and avoiding eating in general were highlighted as
examples of strict dieting which left Cindy vulnerable to binge eating. The
importance of eating consistently throughout the day was stressed, and Cindy
collaborated on developing a plan for having regular meals and snacks consisting
of food and portion sizes that she was prepared to consume without vomiting.
Cindy was easily able to adapt her eating to the scheduled eating pattern she
provided for her young children and found that three meals and three snacks
could be consumed in the presence of others, which would further decrease the
risk of binge eating and vomiting.
The next several sessions focused on implementing Cindy’s plan for
consistent eating while continuing to monitor her weight through weekly
weighing. Cindy was surprised by an improvement in energy and mood as well
as renewed interest in spending time with her children. Given her significant
fear of weight gain and subsequent food avoidance, Cindy was instructed to
develop a hierarchy of feared foods. Each day, Cindy developed a plan for
introducing a food from the bottom of the hierarchy that may trigger binge
eating and engaging in an alternative behavior immediately following the
consumption until the urge to vomit subsided. Weekly weighing provided
Cindy with valuable evidence that these foods could be consumed without
weight gain, which in turn provided motivation for continuing to add previously
avoided foods.
As episodes of binge eating and vomiting decreased and became intermittent,
Cindy’s treatment began to focus on those factors that had been maintaining her
eating disorder, in particular distorted thoughts and beliefs about food, eating,
weight and shape, and the use of dietary restricting, binge eating, and vomiting
as a means of managing life stress and negative mood. The presence of
maladaptive thoughts as cues for eating disorder behaviors was identified using
self-monitoring forms where Cindy had also been monitoring her thoughts
associated with eating episodes. These thoughts were then challenged using
cognitive restructuring. For example, Cindy identified her belief that “eating fat
will make me fat” and more specifically indicated a belief that she would gain
5 pounds if she consumed a food containing more than 5 grams of dietary fat per
serving. Cindy was challenged to test this belief by setting up an experiment in
which she consumed ice cream without vomiting and later in the week
monitored her weight. She was instructed to consume the ice cream in the
presence of her husband, utilize alternative behaviors for a specified period of
time and/or spend time out of the house should the urge to vomit occur. This
process of challenging maladaptive thoughts through empirical testing and
evidence gathering was repeated as Cindy worked through her list of feared
foods, exposed herself to high-risk situations, and confronted negative thoughts
about her weight and shape.
Cindy’s participation in CBT also revealed a longstanding pattern of engaging
in binge eating and vomiting in response to a variety of negative mood states,
including loneliness, frustration, interpersonal conflict, and selfdisparagement.
Cindy was increasingly able to implement cognitive restructuring and
alternative behaviors at such times, but struggled with extreme negative self-
appraisal, which continued to place her at risk for binge eating and vomiting.
Not completing her educational degree and the lack of social contacts since
relocating were identified as cues for self-disparagement and subsequent eating
disorder behavior. Problem solving was applied wherein Cindy clearly identified
why social isolation was problematic, generated options for solving the problem,
including joining a parents-of-preschoolers group, taking courses to finish her
degree, attending community functions, and joining a church. A similar process
of problem solving was applied to Cindy’s struggle with not completing her
degree, as well as a variety of problems of daily life.
After 4 months of treatment, Cindy achieved abstinence from binge eating
and purging, was consistently eating a variety of foods, and reported increasing
confidence as she tackled what had previously been viewed as insurmountable
environmental stressors. The final stage of treatment focused on maintaining the
changes that Cindy had made in treatment and developing a relapse prevention
plan. Cindy considered what situations may place her at risk for relapse (e.g.,
another relocation, her daughter’s transition into school) and developed a plan
for managing eating disorder symptoms should they arise. She was informed of
the possibility of a reemergence of symptoms in times of stress or transition and
was reassured that utilizing the skills she developed in treatment would likely
curtail a relapse. Cindy reviewed the techniques she had learned and devised a
plan for returning to self-monitoring, making sure she was utilizing alternative
behaviors and remaining connected with her social supports. Cindy agreed that
three consecutive weeks of binge eating and vomiting should signal the need to
return to treatment; however, she felt confident that she would be able to
interrupt the cycle of binge eating and vomiting.
CONCLUSION
CBT is a treatment for a wide variety of eating disorder symptoms that has
received considerable empirical support and continues to be the modality of
choice for a variety of eating disorder clinical presentations. CBT may evolve
and be modified in future studies; already applications in telemedicine,
CDROM, and Palm Pilot applications have emerged. Furthermore, variations
of CBT have been developed with a greater emphasis on interpersonal
functioning, emotional responding, and alternative cognitive structures. There
continues to be a need for enhanced approaches to educating mental health
professionals about the applications of CBT to eating-disordered patients, and a
lack of dissemination of these treatments remains a limiting factor for eating-
disordered individuals.
ACKNOWLEDGMENTS
Supported by research grants from the McKnight Foundation and the National
Institute of Mental Health (ROl-MH59234) and the National Institute of
Diabetes, Digestive and Kidney Diseases (R01-DK61912; R01-DK60432;
P30-DK50456), and the Neuropsychiatric Research Institute.
REFERENCES
1. Beck AT, Emery G. Anxiety Disorders and Phobias. New York: Basic Books,
1985.
2. Beck AT, Rush AJ, Shaw BF, Emery G. Cognitive Therapy of Depression. New
York: Guilford Press, 1979.
3. Peterson CP, Mitchell JE. Cognitive behavior therapy. In: Gabbard GO, ed.
Treatment of Psychiatric Disorders. 2d ed. Washington, DC: APA Press, 1995.
4. Fairburn CG, Marcus MD, Wilson GT. Cognitive behavioral therapy for binge
eating and bulimia nervosa: a comprehensive treatment manual. In: Fairburn CG,
Wilson GT, eds. Binge Eating: Nature Assessment and Treatment. New York:
Guilford Press, 1993:361–404.
5. Mitchell JE, Peterson CB. Cognitive-behavioral treatment of eating disorders.
Dickstein LD, Riba MB, Oldham JM, eds. Review of Psychiatry. Vol. 16:.
Washington, DC: APA Press, 1995:1–107–1-134.
6. Crow S, Mitchell JE. Bulimia nervosa. In: Mitchell JE, ed. The Outpatient
Treatment of Eating Disorders: A Guide for Therapists Dietitians and Physicians.
Minneapolis: University of Minnesota Press, 2001:26–59.
7. Lacey JH. An outpatient treatment program for bulimia nervosa. Int J Eat Disord
1983; 286:1609–1613.
8. Connors ME, Johnson CL, Stuckey MK. Treatment of bulimia with brief
psychoeducational group therapy. Am J Psychiatry 1984; 141:1512–1516.
9. Yates AJ, Sambrailo F. Bulimia nervosa: a descriptive and therapeutic study. Behav
Res Ther 1984; 5:505–517.
10. Kirkley BG, Schneider JA, Agras WS, Bachman JA. Comparison of two group
treatments for bulimia. J Consult Clin Psychol 1985; 53:43–48.
11. Ordman AM, Kirschenbaum DS. Cognitive behavioral therapy for bulimia: an
initial outcome study. J Consult Clin Psychol 1985; 53:305–313.
12. Lee NF, Rush AJ. Cognitive-behavioral group therapy for bulimia. Int J Eat Disord
1986; 5:599–615.
13. Wolchik SA, Weiss L, Katzman MA. An empirically validated short-term
psychoeducational group treatment program for bulimia. Int J Eat Disord 1986; 5:
21–34.
14. Wilson GT, Rossiter E, Kleifeld EI, Lindholm L. Cognitive behavioral treatment of
bulimia nervosa: a controlled evaluation. Behav Res Ther 1986; 24:227–238.
15. Fairburn CG, Kirk J, O’Connor M, Cooper PJ. A comparison of two psychological
treatments for bulimia nervosa. Behav Res Ther 1986; 24:629–643.
16. Laessle RG, Waadt S, Pirke KM. A structured behaviorally oriented group
treatment for bulimia nervosa. Psychother Psychosom 1987; 48:141–145.
17. Leitenberg H, Rosen JC, Gross J, Nudelman S, Vara LS. Exposure plus response
prevention treatment of bulimia nervosa. J Consult Clin Psychol 1988; 56:
535–541.
18. Freeman CPL, Barry F, Dunkeld-Turnbull J, Henderson A. Controlled trial of
psychotherapy for bulimia nervosa. Br Med J; 296:521–525.
19. Agras WS, Schneider JA, Arnow B, Raeburn SD, Telch CF. Cognitive behavioral
and response prevention treatments for bulimia nervosa. J Consult Clin Psychol
1989; 57:215–221.
20. Mitchell JE, Pyle RL, Eckert ED, Hatsukami D, Pomeroy C, Zimmerman R. A
comparison study of antidepressants and structured intensive group psychotherapy
in the treatment of bulimia nervosa. Arch Gen Psychiatry 1990; 47:149–157.
21. Fairburn CG, Jones RT, Peveler RC, Carr SJ, Solomon RA, O’Connor ME,
Burran J, Hope RA. Three psychological treatments for bulimia nervosa. Arch Gen
Psychiatry 1991; 48:463–469.
22. Fichter MM, Leibl K, Rief W, Brunner E, Schmidt-Auberger S, Engel RR.
Fluoxetine versus placebo: a double-blind study with bulimic inpatients undergoing
intensive psychotherapy. Pharmacopsychiatry 1991; 24:1–7.
23. Wilson GT, Eldredge KL, Smith D, Niles B. Cognitive behavioral treatment with
and without response prevention for bulimia. Behav Res Ther 1991; 29:575–583.
24. Laessle RG, Beumont PJV, Butow P, Lennerts W, O’Connor M, Pirke KM, Touyz
SW, Waadt S. A comparison of nutritional management with stress management in
the treatment of bulimia nervosa. Br J Psychiatry 1991; 159:250–261.
25. Garner DM, Rockert W, Davis R, Garner MV, Olmsted MP, Engle M.
Comparison of cognitive behavioral and supportive expressive therapy for bulimia
26. Agras WS, Rossiter EM, Arnow B, Schneider JA, Telch CG, Raeburn SD, Bruce B,
Perl M, Koran LM. Pharmacologic and cognitive behavioral treatment for bulimia
nervosa: a controlled clinical comparison. Am J Psychiatry 1992; 149: 82–87.
27. Mitchell JE, Pyle RL, Pomeroy C, Zollman M, Crosby R, Seim H, Eckert ED,
Zimmerman R. Cognitive behavioral group psychotherapy for bulimia nervosa: the
importance of logistical variables. Int J Eat Disord 1993; 14:277–288.
28. Thackwray DE, Smith MC, Bodfish JW, Meyers AW. A comparison of behavioral
and cognitive behavioral interventions for bulimia nervosa. J Consult Clin Psychol
1993; 61:639–645.
29. Goldbloom DS, Olmsted M, Davis R, Clews J, Heinmaa M, Rockert W, Shaw B.
A randomized controlled trial of fluoxetine and cognitive behavioral therapy for
bulimia nervosa. Behav Res Ther 1997; 35:803–811.
30. Walsh BT, Wilson GT, Loeb KL, Devlin MJ, Pike KM, Roose SP, Fleiss J,
Waternaux C. Medication and psychotherapy in the treatment of bulimia nervosa.
Am J Psychiatry 1997; 154:523–531.
31. Agras WS, Walsh BT, Fairburn CG, Wilson GT, Kraemer HC. Multicenter
comparison of cognitive behavioral therapy and interpersonal therapy for bulimia
32. Hsu LK, Rand W, Sullivan S, Liu DW, Milliken B, McDonagh B, Kaye WH.
Cognitive therapy, nutritional therapy and their combination in the treatment of
bulimia nervosa. Psychol Med 2001; 32:871–879.
33. Bulik CM, Sullivan PF, Carter FA, McIntosh VV, Joyce PR. The role of exposure
with response prevention in the cognitive-behavioural therapy for bulimia nervosa.
Psychol Med 1998; 28:611–623.
34. Agras WS, Crow SJ, Halmi KA, Mitchell JE, Wilson CT, Kraemer HC. Outcome
multisite study. Am J Psychiatry 2000; 157:1302–1308.
35. Jacobi C, Dahme B, Dittmann R. Cognitive-behavioural, fluoxetine and combined
treatment for bulimia nervosa: short-and long-term results. Eur Eat Disord
Rev2002; 10:179–198.
36. Agras WS, Rossiter EM, Arnow B, Telch CF, Raeburn SD, Bruce B, Koran L. One
year follow up of psychosocial and pharmacologic treatments for bulimia nervosa.
J Clin Psychiatry 1991; 52:29–33.
37. Peterson CB, Mitchell JE. Cognitive behavioral therapy for eating disorders. In:
Mitchell JE, ed. The Outpatient Treatment of Eating Disorders: A Guide for
Therapists, Dietitians, and Physicians. Minneapolis: University of Minnesota Press,
2001:145–167.
38. Smith DE, Marcus MD, Kaye W. Cognitive behavioral treatment of obese binge
eaters. Int J Eat Disord 1992; 12:257–262.
39. Wilfley DE, Agras WS, Telch CF, Rossiter EM, Schneider JA, Cole AG, Sifford L,
Raeburn SD. Group cognitive behavioral therapy and group interpersonal therapy
for the nonpurging bulimic: a controlled comparison. J Consult Clin Psychol 1993;
61:296–305.
40. Wilfley DE, Welch RR, Stein RI, Spurrell EB, Cohen LR, Saelens BE, Dounchis JZ,
Frank MA, Wiseman CV, Matt GE. A randomized comparison of group cognitive
behavioral therapy and group interpersonal psychotherapy for the treatment of
overweight individuals with binge eating disorder. Arch Gen Psychiatry 2002; 59:
713–721.
41. Garner DM, Vitousek KM, Pike KM. Cognitive behavioral therapy for anorexia
nervosa. In: Garner DM, Garfinkel PE, eds. Handbook of Treatment for Eating
Disorders. 2nd ed. New York: Guilford Press, 1997:94–144.
42. Serpell L, Treasure J. Bulimia nervosa: Friend or foe? The pros and cons of bulimia
43. Channon S, De Silva P, Hemsley D, Perkins R. A controlled trial of
cognitivebehavioural and behavioural treatment of anorexia nervosa. Behav Res
Ther 1989; 27:529–535.
44. Pike KM. Relapse prevention for anorexia nervosa. Paper presented at the Seventh
New York International Conference on Eating Disorders, New York, 1996.
with eating disorders (revision). Am J Psychiatry 2000; 157(Suppl):l-39.
46. D’zurilla TJ, Goldfried MR. Problem solving and behavior modification. J Abnorm
Psychol 1971; 78:197–226.
18
An Overview of Family Evaluation and
Therapy for Anorexia Nervosa, Bulimia
Deborah Marcontell Michel and Susan G.Willard
Tulane University School of Medicine
New Orleans, Louisiana, U.S.A.
Family therapy encompasses a group of psychotherapies that focus on the entire

family unit in terms of interaction and change, instead of focusing on any
individual member. Although the specific theories, treatment approaches, and
goals of particular schools vary, most strive to facilitate change within the family
in order to reduce or eliminate problematic interactions and behaviors.
The roots of family therapy can be traced back to the 1950s when Gregory
Bateson conducted his seminal research with schizophrenics and their families
on levels of communication. His research, along with many notable others such
as Nathan Ackerman, Murray Bowen, Salvador Minuchin, Jay Haley, and Virginia
Satir, led to the family movement. This movement represented more than just a
new approach to therapy. Instead, it was a different way of looking at behavior
that emphasized the examination of communication between individuals in
relationships (1).
In this chapter, we begin by identifying and explaining some of the basic
concepts of family therapy, and by linking these principles to how family
therapists work to facilitate change. Next, we provide a brief historical
perspective on the family treatment of those with eating disorders and
descriptions of common family characteristics. Major models of family therapy
used to treat persons with eating disorders are then presented, along with the
empirical evidence to support their use. We also provide guidelines for clinical
practice of family therapy including important areas of inquiry in family
evaluation, common issues in family therapy, and potential obstacles to
assessment and treatment.
BASIC CONCEPTS AND PRINCIPLES OF FAMILY

THERAPY
To appreciate the family therapist’s perspective of a given individual’s behavior,
a synopsis of the basic concepts and principles of family therapy is in order.
428 MICHEL AND WILLARD
Foley (2) described the family unit as an open system. A system comprises sets of
different parts that are interconnected with all parts affecting the other parts.
Furthermore, each part has a relationship to the other parts in a stable manner.
The family system is considered open because it has a continuous flow of
elements entering and leaving the system. Open systems are further
characterized by wholeness, relationship, and equifinality. Wholeness is the
interdependence between the parts of the system. This principle emphasizes not
only the individuals in the family but also the relationships among them.
Relationship, then, examines the interactions between the parts of the family and
focuses on what is happening, not why it is happening. The family therapist
therefore carefully analyzes interactions among all family members. Finally,
equifinality refers to the tenet that open systems are not dictated by their initial
conditions; thus, difficulties can be eliminated if changes are made at any point
in time within the system. Consequently, a family therapist focuses on the
present, not the past, with the idea that an intervention can be made at any time
to change the system.
According to Foley (2), family relationships are made up of interlocking
triangles that lend stability to the system and function to reduce or increase its
emotional intensity. Thus, whenever the emotional balance between two family
members becomes too distant or too close, a third family member can restore
equilibrium and stability. One of the tasks of the family therapist is to analyze
the existing triangles in the family system and intervene to change the system. The
process by which the system adjusts itself is termed feedback. When negative
feedback is given in response to deviation from the existing system, the deviation
is corrected and balance is restored. By contrast, positive feedback prevents the
system from returning to its original state, forces it to change, and thereby
destroys the former system. Instead of attempting to directly correct a
problematic behavior, the family therapist may use this principle to exaggerate a
symptom to the point at which it causes the original dysfunctional system to
collapse.
HISTORY OF FAMILY THERAPY IN TREATMENT

OF EATING DISORDERS
Today the need for family therapy in treating eating disorders from a
biopsychosocial approach is widely accepted (3). References have been made,
however, to the importance of the role of the family in the development and
maintenance of anorexia nervosa (AN) ever since its appearance in medical
literature during the 1800s. Early writers believed that the families were
harmful and should be excluded from the patient’s treatment (4,5). It was not
until the early 1970s that the family dynamics involved in AN were more fully
FAMILY EVALUATION AND THERAPY 429
appreciated through the observations and writings of Hilde Bruch (6). She noted
the association between eating disorders and difficulty with individuation and
emotional separation from the family of origin. In the mid to late 1970s, Mara
Selvini-Palazzoli (7) and Salvador Minuchin and colleagues (8,9) pioneered the
systematic application of family therapy to the treatment of AN with their schools
of systems therapy and structural therapy, respectively. More recently, a group
of clinicians at the Maudsley Hospital in London have developed the Maudsley
model of family therapy for AN (10).
Obviously, the history of family therapy in the treatment of eating disorders
has focused on the development of methods to treat individuals with AN. These
approaches have also been applied to the family treatment of bulimia nervosa
(BN) and will be discussed later. For a more detailed account of the history and
development of family therapies in the treatment of eating disorders, see Dare
and Eisler (10).
FAMILY CHARACTERISTICS OF INDIVIDUALS

WITH AN EATING DISORDER
Minuchin and his cohorts at the Philadelphia Child Guidance Center presented
family characteristics they found typical of families having a child with
AN (8,9). The attributes included enmeshment, rigidity, overprotectiveness,
lack of conflict resolution, and involvement of the child in parental discord.
Enmeshment referred to the diffusion of boundaries resulting in emotional
overinvolvement. Rigidity in the family organization reflected a lack of
flexibility in the face of developmental stages requiring change. The observed
overprotectiveness often was not limited to the identified patient or the eating
disorder, yet it served to constrain the child’s autonomy and independence. The
rigidity and overprotectiveness of the families left little room for conflict
negotiation. Consequently, problems were usually left unresolved. Involvement
of the ill child in parental conflict (a detrimental triangle) resulted in the child
playing an important role in the family pattern of conflict avoidance. As a result,
the child received powerful reinforcement for remaining ill.
In families having a child with BN, Schwartz et al. (11) observed the same
interaction patterns described by the Philadelphia group in addition to social
isolation, appearance consciousness, and a special meaning attached to food and
eating. In a study of family therapy, these researchers found that the particular
issues around which the attributes were displayed depended on the degree of
“Americanization,” or drive to achieve dominant cultural values, versus ethnic
identity of the family in question. From this research, the authors identified
three family typologies that are very similar to the ones described below.
As cited in Michel and Willard (12), three types of family organizations

commonly seen in families who have a daughter with BN have been described by
Root et al. (13). The authors did not observe differences based on subtype and,
although their focus was on a bulimic population, our experience is that these
characteristics may also be seen in families who have a daughter with AN. One
type of organization is the “perfect family.” Appearance is of utmost
importance, as this family strives to appear perfect to outside observers and the
family reputation is sacred. Family members are therefore expected to think,
act, and feel the “right” way. The family is governed by rigid rules and the
identified patient tends to be a very high achiever relative to family standards.
An eating disorder in this family may represent any or all of the following:
(a) passive rebellion, (b) creation of a separate identity, (c) suppression of or
distraction from feelings by focusing on food, and (d) assertion of personal
control in the midst of familial control.
A second family organization, as described by Root and her colleagues (13),
is the “overprotective family..” This family encourages dependence and refuses
to acknowledge needs for independence. Anger is seldom openly displayed, so
that passive-aggressive behaviors develop, thereby hindering direct expressions
of rebellion. Eating disorder symptoms may reflect a means of passive rebellion
and indirect expression of anger while simultaneously reaffirming dependence
on family members.
The last family organization discussed by Root et al. (13) is the “chaotic
family,” which has virtually no rules, organization, or consistency. The
availability and expression of love is unpredictable, and conflicts are usually
resolved by physical aggression and/or psychological intimidation. Physical and/
or sexual abuse is often present. Consequently, eating disorder symptoms may
function as a safe way to express anger, provide a means of separation from the
abusive situation, and/or assert consistency in one’s life. It can, however, also
represent a form of self-destructive behavior.
There are other family characteristics commonly found among families of an
individual with an eating disorder (12). Weight is often a family issue. One or
more family members may be overweight or excessively thin, and there may be
an overconcern with food and food-related issues in the household. In this case,
it is not uncommon to see one or both parents engaged in chronic dieting and/or
rigid exercise regimens. Adolescents in these families may be especially
vulnerable to the use of dieting and exercising as a means of bonding with
parents who are obviously preoccupied with such issues themselves. In addition,
some parents of children with eating disorders have vicariously experienced
their children’s successes and accomplishments. As signs of puberty ensue, the
parents often get depressed or frightened about what the future will hold when
the children are grown and gone. The unconscious pressure on these youngsters,
then, is to maintain the child role and to not grow up. Family loyalty to retain
homeostasis is often chosen over normal development, which would lead to
separation and independence. Other characteristics that may be observed
include a highly critical parent, parental depression, or marital problems. In the
latter case, the child may receive the spoken or unspoken message that
independence, autonomy, and leaving home will cause the family system to fall
apart.
We are aware of only one study describing the family characteristics of
individuals with binge eating disorder (BED). Hodges et al. (14) compared the
families of those with BED, AN-restricting type (AN-R), AN-binge/eating
purging type (AN—B+P), BN, and normal controls. The investigators found
that those with BED endorsed less family cohesion and less encouragement to
express honest feelings than the other eating disorder groups. They also
described their families as more isolated, more sedentary, and limited in terms
of emphasis on independence. Furthermore, the BED group rated their families
lower on achievement orientation, intellectual-cultural orientation, and moral-
religious emphasis. Finally, they described their families as having less structure,
fewer rules, and less predictability than the other eating disorder groups. When
compared to individuals with AN-R and AN-B+P, the BED group reported
more conflict that was similar to the group with BN.
Compared to subjects without an eating disorder, individuals with BED in
the Hodges et al. (14) study reported more familial conflict and described their
families as more rigid in terms of rules and procedures used to regulate family
life. The BED group also endorsed less independence, cohesiveness, and
expressiveness in addition to less focus on pursuit of political, social, cultural,
and intellectual endeavors or participation in social or recreational activities.
The authors caution that data were gathered via self-report and no other means
of corroboration obtained, yet the patients’ accounts may provide useful
information for treatment.
FAMILY EVALUATION
In order to determine the needs of the family and to decide what therapeutic
interventions are appropriate, an evaluation of the family is in order. This is in
accordance with the APA Practice Guidelines for the Treatment of Eating
Disorders (15). The evaluation should include all members living within the
home and, at times, the clinician may bring in other family members who do
not live in the home but have been identified as family members having
significant relationships with the identified patient. The purpose of the
assessment is to ascertain what role, if any, the family environment played in the
development and maintenance of the eating disorder and to what degree those
issues, if present, remain (16).
Evaluation of the family includes basic psychosocial information about
members such as demographic data, present and past living arrangements,
psychiatric history, medical history, educational and occupational history, and
social history. Inquiry into significant family events is also important as is the
assessment of family traditions. Gathering background information on each
parent’s family of origin is helpful so that multigenerational patterns of relating
and behaving can be identified. Regarding the evaluation of physical or sexual
abuse, our family and individual assessments are conducted separately by
different therapists. Although we acknowledge that not all clinicians are in a
position to practice in this manner, it is our opinion that this arrangement
makes it comparatively easier for the identified patient to disclose such sensitive
information. Our experience in most cases of ongoing abuse, however, is that it
will not be disclosed until a stronger therapeutic relationship has been
established with the individual psychotherapist. Though we are unaware of any
published data to directly support this clinical observation in a psychotherapy
setting, it is consistent with research demonstrating significant delay in child and
adolescent disclosure of sexual abuse, if the abuse is disclosedatall(17).
In addition to conducting a standard psychosocial interview of the family,
Andersen (18) compiled a list of areas which should be investigated including
(a) interactional patterns, (b) flexibility, (c) sensitivity, (d) family supports and
stresses, (e) performance of stage-appropriate tasks, and (f) family knowledge
of the illness. Examination of interactional patterns includes such topics as
quality of the marital relationship, spousal agreement on parenting behavior,
family satisfaction and companionship, communication patterns, and the general
affective atmosphere of the family. Flexibility refers to how easily the family
system allows members to change in terms of communication and roles they
play in response to situations and stressors. In evaluating the sensitivity
component, the practitioner assesses whether family members demonstrate
emotional hypersensitivity and overreactivity, or un involvement and
insensitivity, to one another. Evaluation of family supports and stresses includes
the degree of support, or lack thereof, that family members provide to one
another as well as any sources of significant strength and stress within the
family. In addition, evaluation of outside supports and stressors is important.
Regarding the performance of stage-appropriate tasks, the clinician assesses the
age appropriateness of rules and responsibilities that are assigned to family
members, particularly children. Finally, the clinician needs to know family
members’ understanding of the illness in terms of etiology, treatment, and
recovery as well as their thoughts, feelings, and behaviors associated with it. In
particular, knowledge of family attitudes and behaviors that may hinder the
patient’s ability to recover is critical so that they may be quickly examined and
resolved (16). Similarly, family members’ preoccupation with weight and
appearance, which may sabotage the identified patient’s efforts at recovery,
must be addressed early on (19).
Vanderlinden and Vandereycken (20) stress that family assessment is a
continuous process throughout family treatment. They recommend performing
a functional analysis of the eating disorder within the family system. The goal is
to formulate hypotheses about how the eating disorder functions within the
family. The following questions, then, are important to remember throughout
the process: (a) How does the symptom serve to stabilize the family? (b) What
role does the family play in stabilizing the symptom? (c) Around what central
theme is the problem organized? (d) What consequences will follow change in
the family? (e) What is the therapeutic dilemma?
Standardized, self-report instruments that the practitioner may find helpful in
assessing quality of familial relationships and interactions from an individual
family member’s point of view are available. A selection of these include the
Leuven Family Questionnaire (21), the Family Evaluation Scale (FES; 22), the
Family Adaptation and Cohesion Evaluation Scale (FACES; 23), and the Family
Assessment Measure (FAM; 24). Of course, it is incumbent on the linician to be
familiar with the purposes, strengths, and limitations of any questionnaire
chosen for use as explained by the test developer.
FAMILY THERAPY IN THE TREATMENT OF

ANOREXIA NERVOSA
As mentioned earlier, Minuchin, Selvini-Palazzoli, and their coworkers were
among the first to develop specific family treatments for AN. Minuchin’s group
described a psychosomatic family model of AN (8,9), in which certain factors
combine to produce children with severe psychosomatic illnesses. The factors
include a physical vulnerability in the child, dysfunctional family characteristics
(described above), and an adaptive role that the sick child plays in the family’s
pattern of conflict avoidance. Their approach to treatment was coined structural
family therapy, the goal of which was to alter the dysfunctional family
interactions by limiting certain family interactions and encouraging others. By
doing so, the need for the symptomatic behavior (i.e., anorexic behavior) would
be reduced or eliminated. For example, the therapist determines that a mother
is enmeshed with her ill daughter, preventing the daughter from having
meaningful relationships with her father and brother. The structural therapist
would attempt to increase the emotional distance between the mother and
daughter, while simultaneously encouraging closer relationships between the
daughter and her father, between the siblings, and between the parents.
Minuchin’s group also advocated that the parents take charge of their anorexic
child’s eating by instituting “family meal sessions.” During these sessions, it was
the job of the parents to unite in taking control over their child’s eating
behavior while the therapist monitored family interactions, intervening when
necessary.
Evidence in support of the structural approach, as reported by Minuchin
et al. (9), consists of a detailed study spanning 7 years. During that time,
assessment, treatment, and follow-up data were collected on 50 families having
a child with AN. Six of the identified patients were males. The median age was
14.5 years with a range from 9 to 21 years. All patients received treatment less
than 3 years after the onset of illness. The average length of treatment was
6 months with a range of 2–16 months. Family therapy was typically conducted
on a weekly basis.
Therapists in the study (9) described three trends in treatment that emerged
based on the identified patient’s age. Preadolescents were not seen individually.
Instead, conjoint therapy (identified patient seen with family) was conducted
initially with a shift to marital therapy later. Familial issues centered on
increasing parental control and effectiveness as well as improving the parental
coalition. For the adolescent group, conjoint family therapy and marital therapy
were conducted along with individual sessions. Siblings were seen alone with
the identified patient when judged appropriate. Issues centered on increasing
autonomy and independence in the adolescent. In the young adult sample, the
primary therapeutic issue was separation from the family of origin. Conjoint
family sessions alone were therefore done in the beginning, with a quick shift to
separate individual plus nonconjoint family sessions.
Of the original sample, 80% were followed for 2 years or more. The total
follow-up time ranged from 1.5 to 7 years. Results of the investigation showed
that 86% of the patients achieved full recovery from anorexic symptoms and
correlated psychosocial deficits. Four percent improved somewhat in both
areas, 6% did not improve, and 4% relapsed (9). Martin (as cited in Dare and
Eisler, 10) demonstrated similar findings with a related model of therapy.
Though considered a landmark in family treatment of anorexia nervosa,
Minuchin’s study has been criticized for various methodological weaknesses,
including inadequate sample sizes, overlapping meanings of constructs, a focus
on pathological polarities regarding familial interactions, and a lack of empirical
measures to assess outcome (25,26).
To address some of the criticisms of Minuchin’s psychosomatic model,
authors of a Flemish study operationally defined and measured the concepts by
behavioral methods and self-report (26). The concepts were behaviorally
redefined as follows: “intensity of intrafamilial boundaries” for enmeshment,
“degree of family adaptability” for rigidity, “the family’s way of handling
conflict” for lack of conflict resolution, and “degree of avoidance/recognition of

intrafamilial tension” for overprotectiveness. The study included families having
a child with AN, BN, or eating disorder-not otherwise specified. Evidence of
convergent and discriminant validity for intensity of intrafamilial boundaries,
degree of family adaptability, and the family’s way of handling conflict was
found using the behavioral methods. Self-report showed only convergent
validity for the family’s way of handling conflict. The findings did not support
the concept of overprotectiveness. The authors concluded that it is important for
clinicians to keep in mind the difference in conceptualizations of the family
climate between professional observers and family members when conducting
family therapy.
The Milan systems model developed by Selvini-Palazzoli and her colleagues in
Italy (7,27) emphasized the role of the eating disorder in maintaining
homeostasis in the family system. Homeostasis refers to the balance that occurs
when all family members adhere to their given, often unspoken, rules of
behavior. The Milan group believed that these rules had become overly rigid in
families of anorexic patients and interactions were therefore limited to those
that kept the status quo. According to Dare and Eisler (10), therapy was
designed to introduce a new way of conceptualizing the problematic
interactions and symptomatic behavior, noting their adaptive functions in the
family. To accomplish this task, an end-of-session intervention in the form of
a”message to the family” may be given. In addition to remaining neutral in
relation to the family, the Milan group also believed that it was important to
remain neutral as to whether or not change in the family should occur and, if so,
in what form. Consequently, the Milan family systems therapist gleans
information from the family via interview, encourages them to examine their
interactions, and leads them to ultimately challenge problematic beliefs and
interactions. The Milan group’s studies have been criticized for a lack of
systematic study of interaction, matched control groups, and methodologically
appropriate assessment procedures (28).
Additional support for the Milan approach has been documented by Stierlin
and Weber (as cited in Dare and Eisler, 10) in a study of families having a child
with AN or BN. Family therapy tended to be brief with an average of six sessions
per family. Upon termination, treatment gains in regard to eating disorder
symptoms were limited but at follow-up, which ranged from 2 to 9 years,
approximately two-thirds of the subjects had attained a normal weight and were
menstruating. Similar improvements were noted in family and peer
relationships. The study was limited, however, by the fact that it was not
conducted under controlled conditions.
The Maudsley model of family therapy for adolescents with AN incorporates
structural components but differs from the two aforementioned approaches in
that these practitioners do not assume that the family is dysfunctional (10).
Instead, this London-based group posited that a significant amount of the
“dysfunction” reported in families with a child with AN may be the result of the
development of a life-threatening illness, changes in the child’s mood and
behavior, potential blaming by professionals, and failure of initial therapeutic
efforts (10). Nevertheless, the Maudsley clinicians observed many of
the same family characteristics as described by Minuchin and Palazzoli (29).
Le Grange (30) explains that therapy is time limited and typically proceeds
through three clearly defined phases: (a) refeeding the person with anorexia
nervosa, (b) negotiating new patterns of relationships, and (c) termination.
During the refeeding phase, the focus is on eating disorder symptoms. Like the
structural family therapists, the Maudsley group advocates that the parents take
charge of the child’s eating and a family meal session may be included in this
phase. Once the child begins to cooperate with increased food intake and there
is a discernible change of mood within the family, patterns of relationships and
other family issues can be discussed. Such discussion takes place around how these
problems affect the parents in their efforts to ensure that the child’s weight
increases steadily. Following weight restoration, the termination phase begins
with an emphasis on the child establishing a healthy adolescent or young adult
relationship with the parents that does not include the eating disorder. Some
clinicians also see the patient individually (10).
In adult patients with AN, the approach is modified so that attempts are made
to remove the eating disorder from its central and controlling role in the
relationships between the identified patient and other family members.
Furthermore, there is not typically a push for the parents to take charge of the
adultchild’seating(31).
Recently, the Maudsley approach was used to develop a professional, family-
based treatment manual targeting adolescents with AN (32,33). Favorable
results were reported for weight restoration and associated psychological
features at 6-month follow-up. The authors caution, however, that the results
are preliminary, short term, and include only 19 cases. A larger, longer term
treatment trial is underway at this writing.
The Maudsley group has conducted a number of controlled trials comparing
different forms of therapy for adolescents with AN in an effort to shape their
own approach (10). In addition, they have shown that adolescents under age 19
with anorexia nervosa for less than 3 years respond very well to the Maudsley
model of family therapy in the short term (34,35) and at 5year follow-up (36).
In adults with AN, an early study tentatively demonstrated that adult patients
derive greater benefit from individual supportive therapy than from the
Maudsley model of family therapy for adults (34). A more recent study (31)
found that adult patients benefited from 1 year of family therapy, but not
significantly more so than from 1 year of focal psychoanalytic psychotherapy or

7 months of cognitive-analytical therapy. In addition, benefits were limited in
that some patients were substantially malnourished at follow-up. It was noted
that this particular group carried a poor prognosis as defined by late onset of
illness, a chronic course of illness, and unsuccessful previous treatments, all of
which may have negatively affected the outcome.
The Maudsley group has also been interested in how expressed emotion of
family members affects family treatment for both AN and BN. Expressed emotion
is a pattern of hostile, critical, and intrusive interactions first studied in families
of schizophrenics with high rates of such associated with poor prognosis (37).
One study from the Maudsley researchers found that high maternal expressed
emotion predicted early dropout from family treatment but not from individual
treatment (38). Criticism from either parent at the beginning of treatment
predicted a poor outcome, leading researchers to conclude that for highly
critical families, separate parent counseling may be of more benefit than conjoint
family therapy (39). Similarly, a Dutch study using a variety of family therapies/
counseling approaches to manage adolescent AN and BN found that a critical
maternal attitude was predictive of a poorer outcome at termination and at 1-year
follow-up (40).
A fourth model of family therapy for treating adolescents with AN is
behavioral family systems therapy (41,42). It closely resembles the Maudsley
approach with conjoint family sessions and three phases of treatment. In the first
phase, the parents are encouraged to take control of the youngster’s eating with
coaching from the therapist, much as they would if she were a young child
unwilling to take medicine for a disease. Once successful, the second phase focuses
on family structure, the role of the eating disorder symptoms in the family
system, and cognitive distortions about food, body weight, and family life.
Once the identified patient’s target weight is reached, the third phase starts by
gradually returning control of eating back to the adolescent. Issues of
individuation and communication between the parents and adolescent are also
addressed.
Robin et al. (42) describe an investigation comparing behavioral family
systems therapy to ego oriented individual psychotherapy (41), in which the
adolescent receives weekly individual sessions and the parents attend bimonthly
family sessions alone. They found that the systems approach produced greater
weight gain upon termination. By 1-year follow-up, however, this effect had
diminished. Regarding eating attitudes, depression, and interoceptive awareness
(awareness and identification of internal bodily sensations such as hunger), the
two forms of therapy were equally effective at termination, 1-year follow-up,
and in a limited sample of subjects at 4-year follow-up. With respect to general
family conflict, both groups improved comparably, based on the
researchers’judgments of videotaped interactions. Interestingly, no significant

reports of this type of conflict were reported by family members before or after
treatment. Both groups reported a high level of eating-related conflict prior to
treatment, which improved for both groups following treatment. Thus, because
both groups included family interventions, Dare and Eisler (10) concluded that
the findings tend to underscore the value of family therapy in the treatment
of AN.
Much has been written about the families of those with AN, and overall, the
literature on family treatment, particularly in adolescents, clearly warrants its
use in clinical practice. Despite variations in approaches, all are concerned with
the family’s role in food and weight issues as well as individual concerns,
familial issues, and psychosocial functioning. Our experience, too, corroborates
the necessity for family treatment of AN. More specifically, it is through family
therapy that vital, core issues that may be determined to underlie the eating
disorder symptoms (such as separation and individuation) can be identified and
resolved.
FAMILY THERAPY IN THE TREATMENT OF

BULIMIA NERVOSA
The literature on family treatment for BN is sparse compared to that for AN.
Several reasons for this discrepancy have been postulated (42). First, it is a
comparatively new disorder, having been given a distinct identity in 1980.
Second, those affected tend to be older and may no longer be living with their
families of origin at the time they seek treatment. Nevertheless, in the following
section we present two systemic treatment models developed for a bulimic
population as well as a synopsis of studies examining the effectiveness of family
therapy in the treatment of BN from the structural, Milan, and Maudsley
approaches.
Root et al. (13) developed a systemic-feminist approach to treatment based
on their theoretical descriptions of family types described earlier, although they
did not perform studies to test the efficacy of the therapy. The therapeutic
model incorporated aspects of the structural, Milan, and strategic schools of
family therapy. In their experience, this family population did not always present
with all of the characteristics that Minuchin and his colleagues described.
Instead, they observed the family typologies outlined earlier (perfect,
overprotective, and chaotic). Their model of family therapy centered around
systemic, developmental, and life cycle issues as they played out in the families
they encountered. The issues included individuation and separation, boundaries,
organization, and difficulties expressing and resolving feelings. Enmeshment and
difficulty leaving home often characterized the individuation and separation
issues. The boundary problems encompassed physical space and privacy, rules
about respect for boundaries, and tolerance of emotional distance. Boundaries
were sometimes lacking or impermeable. Regarding organization, systems need
rules or organization to regulate actions or expressions. Thus, a system can be
too rigidly organized or it can be disorganized. In the experience of Root and her
coworkers, all the families they treated had difficulty expressing and resolving
feelings, particularly anger, resentment, jealousy, grief, depression, anxiety,
and insecurity. The clinicians argued that difficulties in this arena resulted in the
actual bulimic symptoms; they postulated that a system that disallows certain
feelings and cannot adapt encourages either an explosive discharge of feelings or
development of psychosomatic illnesses. Thus, a child in this type of system
subsequently learns that directly expressing certain emotions causes a parent to
become explosive or ill.
Another purely theoretical approach to family treatment for BN has been
described by Jack Brandes (43) of the University of Toronto. He asserted that
families having a child with BN have difficulties with attachment and loss. In his
practice, many families related accounts of actual loss and abandonment through
illness, death, change in socioeconomic status, and immigration followed by
social isolation. He hypothesized that the concerns these families have with
appearance, demeanor, and performance may represent worries about being
unacceptable and unappealing to society. Symptomatic behavior, i.e., bulimic
behavior, may therefore be a means of coping with social isolation, separation,
and loss. The treatment model Brandes (43) proposed to treat families having a
child with BN consisted of four phases that embraced structural, cognitive, and
psychoeducational components. He noted that the family may be seen
exclusively, or individual and/or marital sessions may be added as clinically
appropriate. The first phase of treatment is assessment in which the therapist
“joins” with the family (establishes a therapeutic alliance), performs a family
history, works to reduce shame and blame, and educates the family about BN.
The second phase, early treatment, focuses on dietary reeducation and
encouraging the identified patient to own the bulimic symptoms. There is also
an emphasis on loss and abandonment themes, self-sacrifice of the identified
patient, and individual identity in the family. The third stage of treatment
addresses individual concerns, marital issues, separation, and integration into
the community. Termination, the final phase of treatment, focuses on feelings
of loss and abandonment regarding treatment. To ease the difficulty of
termination, the therapist contracts for future contacts with the family.
An investigation by Schwartz et al. (11) involved 30 families having an
adolescent or adult child with BN treated from a structural-systemic approach.
The treatment model focused on patient differentiation from the family of
origin, with work in this area initiated prior to directly targeting bulimic
symptoms so that symptoms would not be reactivated by family dynamics.

Treatment occurred in the following stages: (a) motivating the patient and
family for differentiation, (b) guiding the differentiation process, (c) targeting
the symptoms with specific interventions, and (d) relapse prevention. In this
study, the mean frequency of bulimic episodes was 19 per week, ranging from 5
to 63. Individual sessions were often alternated with family sessions, and
treatment lasted an average of 27 sessions over 9 months with a range of 2–90
sessions. The average length of illness was 6.8 years with a range of 1–23.
Follow-up spanned from 1 to 42 months with a mean of 16 months. At follow-
up, 66% of the identified patients were either abstinent or had 1 bulimic
episode per month, 10% had 2 episodes per month up to 1 per week, 10%
reported 2–4 episodes per week, and 14% suffered from more than 5 episodes
per week. Of the 7 patients functioning at the bottom two levels, 4 dropped
out of treatment with fewer than 8 sessions. Thus, overall, the majority had a
good outcome with a structural-systemic approach even though the severity
level was high and the identified patients had a chronic course of illness. The
study is limited by its relatively small sample size and lack of controlled
experimental conditions.
Further support for the structural model as it relates to BN is seen in the
work of Kog et al. (26). As previously discussed in the section on AN, evidence
for three concepts of the psychosomatic model from which the structural
approach stems was found.
German investigators tested the Milan model in a bulimic population by
comparing a 2-month trial of inpatient psychoanalytical group therapy to 1 year
of systemic outpatient therapy (44). Outcome was measured at 14, 26, and
38 months after treatment initiation. Results indicated that both therapies were
effective in reducing symptoms of BN and correlated features, with only a slight
advantage for the inpatient treatment regimen despite the other therapies
associated with it. Nevertheless, the researchers were unable to differentiate the
effects of the two treatment regimens.
An early study by the Maudsley group (34) compared response to family
therapy and to individual supportive psychotherapy in young adult women with
BN. No substantial differences were found between the groups, and both had a
poor outcome. The authors speculated that the sole use of the given
interventions was not sufficient to produce a more positive outcome. A later
study by the group investigated the Maudsley model in eight female adolescents
with BN (45). As in treatment for AN, the parents are charged with the task of
normalizing the child’s eating and reducing binge-purge episodes with the
guidance of the therapist. After eating patterns have been regulated, control
over eating behavior is given back to the adolescent. Other issues are then
addressed, as outlined earlier in the section on AN. Results indicated that one
subject had a good outcome, five had an intermediate outcome, and one had a
poor outcome. Overall, average outcome scores significantly improved
following treatment, although the study was limited by a small sample and lack
of a control group.
Despite the relative dearth of empirical information on the effectiveness of
family treatment in BN, family therapy is clinically accepted as an important
component in the comprehensive treatment of BN, particularly in younger
patients. Our experience has shown family therapy to be quite helpful in
resolving issues that may contribute to the development and maintenance of the
eating disorder. It is particularly useful when used in conjunction with
individual psychotherapy focused on individuation and identity formation
combined with the use of cognitive-behavioral techniques to address symptom
reduction.

BED is a newly described phenomenon that is currently diagnosable only with
the eating disorder-not otherwise specified category of the Diagnostic and
Statistical Manual of Mental Disorders, 4th ed, (46). It was also proposed as an area
for further study. Subsequently, much research has begun in this area on clinical
correlates, etiology, and treatment. As of this date, we are unaware of any case
descriptions or controlled trials on the use of family therapy for this disorder.
Although there are no investigations on the use of family therapy in BED, one
study examined the effectiveness of spousal involvement in treatment.
Gorin et al. (47) conducted a study of 94 females comparing the use of
manualized, standard group cognitive-behavioral treatment (CBT) of BED,
group CBT with spouse involvement, and a wait-list control group. The average
age of participants was 45 years and subjects were randomly assigned to the
treatment conditions. Spouses were asked to attend all sessions and assist in
setting goals to help with binge eating cessation and weight stabilization.
Frequency of binge eating, body mass index (BMI), eating psychopathology,
general psychopathology, and marital satisfaction were evaluated pre-
and posttreatment as well as at 6-month follow-up. Although both CBT groups
significantly improved in terms of binge eating, BMI, eating psychopathology,
and general psychopathology when compared to controls, spousal involvement
did not result in superior results compared to standard CBT. The authors
speculated that the type of spousal intervention may not have been effective or
that there may be something about the nature of BED that limits spousal
influence on treatment. They also hypothesized that spousal involvement may
have served as a catalyst for binge eating in some subjects if the spouses were
perceived to be critical or unsupportive. The study was limited by attrition
across the experimental conditions (34%) and by the lack of a specialized

measure to diagnosis BED.
We have approached family treatment for BED based on need as judged
through clinical assessment of the patient. As with the other eating disorders, if
it is determined that the family plays a major role in the origin and maintenance
of the disorder, family therapy is recommended. Given that patients presenting
with BED tend to be older, family therapy may not necessarily include the
family of origin. Instead, the patient’s current family unit is a more likely
candidate. Nevertheless, any family members may be asked to participate if such
is clinically expected to be useful.
FAMILY THERAPY IN CLINICAL PRACTICE

Our own approach to family therapy with those afflicted with an eating disorder
stems from a combined structural and Milan systems conceptualization. We
have stated elsewhere (12) that the family system is typically the context out of
which an eating disorder has developed and therefore requires some degree of
change if the affected adolescent is to overcome her eating disorder in that
environment. For patients who are in a relatively older age range and no longer
live with their families of origin, family therapy may not be necessary. In some
cases, patients do well in treatment without family participation. When a patient
still lives with her family, however, family therapy is often very helpful for
recovery. It is also indicated if the family is living separately but is obviously
functioning as a stimulating factor in the illness. If the patient is married, marital
therapy may be suggested as an adjunctive treatment.
Perednia et al. (48) described a pyramid model of family treatment in which
there are different levels of intervention based on familial need. All parents
receive guidance and education counseling, a subset participate in family therapy
(which may evolve into marital therapy for some couples), and, finally, a few
parents are referred for individual psychotherapy. This model allows for
maximum flexibility in treatment from a family-oriented position, with the
authors asserting that not all families are “disturbed.” Instead, they posit that
some suffer from a temporary, situational crisis, whereas others exhibit more
basic psychopathology.
From a structural-systemic perspective, family therapy is typically geared to
understanding the role that the individual with an eating disorder has
characteristically held within the family system and how the illness has
contributed to maintaining whatever homeostasis has been achieved. Table 1
lists common therapeutic issues. Like Perednia et al. (48), we have noted (12)
that family therapy sometimes evolves into marital therapy for the parents or
individual therapy for one or both parents, particularly as the child gets better.
TABLE 1 Common Issues in Family Therapy of AN and BN
Source: Ref. 12.
In our experience, it is not unusual for a parent to be referred for individual

therapy at the outset of treatment, if it is clear that the primary family problem
lies within that particular parent-child relationship.
Whether or not dysfunctional family characteristics are part of the cause of
eating disorders in offspring or if these observed problems are a result of the
stress associated with having a child with an eating disorder is controversial (3).
In fact, there is a great deal of information accumulating on the genetic
transmission of eating disorders (see Chapter 7), though experts recognize that
it is likely to be the interaction between genetics and environment, including
sociocultural influence, that is responsible for the development of these
disorders (49). Regardless, it is never helpful or appropriate to place blame on
family members for the development of the illness (12). Explaining the
biological propensity that an individual may have toward an eating disorder can
provide the practitioner with a nonblaming means of joining with the family,
and any stigma which may be felt is likely to be reduced. By educating the
family about eating disorders in this manner and enlisting their support in
therapeutic endeavors, any potential defensiveness may be reduced or avoided.
Consequently, the family’s insight and esteem can be enhanced, facilitating full
engagement in the therapeutic process. Most families experience strong
emotional responses in reaction to a child’s development of an eating disorder.
These responses may include feelings of fear, frustration, guilt, demoralization,
desperation, and anger, and many parents blame themselves or others (50).
Family therapy may therefore serve as a much-needed forum for parents and
siblings to identify, express, and resolve their thoughts, feelings, and beliefs as
well as receive education about the nature of eating disorders and recovery.
In our practices, family therapy is conducted in conjunction with individual

psychotherapy using separate therapists. It has been acknowledged that the use
of family therapy with individual psychotherapy greatly improves the chance of a
full recovery (19), and this combination of treatment has been recommended in
the Practice Guideline for the Treatment of Eating Disorders (15). We believe
that having separate therapists facilitates the individuation-separation process.
More specifically, this method allows the patient to have a person, i.e.,
psychotherapist, of his or her own with whom to work on individuation,
separation, and identity issues. Simultaneously, the family works with a therapist
of their own, with whom they can address their concerns and find support and
guidance. In addition, we argue that this arrangement makes it is easier for the
individual therapist to manage confidentiality issues between the identified
patient and other family members. It also eases the therapeutic burden of
treating such complicated and challenging illnesses by having two therapists
dividing the labor (43). Finally, we believe, like others (51), that this
combination of treatment addresses family dysfunction (if it is present) on two
levels, namely, the individual level where it has been introjected, as well as in
the current family system. As part of this team approach, the family therapist
does not see the identified patient individually, but may see (a) all family
members, (b) the identified patient with the parents, (c) the identified patient with
the siblings, (d) the parents alone, or (e) the siblings alone. In other words,
different combinations of individuals may be seen at any given time depending
on the therapeutic issues at hand. Of course, regular communication between
the individual psychotherapist and family therapist is essential to avoid splitting
and to ensure that everyone is headed in the same direction with
mutual goals (12).
Again, we recognize that not all practitioners are in a practical or theoretical
position to utilize separate individual and family therapists. Positive aspects of
one psychotherapist treating the identified patient and the family include
immediate knowledge of individual and family issues, efficiency, and cost
containment. Thus, successful treatment is certainly possible with one
psychotherapist, although the sole clinician may have a greater number of
therapeutic issues to process simultaneously.
Goldner and Birmingham (35) noted that while some schools of family
therapy advocate parental control over the child’s eating, most clinicians do not
follow this approach. Their belief is that the family can be most helpful by
relinquishing the battles around food and weight, and strengthening their
relationships in other areas. They encourage families to transfer the
responsibility of food and weight concerns to the identified patient and the
treatment team. Their experience with families using this approach has been
positive, with families reporting the diminishing of long-standing battles
therefore allowing them to work on communication skills and improved

relationships. We, too, subscribe to this approach, with food issues handled by
an experienced registered dietician who is a critical member of our treatment
team. Our experience has been similarly positive.
FAMILY OBSTACLES TO ASSESSMENT AND

TREATMENT
Although many families welcome help for themselves and the identified patient,
some may present obstacles to assessment and treatment (52). Denial of an
eating disorder, minimization of the problem, and/or denial of the psychological
underpinnings may be present (15,53,54). For some families, according to
Michel (37), education about eating disorders may remove these obstacles
altogether, or at least help diminish them. For others, the psychological origins
of the disorder may be too disturbing for family members to deal with directly.
Consequently, therapeutic efforts may be resisted as the family struggles to
maintain homeostasis. Treatment resistance, as well as a lack of familial
motivation to change, may be related to denial and/or a need to maintain the
identified patient’s role in the family system. Regardless of the etiology of the
defensiveness or resistance, these issues must be resolved before a therapeutic
alliance can be established, thereby increasing the likelihood of achieving a
positive outcome (12,55).
SUMMARY
The family therapy movement began in the 1950s and represented a change in
the way in which behavior was viewed, emphasizing present interactions and
relationships between family members. The majority of family treatment
models and studies on the efficacy of family therapy in treating eating disorders
have focused on AN. Nevertheless, it is widely accepted in clinical practice that
family therapy is required in comprehensive treatment for both AN and BN,
particularly in younger patients still living at home. Empirical support for family
treatment of AN and BN exists, but well-controlled outcome investigations are
somewhat limited (3). In the case of BED, a recently defined problem, family
treatment has not been studied. However, there may be a role for family
therapy in treatment if the clinician determines that family members have a role
in the origin or maintenance of the illness.
Dysfunctional family characteristics of those with eating disorders have been
identified and, for those with AN or BN, clinicians have described family
typologies. Whether or not these attributes are part of a dysfunctional system
that produces an individual with an eating disorder is debatable, as some experts
believe that it may be the result of the stress associated with having a child with
AN or BN. In addition, the importance of biological factors in the development
of AN and BN is becoming more evident, along with environmental influences.
Regarding the environmental factors, we have previously noted (12) that the
family may contribute to the development of, or exacerbate, an eating disorder
by providing an environment that can hinder an adolescent in establishing an
identity, practicing effective communication skills, and/or learning adaptive
coping skills. Within the context of the family, his or her eating disorder may
function to establish an identity separate from the family, to cope with
stressors, to distract from negative feelings, and to provide what the patient
considers to be a means of “safe” self-expression of feelings.
With regard to clinical practice, it is usually essential for patients living with
their families of origin to participate in family therapy, as everyone in the family
is affected by these disorders. For those no longer living at home, family therapy
may be limited to couples therapy, marital therapy, or the current family system.
If, however, members of the family of origin are found to still stimulate the
eating disorder, it will be necessary to include them in family treatment. In very
critical or abusive families, family therapy may need to take the form of parent
counseling or nonconjoint family therapy. A thorough evaluation of the family
and family dynamics at the outset will assist the clinician in sorting out these
issues and in conducting effective family therapy. In addition, the combination of
family therapy and individual therapy appears to be quite prevalent as each form
of treatment can provide a useful and complementary function, which facilitates
recovery. Finally, some families present obstacles to assessment and treatment,
including denial of the eating disorder, minimization of the problem, and/or
denial of the psychological underpinnings. These issues must be resolved and the
family engaged in a therapeutic alliance in order to increase the likelihood of
achieving a successful outcome.
REFERENCES
1. Hoffman L. Foundation of Family Therapy: A Conceptual Framework for Systems
Change. New York: Basic Books, 1981:16–19.
2. Foley VD. Family therapy. In: Corsini RJ, ed. Current Psychotherapies. 3rd ed.
Itasca, IL: F.E. Peacock, 1984:447–490.
3. Lemmon CR, Josephson AM. Family therapy for eating disorders. Child Adol
Psychiatry Clin of North Am 2001; 10:519–542.
4. Gull WW. Anorexia nervosa (apepsia hysterica, anorexia hysterica). Trans Clin Soc
London 1874; 7:22–28.
5. Lasegue C. De 1’anorexie hysterique. Arch Gen Med 1873; 1:384–403.
7. Selvini-Palazzoli M. Self-Starvation. New York: Jason Aronson, 1974.
8. Minuchin S, Baker L, Rosman BL, Liebman R, Milman L, Todd TC. A conceptual
model of psychosomatic illness in children. Arch Gen Psychiatry 1975; 32:
1031–1038.
9. Minuchin S, Rosman BL, Baker L. Psychosomatic Families: Anorexia Nervosa in
Context. Cambridge, MA: Harvard University Press, 1978.
10. Dare C, Eisler I. In: Garner DM, Garfinkel PE, eds. Handbook of Treatment for
Eating Disorders. New York: Guilford Press, 1997:307–324.
11. Schwartz RC, Barrett MJ, Saba G. Family therapy for bulimia. In: Garner DM,
Garfinkel PE, eds. Handbook of Psychotherapy for Anorexia Nervosa and Bulimia.
New York: Guilford Press, 1985:280–307.
12. Michel DM, Willard SG. When Dieting Becomes Dangerous: Understanding and
Treating Anorexia and Bulimia. New Haven, CT: Yale University Press, 2003.
13. Root MPP, Fallon P, Friedrich WN. Bulimia: A Systems Approach to Treatment.
New York: W.W. Norton and Co., 1986.
14. Hodges EL, Cochrane CE, Brewerton TD. Family characteristics of binge-eating
disorder patients. Int J of Eat Disord 1998; 23:145–151.
15. American Psychiatric Association Work Group on Eating Disorders. Practice
guideline for the treatment of patients with eating disorders. Rev Am J Psychiatry
2000; 57(suppl l):l-39.
16. Woodside DB, Shekter-Wolfson LF, Garfinkel PE, Olmsted MP. Family
interactions in bulimia nervosa II: complex intrafamily comparisons and clinical
significance. Int J Eat Disord 1999; 17:117–126.
17. Paine ML, Hansen DJ. Factors influencing children to self-disclose sexual abuse.
Clin Psychol Rev 2002; 22:271–295.
18. Andersen AE. Practical Comprehensive Treatment of Anorexia Nervosa and
Bulimia. Baltimore, MD: The Johns Hopkins University Press, 1985:135–148.
19. Pelch BL. Eating disordered families: Issues between the generations. In: Lemberg
R, Cohn L, eds. Eating Disorders: A Reference Sourcebook. Phoenix: OryxPress,
1999:121–123.
20. Vanderlinden J, Vandereycken W. Family therapy within the psychiatric hos pital:
Indications, pitfalls, and specific interventions. In: Vandereycken W, Kog E,
Vanderlinden J, eds. The Family Approach to Eating Disorders: Assessment and
Treatment of Anorexia Nervosa and Bulimia. New York: PMA Publishing, 1989:
263–310.
21. Kog E, Vertommen H, DeGroote T. Family interaction research in anorexia
nervosa: the use and misuse of a self-report questionnaire. International Journal of
Family Psychiatry 1985; 6:227–243.
22. Moos RH, Moos BS. Family Environment Scale Manual. 2d ed. Palo Alto, CA:
Consulting Psychologists Press, 1986.
23. Waller G, Slade P, Calam R. Family adaptability and cohesion: relation to eating
attitudes and disorders. Int J Eat Disord 1990; 9:225–228.
24. Skinner H, Santa-Barbara J, Steinhaur P. The family assesment measure. Can J

Commun Ment Health 1983; 2:91–105.
25. Kog E, Vandereycken W, Vertommen H. Multimethod investigation of eating
disorder families. In: Vandereycken W, Kog E, Vanderlinden J, eds. The Family
Approach to Eating Disorders: Assessment and Treatment of Anorexia Nervosa and
Bulimia. New York: PMA Publishing, 1989:81–106.
26. Kog E, Vertommen H, Vandereycken W. Minuchin’s psychosomatic family model
revised: a concept-validation study using a multitrait-multimethod approach. Fam
Proc 1987; 26:235–253.
27. Selvini-Palazzoli M, Boscolo L, Cecchin G, Prata G. Paradox and Counterparadox.
New York: Jason Aronson, 1978.
28. Kog E, Vandereycken W. The facts: a review of research data on eating disorder
families. In: Vandereycken W, Kog E, Vanderlinden J, eds. The Family Approach
to Eating Disorders: Assessment and Treatment of Anorexia Nervosa and Bulimia.
New York: PMA Publishing, 1989:25–68.
29. Dare C, Le Grange D, Eisler I, Rutherford J. Redefining the psychosomatic family:
Family process of 26 eating disorder families. Int J Eat Disord 1994; 16: 211–226.
30. Le Grange D. Family therapy for adolescent anorexia nervosa. 1999; 55:727–739.
31. Dare C, Eisler I, Russell G, Treasure J, Dodge L. Psychological therapies for adults
with anorexia nervosa: randomised controlled trial of out-patient treatments. Br J
Psychiatry 2001; 178:216–221.
32. Lock J, Le Grange D. Can family-based treatment of anorexia nervosa be
manualized? J Psychother Pract Res 2001; 10:253–261.
33. Lock J, Le Grange D, Agras WS, Dare C. Treatment Manual for Anorexia Nervosa:
A Family Based Approach. New York: Guilford, 2002.
34. Russell GFM, Szmukler GI, Dare C, Eisler I. An evaluation of family therapy in
anorexia nervosa and bulimia nervosa. Arch Gen Psychiatry 1987; 44:1047–1056.
35. Dare C, Eisler I, Russell GFM, Szmukler GI. Family therapy for anorexia nervosa:
implications from the results of a controlled trial of family and individual therapy.
J Marital Fam Ther 1990; 16:39–57.
36. Eisler I, Dare C, Russell GFM, Szmukler GI, Le Grange D, Dodge E. Family and
individual therapy in anorexia nervosa: a five-year follow-up. Arch Gen Psychiatry
1997; 54:1025–1030.
37. Kaplan HI, Sadock BJ. Contributions of the psychosocial sciences to human behavior.
In: Synopsis of Psychiatry: Behavioral Sciences Clinical Psychiatry. 6th ed.
Baltimore, MD: Williams & Wilkins, 1991:104–154.
38. Szmukler GI, Eisler I, Russell GF, Dare C. Anorexia nervosa, parental “expressed
emotion” and dropping out of treatment. Br J Psychiatry 1985; 147:265–271.
39. Le Grange D, Eisler I, Dare C, Hodes M. Family criticism and self-starvation: A
study of expressed emotion. J Fam Ther 1992; 14:177–192.
40. Van Furth EF, Van Strien DC, Martina LM, Van Son MJ, Hendrickx JJ, Van
Engeland H. Expressed emotion and the prediction of outcome in adolescent eating
41. Robin AL, Siegal PT, Koepke T, Moye AW, Tice S. Family therapy versus
individual therapy for adolescent females with anorexia nervosa. J Dev Behavi
Pediat 1994; 15:111–116.
42. Robin AL, Gilroy M, Dennis AB. Treatment of eating disorders in children and
adolescents. Clin Psychol Rev 1998; 18:421–446.
43. Brandes J. Outpatient family therapy for bulimia nervosa. In: Woodside DB,
Shekter-Wolfson L, eds. Family Approaches in Treatment of Eating Disorders.
Washington, DC: American Psychiatric Press, 49–66.
44. Jager B, Liedtke R, Kunsebeck HW, Lempa W, Kersting A, Seide L.
Psychotherapy and bulimia nervosa: evaluation and long-term follow-up of two
conflict-orientated treatment conditions. Acta Psychiatr Scand 1996; 93:268–278.
45. Dodge E, Hodes M, Eisler I, Dare C. Family therapy for bulimia nervosa in
adolescents: an exploratory study. J Fam Ther 1995; 17:59–77.
47. Gorin AA, Le Grange D, Stone AA. Effectiveness of spouse involvement in
cognitive behavioral therapy for binge eating disorder. Int J Eat Disord 2003: 421–
433.
48. Perednia C, Van Vreckem E, Vandereycken W. Parent counseling: From guidance
to treatment. In: Vandereycken W, Kog E, Vanderlinden J, eds. The Family
Approach to Eating Disorders: Assessment and Treament of Anorexia Nervosa and
Bulimia. New York: PMA Publishing, 1989:249–261.
49. Klump K. A genetic link to anorexia. In: DeAngelis T, ed. Monitor on Psychology.
March 2002; 33(3):34–36.
50. Goldner EM, Birminghan CL. Anorexia nervosa: methods of treatment. In:
Alexander-Mott L, Lumsden DB, eds. Understanding Eating Disorders: Anorexia
Nervosa, Bulimia Nervosa, and Obesity. Washington, DC: Taylor & Francis, 1994:
135–157.
51. Gowers S, Norton K, Halek C, Crisp AH. Outcome of outpatient psychotherapy in
a random allocation treatment study of anorexia nervosa. Int J Eat Disord 1994; 15:
165–177.
52. Michel DM. Psychological assessment as a therapeutic intervention in hospitalized
patients with eating disorders. Prof Psychol Res Pract. In press.
53. Casper RC, Troiani M. Family functioning in anorexia nervosa differs by subtype.
Int J Eat Disord 2001; 30:338–342.
54. Powers P. Management of patients with comorbid medical conditions. In: Garner
DM, Garfinkel PE, eds. Handbook of Treatment for Eating Disorders. 2d ed. New
55. Willard SG. Anorexia and Bulimia: The Potential Devastation of Dieting. Plainfield,
NJ: Patient Education Press, 1990.
19
Interpersonal Psychotherapy for Anorexia
Disorder
M.Joy Jacobs
San Diego State University/University of California San Diego
Joint Doctoral Program in Clinical Psychology
San Diego, California, U.S.A.
R.Robinson Welch and Denise E.Wilfley
Washington University School of Medicine
St. Louis, Missouri, U.S.A.
INTRODUCTION
Originally developed by Gerald Klerman and colleagues (1) for the
management of unipolar depression, interpersonal psychotherapy (IPT) is a
brief, time-limited therapy that has been successfully adapted for the
management of eating disorders. First successfully adapted for the management
of bulimia nervosa (BN) (2,3), IPT has since proven effective, via an innovative
group format, for binge eating disorder (BED). Currently, the role of IPT in the
management of anorexia nervosa remains unclear.
THEORETICAL FOUNDATIONS OF
INTERPERSONAL PSYCHOTHERAPY
IPT was initially developed not as a novel therapy, but as an attempt to reflect
interpersonally focused treatment for depression already in practice in the
1970s and 1980s (4,5). IPT is theoretically rooted in theories developed by
Adolf Meyer, Henry Stack Sullivan, and John Bowlby. In the 1950s, Meyer
postulated that psychopathology was rooted in maladjustment to one’s social
environment (6). During the same time period, Henry Stack Sullivan (who was
responsible for popularizing the term “interpersonal”) theorized that a patient’s
interpersonal relationships, rather than intrapsychic processes alone, constituted
the relevant focus of therapeutic attention. Sullivan believed that individuals could
not be understood in isolation from their interpersonal relationships and posited
452 JACOBS ET AL.
that enduring patterns in these relationships could either encourage self-esteem

or result in anxiety, hopelessness, and psychopathology (7). The work of John
Bowlby, specifically his attachment theory, is also associated with IPT. Bowlby
emphasized the importance of early attachment in the later development of
interpersonal relationships and emotional well-being (8). According to Bowlby,
failures in attachment resulted in psychopathology. Incorporating aspects of the
theories posited by Meyer, Sullivan, and Bowlby, IPT acknowledges a two-way
relationship between social functioning and psychopathology; interpersonal
dysfunction results in psychopathology and psychopathology results in a
deterioration in interpersonal functioning.
INTERPERSONAL FUNCTIONING AND EATING

DISORDERS: EMPIRICAL BASIS
IPT for eating disorders is based on compelling evidence that interpersonal
factors have a significant role in the etiology and maintenance of these disorders.
Eating-disordered individuals typically have a history of more frequently
difficult social experiences, including problematic family histories and specific
interpersonal stressors, than non-eating-disordered individuals (9–12). These
individuals also experience a wide range of social problems, including
loneliness, lack of perceived social support, low self-esteem, low social
adjustment, and poor social problem-solving skills (13–20); this combination of
factors may inhibit their ability to cope with interpersonal stressors (21–24).
Furthermore, interpersonal difficulties, low self-esteem, and negative
affect (25–28) may be interconnected and associated with dysfunctional eating
patterns (29). These related factors may then create a vicious cycle, each
exacerbating the other and combining to precipitate and/or maintain eating
disorder symptoms. IPT aims to improve interpersonal functioning, selfesteem,
and negative affect as they relate to each other and to eating disorder
symptoms.
INTERPERSONAL PSYCHOTHERAPY FOR EATING

DISORDERS
Ideally, in each session of IPT for eating disorders, symptoms should be
explicitly and repeatedly linked to problems in interpersonal functioning.
Research versions of IPT for eating disorders to date have avoided an explicit
symptom focus in order to clearly distinguish IPT from cognitive-behavioral
therapy (CBT) in comparison studies. Because of this, research studies
investigating the use of IPT for eating disorders have typically not adequately
addressed specific eating disorder symptoms. In clinical settings, however,
INTERPERSONAL PSYCHOTHERAPY 453
consistent attention to the relationship between eating disorder symptoms and

problems in interpersonal functioning is recommended for maximal therapeutic
impact.
The fundamental structure and techniques of IPT are similar for all three
eating disorders. The essentials concepts of IPT and the specific tasks of each
phase of treatment are described in detail below.
Basic Interpersonal Psychotherapy Concepts
Interpersonal Problem Areas

IPT is designed to help patients identify and address current interpersonal
problems. Treatment focuses on the resolution of problems within four social
domains that are associated with the onset and/or maintenance of the eating
disorder, namely, interpersonal deficits, interpersonal role disputes, role
transitions, and grief (Table 1). Interpersonal deficits apply to those patients who
are socially isolated or who are in chronically unfulfilling relationships. For
patients with this problem area, unsatisfying relationships and/or inadequate
social support are frequently the result of poor social skills. Interpersonal role
disputes are conflicts with a significant other (e.g., a partner, other family
member, coworker, or close friend) that emerge from differences in
expectations about the relationship. Role transitions include difficulties associated
with a change in life status (e.g., graduation, leaving a job, moving, marriage/
divorce, retirement, change in health status). Griefis identified as the problem
area when the onset of the patient’s symptoms is associated with the loss of a
person or a relationship, either recent or past. IPT for eating disorders focuses
on identifying and changing the maladaptive interpersonal context in which the
eating problem has been developed and maintained.
Percentages of problem areas were obtained from psychotherapy studies for
BN and BED (30,31) and from an interview-based study for AN (32). For all
three disorders, grief was uncommon whereas role disputes were fairly
common. The notable differences among the disorders were that inter personal
deficits were more likely to be found among BED and AN patients, and role
transitions were more common among BN patients.
TABLE 1 Interpersonal Problem Areas: Description, Goals, and Strategies (Klerman et al., 1984; Weissman et al., 2000)
454 JACOBS ET AL.
Source: Adapted from Wilfley DE, Stein Rl, Welch RR. Interpersonal psychotherapy for the treatment of eating disorders. In: Treasure, J,
Schmidt U, Dare C, Van Furth E, eds. Handbook of Eating Disorders, 2nd ed. Sussex: John Wiley & Sons, (in press).
456 JACOBS ET AL.
Treatment Structure
IPT for eating disorders typically lasts 15–20 sessions over a 4- to 5-month
period (33,34). The treatment is demarcated by three phases. The initial phase is
dedicated to identifying the problem area that will be the target for treatment.
The intermediate phase is devoted to working on the target problem area(s). The
termination phase is devoted to consolidating gains made during treatment and
preparing patients for future work on their own. Each phase of treatment for
eating disorder patients, along with clinical vignettes illustrating implementation
of the treatment, will be described in detail below.
Therapeutic Stance
Similar to other therapies, IPT places importance on establishing a positive
therapeutic alliance between therapist and patient. Specifically, the IPT
therapeutic stance is one of warmth, support, and empathy. The therapist is
active and advocates for the patient rather than remaining neutral. By phrasing
things positively, the therapist helps the patient feel comfortable and aims to
foster a safe and supportive working environment. Confrontations and
clarifications are offered in a gentle and timely manner, and the therapist is
careful to encourage the patient’s positive expectations of the therapeutic
relationship. In addition, the therapist conveys a hopeful stance and optimistic
attitude about the patient’s ability to recover.
Implementing Interpersonal Psychotherapy for

Eating Disorders
Tasks of the Treatment Phases (Table 2)

The Initial Phase. Sessions 1–5 typically constitute the initial phase of IPT for
eating disorders. After assessing the patient’s current eating disorder symptoms
and obtaining a history of these symptoms, the therapist gives the patient a formal
diagnosis. Therapist and patient then discuss the diagnosis as well as what might
be expected from treatment. As described below, assignment of the sick role
during this phase serves the dual function of granting the patient the permission
to recover, as well as the responsibility to recover. The therapist explains the
rationale of IPT, underscoring that therapy will focus on the identifying and
altering dysfunctional interpersonal patterns related to eating disorder
symptomatology. As discussed below, in order to determine the precise focus
of treatment, the therapist conducts an interpersonal inventory with the patient
and develops an interpersonal formulation based on this. In the interpersonal
TABLE 2 Therapist Tasks for Interpersonal Psychotherapy
Source: Adapted from Wilfley DE, Stein Rl, Welch RR. Interpersonal psychotherapy for
the treatment of eating disorders. In: Treasure, J, Schmidt U, Dare C, Van Furth E, eds.
Handbook of Eating Disorders, 2nd ed. Sussex: John Wiley & Sons, 2003; 253–270.
formulation, the therapist links the patient’s eating disorder to one of the four
interpersonal problem areas. The patient’s concurrence with the therapist’s
identification of the problem area and agreement to work on this area are essential
before beginning the intermediate treatment phase.
DIAGNOSIS AND ASSIGNING THE SICK ROLE. After a thorough
psychiatric review has been conducted, the patient is formally diagnosed with an
eating disorder and assigned the “sick role.” The purposes of assigning the sick
role are both theoretical and practical. Consistent with the medical model,
receiving a formal diagnosis reinforces the idea that the patient has a known
condition that can be managed. Accurate diagnosis is essential to effective
treatment. The giving of a diagnosis also explicitly identifies the patient as in
need of help. The sick role is assigned not to condescend to the patient but
rather to temporarily exempt the individual from other responsibilities in order
to devote full attention to recovery. This is particularly important for eating-
disordered patients, many of whom tend to set aside their own needs and
desires in order to care for and please others. If this applies in a particular case,
the therapist may explicitly highlight the patient’s excessive caretaking
458 JACOBS ET AL.
tendencies and encourage the patient to redirect this energy from others to her
own recovery. In doing so, the therapist clarifies the rationale behind IPT—that
by improving the patient’s patterns of interpersonal functioning, the patient’s
eating disorder symptoms are expected to improve as well.
THE INTERPERSONAL INVENTORY. At the beginning of IPT, an
interpersonal inventory examining the patient’s interpersonal history is
conducted. The interpersonal inventory may take 1–3 sessions to complete. A
thorough interpersonal inventory is essential for adequate case formulation and
development of an optimal treatment plan. The clinical importance of investing
the time involved in conducting a comprehensive interpersonal inventory
cannot be overemphasized; accurate identification of the patient’s primary
problem area can be difficult and is key to success in therapy.
Included in the interpersonal inventory is a review of the patient’s current
close relationships, current social functioning, relationship patterns, and
expectations. Changes in interpersonal relationships are explored and discussed
with reference to the onset and maintenance of eating disorder symptoms. For
each person who is important in the patient’s life, the following information is
assessed: frequency of contact, activities shared, satisfactory and unsatisfactory
aspects of the relationship, and ways that the patient would like to change the
relationship. The therapist obtains a chronological history of significant life
events, fluctuations in mood and self-esteem, interpersonal relationships, and
eating disorder symptoms. From this review, the therapist can work with the
patient to make connections between certain life experiences and eating disorder
symptoms. Thorough exploration of this interrelationship typically helps
patients to more clearly understand the rationale behind IPT. Upon completion
of the interpersonal inventory, the therapist should have helped the patient
identify a primary interpersonal problem area(s).
THE INTERPERSONAL FORMULATION. Upon completion of the
interpersonal inventory, the therapist should have developed an individualized
interpersonal formulation, including identification of the patient’s primary
problem area. Although some patients may present for treatment with
difficulties in several problem areas, the time-limited nature of the treatment
necessitates a focused approach. The therapist, with the agreement of the
patient, should assign one, or at most two, problem area(s) upon which to
develop a treatment plan. The goals developed at this stage will be referenced at
each future session and will guide the day-to-day work of the treatment. For
examples of individual case formulations, the reader is referred to (35). If more
than one problem area is identified, the patient may choose to work
simultaneously on both or may decide to first address the problem area
that seems most likely to be responsive to treatment. For example, when a
patient has role disputes and interpersonal deficits, clinical attention might first
be focused on role disputes, since interpersonal deficits reflect long-term

patterns that may require considerably more time and effort to change. Once
the role dispute has been resolved, therapist and patient would then decide how
to best address the more entrenched interpersonal deficits.
The Intermediate Phase. Once the patient and therapist have agreed on the
primary problem area and have set treatment goals, the intermediate phase of
treatment begins. An essential task throughout the intermediate phase is to
strengthen the connections patients make between difficulties in their
interpersonal lives and eating problems. The intermediate phase typically lasts a
total of 8–10 sessions and constitutes the “work” of the therapy. The strategies
and goals of the intermediate phase are shaped by the primary problem area
targeted in the treatment.
THERAPEUTIC GOALS AND STRATEGIES BY PROBLEM AREA. The
eating disorders therapist should also implement specific treatment strategies
based on the identified problem area. These are discussed below.
Interpersonal Deficits. If interpersonal deficits is the primary problem area, as is
frequently the case with BED and AN patients, treatment strategies should be
utilized in order to reduce the patient’s social isolation, to enhance the quality
of existing relationships, and to encourage the formation of new relationships. It
is crucial to determine why the patient is having difficulty forming and/or
maintaining relationships. Conducting a review of past significant relationships
will be particularly useful in making this assessment. During this review,
attention should be given to both the positive and negative aspects of the
relationships, as well as an investigation of potentially recurrent patterns in
these relationships. If appropriate, the therapist should then relate the
problematic patterns occurring in the patient’s present relationships to
relationship patterns already known to be problematic. Use of the therapeutic
relationship (discussed in greater detail in “General Therapeutic Techniques,”
below) is also helpful in the treatment of patients with interpersonal deficits.
Use of this technique provides an illustration of interpersonal patterns that may
be the source of difficulties in other relationships and provides the patient
helpful feedback on his or her interactive style.
Interpersonal Role Disputes. Interpersonal role disputes occur frequently in
eating disorder patients and may be especially common among BN
patients (11). Patients who are particularly affected by this problem area
are typically involved in conflicts with a significant other (i.e., spouse, other
family member, close friend, or coworker). The goals of treatment include
clearly identifying the nature of the dispute and exploring options for its
resolution. Important in this will be making a determination regarding the stage
of the dispute. Once the stage of the dispute becomes clear, it may be important
to modify the patient’s expectations and remedy faulty communication in order
460 JACOBS ET AL.
to bring about adequate resolution. Particularly helpful in leading to a resolution

may be an exploration of how nonreciprocal role expectations relate to the
dispute. If resolution is not possible, it will be important to encourage the
patient to either reassess expectations of the relationship or to consider
dissolving the relationship and mourning its loss.
Role Transitions. Role transitions typically involve major life cycle changes that
effect an important aspect of the patient’s self-identification. Common role
transitions include a career change (i.e., promotion, firing, retirement), a
family change (marriage, divorce, birth of a child), the beginning or end of an
important relationship, a move, graduation, or diagnosis of a medical illness.
The goals of therapy include mourning and accepting loss of the old role,
recognizing the positive and negative aspects of both the old and new roles, and
restoring the patient’s self-esteem by having him or her develop a sense of
mastery in the new role. Key strategies in achieving these goals include a thorough
exploration of the patient’s feelings related to the role change and encouraging
the patient to develop new skills and adequate social support for the new role.
Grief. Grief is the least common primary problem area among eating disorder
patients. Grief is most obviously identified as the problem area when the onset
of the patient’s symptoms is associated with the death of a loved one, recent or
past. However, grief may not be limited to the physical death of a loved one.
Grief can also result from the loss of a significant relationship or the loss of an
important aspect of one’s identity. Compared to the other problem areas, grief
can be resolved relatively quickly. Goals for managing a complicated
bereavement include facilitating the mourning process and helping the patient to
identify new relationships and activities to compensate for the loss. Patients
must be educated about the grieving process and be encouraged to explore all
feelings that arise. As well as profound sadness, grief may evoke feelings of
anger toward the deceased and subsequent guilt for this anger; patients must be
encouraged to explore these emotions as well. During the grieving process,
patients frequently idealize what has been lost; a thorough reconstruction of
both the positive and negative aspects of what has been lost may help the patient
to develop a more balanced view. As patients become less focused on the past,
they should be encouraged to consider new ways of increasing their
involvement with others and to establish new interests.
Therapeutic Techniques.
FOCUSING ON GOALS. IPT is a directed, goal-oriented therapy. Thus, it
is important that the therapist maintain a focus each week on how the patient is
working on his or her goals between sessions. Phrases such as “moving forward
on your goals” and “making important changes” are used to encourage patients
to be responsible for their treatment while reminding them that altering

interpersonal patterns requires attention and persistence. In session, unfocused
discussions are redirected to the key interpersonal issues. By explicitly
addressing goals each week, the patient can begin working toward necessary
changes. This goal-oriented focus has been supported by research on IPT
maintenance treatment for recurrent depression, which has demonstrated that
the therapist’s ability to maintain focus on interpersonal themes is associated
with better outcomes (36).
In the following vignette, note how the therapist initiates the discussion
about goals and helps a patient in treatment for BN with interpersonal deficits
work on her goals*:
Therapist: Victoria, now that we have just started this middle phase of
therapy, I wanted to check in with you to see how your work is
coming on your goals. You mentioned last week that you are
starting to become more aware of interpersonal triggers of bingeing
and purging.
Victoria: This week I started paying more attention to what is happening as
I’m having the urge to binge. It’s a little overwhelming, since so
much seems to be going on—a fight with my mother, missing my
boyfriend, feeling stressed out about school. I’m starting to get a
better sense of what’s happening with me when I get the urge to
binge, but I don’t know what to do with that information once I
have it!
Therapist: That’s a great start in identifying feelings that become triggers,
Victoria. Clearly a lot of things are playing into your desire to
binge. How were you able to become more aware of what was
happening with you when you felt the urge to binge?
Victoria: Well, I’m used to just reacting, to just giving in to the urge to binge
as soon as I feel it—just like we talked about in our first ses sions.
Now I’m aware that a switch seems to go off inside of me when I
have the urge to binge. Stopping and checking in with myself about
what I’m feeling slowed down the frenzy that usually characterizes
my binges. Although I didn’t avoid binge episodes completely this
week, I can identify at least one or two occasions where I lost the
desire to binge once I took a breather and checked in with myself.
Therapist: What specifically did you notice was happening with you?
* Please note that in order to most clearly illustrate the therapeutic principles discussed
in the text, this and the other clinical vignettes included below appear to offer more
rapid results than may typically be expected clinically.
462 JACOBS ET AL.
Victoria: I noticed how difficult it is for me to confront my mother and how

dependent I feel on my boyfriend being around to compliment me
for my self-esteem. I realized that sometimes I’m not
communicating what I want or need from these relationships. I
don’t feel like I know how to do this and end up expressing—or is
it suppressing?—my frustration through food.
Therapist: Good work, Victoria. You’ve discovered some really important
clues to your interpersonal triggers. Now that you’re becoming
more aware of the circumstances surrounding your urge to binge,
we can begin to work on helping you find more effective ways to
manage your feelings and relationships. As we discussed earlier,
this will help you eliminate your binge eating.
Victoria: That sounds great.
As the above vignette illustrates, critical to IPT for eating disorders is helping
to facilitate and strengthen the connections patients make between their
problematic eating and difficulties in their interpersonal lives. Focusing on
specific goals provides a structure in which to do this.
MAKING CONNECTIONS. A crucial task of the intermediate phase is
helping patients recognize the connections between eating difficulties and
interpersonal events during the week. As patients learn to make these
connections and develop strategies to alter the interpersonal context in which
the eating behavior occurs, the cycle of the eating disorder is interrupted. In the
following vignette, also in the intermediate phase, the therapist encourages a
patient with interpersonal role disputes in treatment for AN to talk about the
connections she has made between her desire to restrict her food intake and
difficulties she has with her alcoholic father:
Therapist: How has the week gone with your goals, Anna?
Anna: What I’ve noticed is that whenever my dad is drinking, I just want
to starve myself. I get busy making sure his bills are paid and that
his other things get done, and then I want to restrict more than
ever! I sense that this is my way of shutting down—focusing on
not eating is a way for me to block out all the stuff that’s happening
with my dad. I never realized that connection until recently. I just
focused on wanting to be thin.
Therapist: This is great work, Anna! One of the things we’ve been working on
is getting you to become more aware of what’s happening around
you when you feel the desire to restrict most intensely. You’ve just
made an important connection between your stress and fear related
to your father’s behavior and your wanting to restrict. Now that
you can more clearly see that connection, how would you like to
start working on your relationship with your father?
REDIRECTING ISSUES RELATED TO EATING, SHAPE, AND/OR
WEIGHT. Eating disorder patients may frequently bring up issues in session
relating to distressing eating behavior (e.g., binge episodes) or may want to
engage in extended discussion relating to eating, shape, and weight. Although
these issues are relevant insofar as they reflect the clinical status of the patient’s
eating disorder, the therapist must be attentive to keep the session “on track,”
i.e, focused on the patient’s treatment goals. In such situations, the therapist
should gently but firmly redirect the patient back to work on the treatment
goals. As discussed above, dialogue related to eating disorder symptomatology
must be repeatedly linked to its functional role in the interpersonal domain. The
following example illustrates how a therapist can redirect discussion and help
the patient focus on the treatment goals:
Therapist: What did you want to work on today, Jim?
Jim: I’ve been really stressed out lately. I have two exams this week and
a research paper due. I didn’t have much time to eat, which was
good, but I had a huge binge on Saturday night. I found myself in
the kitchen gorging on chips, cookies, pizza, basically anything that
didn’t move. I was so embarrassed when my roommate walked in
and saw what I was doing. I probably would still be eating if he
hadn’t walked in.
Therapist: One of the things you shared with me last week was that eating is a
way for you to relieve stress, to unwind. Instead of allowing
yourself a break or sharing your feelings with friends, you’ll turn to
food.
Jim: I definitely did that. You wouldn’t believe how I was shoving food
in by the mouthful, I couldn’t eat fast enough, I wanted…
Therapist: Jim, let me refocus you for a moment, back to your goals. How
have your efforts to reach out to other students been coming?
Jim: Pretty good. I’ve started studying in the library, so that I can hang
out in the student lounge when I need a break. I’ve met a few
people that way. It’s hard for me though because I always feel
insecure in social situations. I feel like other people may be judging
me or don’t like me.
Therapist: As we discussed last week, for a long time you’ve been sheltering
yourself from that insecurity by avoiding others and using food to
cope with stress. I wonder if as you practice being more social and
build up your support system, you’ll start to feel more comfortable
socially. You’ll have more friends to turn to when you’re stressed
464 JACOBS ET AL.
and be less likely to turn to food. In the short time we’ve been
working together, you’ve already met several new people. Not
surprisingly, you only had one overeating episode this past week.
You said earlier that you were bingeing nightly during similarly
stressful times in the past.
The above discussed therapeutic techniques (i.e., focusing on goals;
connecting eating symptoms and interpersonal problems; redirecting eating,
weight, or shape issues) are utilized with patients in all four problem areas.
By the end of the intermediate phase, patients are often acutely aware that
treatment will soon be ending. The therapist should begin to discuss termination
explicitly and address any anxiety the patient may be experiencing regarding it.
The therapist should begin to prepare the patient for emotions that may arise
with termination, including grief related to the ending of treatment.
The Termination Phase. The termination phase typically lasts four to five
sessions. During this phase, the therapist should encourage the patient to reflect
on progress made during therapy and to outline goals for remaining work.
Patients are encouraged to identify early warning signs of relapse (e.g.,
overeating, restricting, negative mood) and to identify potential plans of action.
Patients are reminded that eating disorder symptoms tend to arise in times of
difficulty and are encouraged to view such symptoms as important early
warning signals. Identifying potential plans of action in such situations will serve
to increase the patient’s sense of competence and security. Nevertheless, it is
also essential to assist patients in identifying warning signs and symptoms that
may indicate a need for professional intervention in the future.
General Therapeutic Techniques

Throughout therapy, the therapist should maintain a consistent focus on the
interpersonal context of the patient’s life. Eating disorder symptoms should
thus be analyzed in light of the interpersonal difficulties constituting the
backdrop of those symptoms. Although this approach is unique to IPT, some of
the therapeutic techniques utilized in IPT are similar to techniques used in other
therapies. In addition to the therapeutic techniques described above, other
helpful techniques include exploratory questions, encouragement of affect,
clarification, communication analysis, and use of the therapeutic relationship
[for a more detailed description of these and other techniques, please see (1)].
Exploratory Questions. Use of general, open-ended questions can facilitate the
free discussion of material. This is particularly true in the initial phases of a
session. For example, the therapist might say, “Tell me about your relationship
with your husband.” Once this has generated discussion, progressively more
specific questioning would ensue.
Encouraging Affect. IPT’s focus on affect evocation and exploration is especially
relevant for eating disorder patients, given that problematic eating often
functions as a way to regulate negative affect. Specifically, the IPT therapist
helps patients: (a) acknowledge and accept painful affects, (b) use affective
experiences to bring about desired interpersonal changes, and (c) experience
suppressed affects.
Frequently, eating-disordered patients are emotionally constricted in
situations when others would typically experience strong emotions. For
example, sometimes patients will deny feeling upset when it is clear that an
upsetting interaction has just occurred. In this situation, the therapist might say,
“Although you said you were not upset, it appears to me that you have shut down
since mentioning the situation with your husband.” By explicitly noting the
discrepancy, the therapist will attempt to draw out affect that has been
suppressed.
Clarification. This technique is useful in calling attention to contradictions that
may have occurred in the patient’s presentation of material and increases the
patient’s awareness about what she or he has actually communicated. For
example, contradictions between the patient’s affect and speech may be
noteworthy (i.e., “While you were telling me how upset you are about your
father, you had a smile on your face. What do you think that’s about?”).
Communication Analysis. This technique is used to identify any communication
difficulties that the patient may be experiencing and to help the patient alter
ineffective communication patterns. Typically, the therapist will ask the patient
to recall in great detail a recent interaction or argument with a significant other.
Together, patient and therapist work to identify any communication difficulties
and to find more effective communication strategies.
Use of the Therapeutic Relationship. The premise behind this technique is that
people have characteristic patterns of interacting with others. The technique is
utilized by exploring the patient’s thoughts, feelings, expectations, and behavior
in the therapeutic relationship and relating these to the patient’s characteristic way
of behaving and/or feeling in other relationships. This technique is particularly
relevant and useful for patients with interpersonal deficits and interpersonal role
disputes. Use of this technique offers the patient the opportunity to understand
the nature of his or her difficulties in interacting with others and provides the
patient with helpful feedback on his or her interaction style.
466 JACOBS ET AL.
REVIEW OF OUTCOME STUDIES

IPT has been adapted for the treatment of BN, BED, and AN, respectively.
Although IPT as tested for major depression includes a consistent focus on
depressive symptomatology, research applications of IPT for eating disorders
have not included a strong focus on eating disorder symptoms. This focus has
been avoided in order to minimize procedural overlap with cognitivebehavioral
therapy (CBT). However, clinical experience strongly suggests that significant
therapeutic benefit is derived from consistent attention to the role of specific
eating disorder symptoms as “red flags,” signaling interpersonal difficulties in
need of clinical attention.
Interpersonal Psychotherapy for Bulimia Nervosa

CBT is currently the most researched, best established treatment for BN (37).
To date, IPT is the only psychological treatment for BN that has demonstrated
long-term outcomes comparable to those of CBT. Thus far, all controlled
studies of IPT have been comparison studies with CBT. Early studies indicated
similar short- and long-term outcomes for binge eating between CBT and
IPT (3,38). A more recent multisite study (39) comparing CBT and IPT as
treatments for BN found that in the short term (posttreatment) patients
receiving CBT demonstrated higher rates of abstinence from binge eating and
lower rates of purging. By long-term follow-up (8 months and 1 year following
treatment), these rates were equivalent. The more rapid effects of CBT
compared to IPT may be at least partially explained by the relative lack of focus
on eating disorder symptomatology in the research version of IPT. Despite
relatively slower response rates, however, IPT patients rated their treatment as
more suitable and expected greater success than did CBT patients. This finding
suggests that BN patients may perceive the interpersonal focus of IPT as more
relevant to their disorder and to their treatment needs than a more cognitive-
behavioral focus on distortions related to weight and shape.
When choosing between IPT and CBT for an individual patient, the therapist
and patient should together evaluate the advantages and disadvantages of each
treatment. To date, more data support the efficacy of CBT. CBT has been
shown to produce more rapid effects for BN, although IPT produces equivalent
outcomes over the long term. IPT may be particularly well suited to patients
presenting with interpersonal difficulties or for patients who express a distaste
for elements of CBT (i.e., keeping food diaries, thought recording). Finally,
therapist expertise is relevant to the choice of treatment; both IPT and CBT are
specialty treatments and should be administered only by trained practitioners.
IPT for Binge Eating Disorder

The use of IPT for BED was based on the earlier success of IPT in BN (40).
Wilfley and colleagues first adapted IPT to a group format for BED
patients (41,42). New strategies were developed to specifically address
interpersonal deficits, an interpersonal problem area that occurs more
frequently in BED patients than BN patients. In IPT for BED, group members with
interpersonal deficits are encouraged to use the group as a “live” social network.
This social milieu is designed to decrease social isolation, support the formation
of new social relationships, and serve as a model for initiating and sustaining
social relationships outside of the therapeutic context (33). Self-stigmatization is
common among BED patients; this stigmatization contributes to the
maintenance of the disorder. Group therapy thus offers a radically altered social
environment for BED patients, who typically endeavor to keep shameful eating
behaviors hidden from others.
Similar to BN, CBT has been shown to have specific and robust treatment
effects in BED (42–49). Two studies have compared IPT with CBT and found
that IPT has comparable effects to CBT in the management of BED. The first
study, comparing group CBT and IPT, revealed that both treatments were more
effective than a wait-list control group at reducing binge eating and had
equivalent significant reductions in binge eating in both the short and long
term (42). In a second study, which included a substantially larger sample size,
both CBT and IPT demonstrated equivalent short- and long-term efficacy in
reducing binge eating and associated specific and general psychopathology, with
approximately 60% of patients remaining abstinent from binge eating at 1-year
follow-up (31). The time course of almost all outcomes with IPT was identical
to that of CBT. The one treatment-specific difference was a significantly greater
reduction in dietary restraint among CBT patients at posttreatment and
4-month follow-up; however, IPT demonstrated parity with CBT by the 8- and
12-month follow-ups. Findings from the two aforementioned studies indicate
that IPT represents an efficacious treatment alter native for BED. Similar to BN,
considerations relevant to the choice among treatment options for a given BED
patient include the individual patient’s symptom profile and preferences, as
well as therapist expertise.
Interpersonal Psychotherapy for Anorexia Nervosa

Given the dearth of randomized controlled treatment trials studying AN, some
have argued that it may be impossible to generate empirically based
recommendations regarding treatment (37). To date, family therapy has been
the most studied treatment for AN (37). Specifically, family therapy has been
468 JACOBS ET AL.
found to be efficacious for adolescents with a short duration of illness (50,51).

However, the often chronic nature of AN, particularly in adult patients, limits
the generalizability of these findings.
No controlled studies have yet demonstrated the efficacy of IPT for AN. In
the first trial to compare IPT, CBT, and nonspecific supportive clinical
management (NSCM) for AN, McIntosh and colleagues (52) found that NSCM
was superior to both IPT and CBT for improving the core features of AN. As
developed and manualized for this study, NSCM was delivered by highly trained
eating disorder specialists; the treatment focused on the patient’s presenting
eating disorder symptoms. Similar to other treatment trials using IPT for eating
disorders, the IPT condition in this study did not include a focus on eating
disorder symptomatology. However, McIntosh and colleagues (32) advise, that
optimal delivery of IPT with anorexic patients in clinical settings should include
an ongoing review of the connections between the interpersonal problem areas
of interest and core anorexic symptoms. Given the ego syntonic nature of AN,
the relative lack of focus on eating disorder symptoms in this research trial may
have blunted IPT’s effect and avoided the essential work of therapy (52).
Alternatively, it may be that the supportive, nondirective approach offered in
NSCM is superior to a specialized psychotherapy (52), particularly in the weight
regain phase of AN treatment.
Nevertheless, given the importance of interpersonal functioning in etiological
theories of AN (32), it makes sense to continue to explore IPT’s utility in
treatment of the disorder. Specifically, investigation of the clinical effectiveness
of IPT for AN that includes a focus on eating disorder symptoms as they relate
to interpersonal problems is warranted. It may be that IPT for AN is optimally
delivered in the context of other adjunctive treatments (e.g., pharmacological,
nutritional) rather than as a “stand-alone” treatment. Staging of treatment may
also be important; perhaps IPT is more suitable for the maintenance and relapse
prevention stages of treatment than for the weight regain phase. Alternatively, a
combination therapy including components of both IPT and CBT may provide a
more efficacious treatment than IPT or CBT alone. Given the state of current
research, however, no definitive clinical recommendations regarding IPT for AN
can be made.
SUMMARY AND FUTURE DIRECTIONS

IPT for eating disorders is a focused, time-limited treatment that targets
interpersonal problem(s) associated with the onset and/or maintenance of the
eating disorder. The interpersonal focus is highly relevant to eating disorder
patients, many of whom experience difficulties in interpersonal functioning.
Depending on the patient’s primary problem area, specific treatment strategies
and goals are incorporated into the treatment plan. The primary problem area is
determined by conducting a thorough interpersonal inventory and devising an
individualized interpersonal formulation for each patient. IPT has resulted in
significant and well-maintained improvements for the management of BN and
BED; its role in the management of AN has yet to be determined.
The interpersonal focus of IPT has traditionally been more readily embraced
by clinicians than CBT; IPT may also be an easier therapy in which to become
proficient. In the future, the efficacy of IPT in eating-disordered populations
may be further enhanced by including a specific focus on eating disorder
symptomatology in the treatment. In addition, further adaptations of the group
format, which has demonstrated promise with BED patients, may be usefully
extended to other patient subpopulations who would likely benefit from the
support of a group modality (i.e., adolescents). Additional research is needed
regarding the mechanisms by which IPT achieves its effects, predictors of
treatment outcome, and the effectiveness of IPT for eating disorders in clinical
settings. Finally, advances in neuroscience, specifically increased sophistication
in neuroimaging techniques, offer the potential to investigate the impact of IPT
on brain function in eating disorder patients. Such studies have revealed changes
in brain function in depressed patients treated with IPT (53,54); similar studies
of IPT for eating disorders may provide useful information regarding
physiological mechanisms influencing treatment response. Such empirically
based refinements of the content and delivery of IPT may thus further enhance
its clinical utility in the management of eating disorders.
REFERENCES
1. Klerman GL, Weissman MM, Rounsaville BJ, Chevron ES. Interpersonal
Psychotherapy of Depression. New York: Basic Books, 1984.
2. Fairburn CG, Jones R, Peveler RC, Carr SJ, Solomon RA, O’Connor ME, Burton
J, Hope A. Three psychological treatments for bulimia nervosa: a comparative
trial. Arch Gen Psychiatry 1991; 48:463–469.
3. Fairburn CG, Peveler RC, Jones R, Hope RA, O’Connor ME. Predictors of
12-month outcome in bulimia nervosa and the influence of attitudes to shape and
weight. J Consult Clinical Psychol 1993; 61:696–698.
4. Klerman GL, Weissman MM. Interpersonal psychotherapy for depression:
background and concepts. In: Klerman GL, Weissman MM, eds. New Applications
of Interpersonal Psychotherapy. Washington, DC: American Psychiatric
Association, 1993:3–26.
5. Frank E, Spanier C. Interpersonal psychotherapy for depression: overview, clinical
efficacy, and future directions. Clin Psychol Sci Pract 1995; 2:349–369.
6. Meyer A. Psychobiology: A Science of Man. Springfield, IL: Charles C Thomas,
1957.
470 JACOBS ET AL.
7. Sullivan HS. The Interpersonal Theory of Psychiatry. New York: WW Norton,

1953.
8. Bowlby J. Attachment and Loss. Vol. 1. Attachment. New York: Basic Books,
1982.
9. Schmidt U, Tiller J, Blanchard M, Andrews B, Treasure JL. Is there a specific
trauma precipitating anorexia nervosa? Psychol Med 1997; 27:523–530.
for binge eating disorder: a community-based, case-control study. Arch Gen
Psychiatry 1998; 55:425–432.
11. Fairburn CG, Welch SA, Doll HA, Davies BA, O’Connor ME. Risk factors for
bulimia nervosa. Arch Gen Psychiatry 1997; 54:509–517.
12. Segrin C. Interpersonal Processes in Psychological Problems. New York: Guilford
Press, 2001.
13. Gual P, Perez-Gaspar M, Martinez-Gonzalez MA, Lahortiga F, Irala-Estevez J,
Cervera-Enguix. Self-esteem, personality, and eating disorders: baseline
assessment of a prospective population-based cohort. Int J Eat Disord 2002; 31:
261–273.
14. Herzog DB, Keller MB, Lavori PW, Ott IL. Social impairment in bulimia. Int J Eat
Disord 1987; 6:741–747.
15. O’Mahony JF, Hollwey S. The correlates of binge eating in two nonpatient
samples. Addict Behav 1995; 20:471–480.
16. Ghaderi A, Scott B. Prevalence and psychological correlates of eating disorders
among females aged 18–30 years in the general population. Acta Psychiatr Scand
1999; 99:261–266.
17. Rorty M, Yager J, Buckwalter JG, Rossotto E. Social support, social adjustment,
and recovery status in bulimia nervosa. Int J Eat Disord 1999; 26:1–12.
18. Troop NA, Holbrey A, Trowler R, Treasure JL. Ways of coping in women with
eating disorders. J Nerv Ment Dis 1994; 182:535–540.
19. Grissett NL, Norvell NK. Perceived social support, social skills, and quality of
relationships in bulimic women. J Consult Clin Psychol 1992; 60:293–299.
20. Tiller JM, Sloane G, Schmidt U, Troop N, Power M, Treasure JL. Social sup port
in patients with anorexia nervosa and bulimia nervosa. Int J Eat Disord 1997; 21:
31–38.
21. Tanofsky-Kraff M, Wilfley DE, Spurell E. Impact of interpersonal and egorelated
stress on restrained eaters. Int J Eat Disord 2000; 27:411–418.
22. Tuschen-Caffier B, Vogele C. Psychological and physiological reactivity to stress:
an experimental study on bulimic patients, restrained eaters and controls.
23. Telch CF, Agras WS. Obesity, binge eating and psychopathology: are they related?
24. Garner DM, Vitousek KM, Pike KM. Cognitive-behavioral therapy for anorexia
nervosa. In: Garner DM, Garfinkel PE, eds. Handbook of Treatment for Eating
Disorders. 2nd ed. New York: Guilfor Press, 1997:94–144.
25. Stice E, Akutagawa D, Gaggar A, Agras WS. Negative affect moderates the relation
between dieting and binge eating. Int J Eating Disord 2000; 27:218–229.
26. Telch CF, Agras WS. Do emotional states influence binge eating in the obese? Int J
Eat Disord 1996; 20:271–279.
27. Powell AL, Thelen MH. Emotions and cognitions associated with bingeing and
weight control behavior in bulimia. J Psychosom Res 1996; 40:317–328.
28. Schupak-Neuberg E, Nemeroff CJ. Disturbances in identity and self-regulation in
bulimia nervosa: implications for a metaphorical perspective of “body as self.” Int J
Eat Disord 1993; 13:335–347.
29. Steiger H, Gauvin L, Jabalpurwala S, Seguin JR, Stotland S. Hypersensitivity to
social interactions in bulimic syndromes: relationship to binge eating. J Consult
Clin Psychol 1999; 67:765–775.
30. Fairburn CG, Peveler RC, Jones R, Hope RA, O’Connor ME. Predictors of
12-month outcome in bulimia nervosa and the influence of attitudes to shape and
weight. J Consult Clinical Psychol 1993; 61:696–698.
31. Wilfley DE, Welch RR, Stein RI, Spurrell EB, Cohen LR, Saelens BE, Dounchis JZ,
Frank MA, Wiseman CV, Matt GE. The psychological treatment of binge eating
disorder (BED): a comparison group of cognitive behavioral therapy and
interpersonal psychotherapy. Arch Gen Psychiatry 2002; 59:713–721.
32. McIntosh VV, Bulik CM, McKenzie JM, Luty SE, Jordan J. Interpersonal
psychotherapy for anorexia nervosa. Int J Eat Disord 2000; 27:125–139.
33. Wilfley DE, Frank MA, Welch R, Spurrell EB, Rounseville BJ. Adapting
interpersonal psychotherapy to a group format (IPT-G) for binge eating disorder:
toward a model for adapting empirically supported treatments. Psychother Res
1998; 8:379–391.
for binge eating disorder: a community-based, case-control study. Arch Gen
Psychiatry 1998; 55:425–432.
35. Wilfley DE, Stein RI, Welch RR. Interpersonal psychotherapy for the treatment of
eating disorders. In: Treasure J, Schmidt U, Dare C, Van Furth E, eds. Handbook
of Eating Disorders. 2d ed. Sussex: John Wiley & Sons, 2003; 253–270.
36. Frank E, Kupfer DJ, Wagner EF, McEachran AB, Cornes C. Efficacy of
interpersonal psychotherapy as a maintenance treatment of recurrent depression:
contributing factors. Arch Gen Psychiatry 1991; 48:1053–1059.
37. Wilson GT, Fairburn CG Eating disorders. In: Nathan PE, Gorman JM, eds.
Treatments That Work. 2d ed. New York: Oxford University Press, 2002.
38. Fairburn CG, Norman PA, Welch SL, O’Connor ME, Doll HA, Peveler RC. A
prospective study of outcome in bulimia nervosa and the long-term effects of three
psychological treatments. Arch Gen Psychiatry 1995; 52:304–312.
39. Agras WS, Walsh BT, Fairburn CG, Wilson GT, Kraemer HC. A multicenter
comparison of cognitive-behavioral therapy and interpersonal psychotherapy for
bulimia nervosa. Arch Gen Psychiatry 2000; 57:459–466.
472 JACOBS ET AL.
40. Fairburn CG, Jones R, Peveler RC, Carr SJ, Solomon RA, O’Connor ME, Burton
J, Hope A. Three psychological treatments for bulimia nervosa: a comparative
41. Wilfley DE, MacKenzie KR, Welch RR, Ayres VE, Weissman MM. Interpersonal
Psychotherapy for Group. New York: Basic Books, 2000.
42. Wilfley DE, Agras WS, Telch CF, Rossiter EM, Schneider JA, Cole AG, Sifford L,
Raeburn SD. Group cognitive-behavioral therapy and group interpersonal
psychotherapy for the nonpurging bulimic individual: a controlled comparison.
43. Grilo CM, Masheb RM, Heninger G, Wilson GT. Controlled comparison of
cognitive behavior therapy and fluoxetine for binge eating disorder. Paper
presented at the Academy for Eating Disorders International Conference on Eating
Disorders, Boston, MA, April 2002.
44. Devlin MJ. Psychotherapy and medication for binge eating disorder. Paper
45. Ricca V, Mannucci E, Mezzani B, et al. Fluoxetine and fluvoxamine combined with
individual cognitive-behavior therapy in binge eating disorder: a one-year follow-up
study. Psychother Psychosom 2001; 70:298–306.
46. Nauta H, Hospers H, Kok G, Jansen A. A comparison between a cognitive and a
behavioral treatment for obese binge eaters and obese non-binge eaters. Behavior
Therapy 2000; 21:441–461.
47. Kenardy J, Mensch M, Bowen K, Green B, Walton J. Group therapy for binge
eating in type 2 diabetes: A randomized trial. Diabet Med March 2002; 19(3):
234–239.
48. Marcus MD, Wing RR, Fairburn CG. Cognitive behavioral treatment of binge
eating vs. behavioral weigh control on the treatment of binge eating disorder. Ann
Behav Med 1995; 17:SO90.
49. Telch CF, Agras WS, Rossiter EM, Wilfley DE, Kenardy J. Group
cognitivebehavioral treatment for the non-purging bulimic: an initial evaluation.
50. Eisler I, Dare C, Russell GF, Szmukler GI, Le Grange D, Dodge E. Family and
individual therapy in anorexia nervosa: a 5-year follow-up. Arch Gen Psychiatry
1997; 54:1025–1030.
51. Eisler I, Dare C, Hodes M, Dodge E, Russell G, Le Grange D. Family therapy for
adolescent anorexia nervosa: the results of a controlled comparison of two family
interventions. J Child Psychol Psychiatry 2000; 41:727–736.
52. McIntosh VV, Jordan J, Carter FA, Luty SE, McKenzie JM, Bulik CM, Joyce PR.
Three psychotherapies for anorexia nervosa: a randomized controlled trial. Paper
53. Brody AL, Saxena S, Stoessel P, et al. Regional brain metabolic changes in patients
with major depression treated with either paroxetine or interpersonal therapy.
54. Martin SD, Martin E, Rai SR, et al. Brain blood flow changes in depressed patients
treated with interpersonal psychotherapy or venlafaxine hydrochloride. Arch Gen
Psychiatry 2001; 58:641–648.
474 JACOBS ET AL.
20
Use of Dialectical Behavior Therapy in the
Eating Disorders
Marsha D.Marcus and Michele D.Levine
Western Psychiatric Institute and Clinic, University of Pittsburgh
School of Medicine
Dialectical behavior therapy (DBT), developed by Marsha Linehan, is a

comprehensive treatment program based on cognitive and behavioral principles
that are complemented by acceptance-based strategies derived primarily from
Zen Buddhism. Originally designed for individuals with borderline personality
disorder (BPD) and self-injurious behaviors, DBT has been shown to
significantly improve the outcome in this population. Because of its efficacy in
treating BPD, clinicians and researchers have become interested in the
application of DBT to other difficult, refractory, or chronic conditions,
including eating disorders. Moreover, comorbid BPD is common among eating
disorder patients, and self-injurious and suicidal behaviors also are common (see
Chapter 9). One recent report found that approximately 20% of women with
an eating disorder diagnosis endorsed an episode of self-harm in the previous
6 months, and one-third of these indicated self-injuring at least several times per
month (1).
However, it is important to note that our use of DBT in the treatment of
patients with eating disorders is not limited to individuals with comorbid BPD or
parasuicidal behavior. We have found DBT to be useful in the treatment of any
chronic eating disorder, including restricting anorexia nervosa, because of its
acceptance of and tolerance for refractory symptoms and matter-of-
fact emphasis on the need for change. DBT may have particular utility for
anorexia nervosa (2) because chronically restricting patients often find it
excruciatingly difficult to modify symptoms, and consequently avoid or resist
treatment. DBT incorporates specific techniques for working with patients who
are ambivalent about change, and specific strategies for maintaining commitment
to treatment. Similarly, chronic anorexia patients often have limited ability to
negotiate emotionally or interpersonally difficult situations. DBT emphasizes
the need for acquisition of life skills to promote effective functioning and
recovery. Finally, DBT provides an effective methodology for dealing with
therapists’ responses to patients, which is often helpful when treating numbers
476 MARCUS AND LEVINE
of chronically ill patients with anorexia nervosa. Accordingly, we have used

DBT as an outpatient treatment for eating disorder patients, irrespective of
eating disorder diagnosis, who have not responded to front-line treatments, and
have incorporated DBT principles in the treatment of patients in our inpatient
and partial hospitalization programs.
In this chapter, we first provide a brief overview of the philosophy and
assumptions of DBT and a description of the treatment with comments regarding
the applicability of DBT to the management of eating disorders. Next, we
present information on the efficacy of DBT in BPD and other disorders, along with
evidence bearing on its utility in managing disordered eating. Finally, we discuss
additional aspects of the implementation of DBT and its principles in our
program. Although we describe the basics of DBT, a detailed presentation of its
principles and procedures is beyond the scope of this chapter. These are
explicated in full in Dr. Linehan’s text, Cognitive-Behavioral Treatment of
Borderline Personality Disorder (3), and skills manual, Skills Training Manual for
Treating Borderline Personality Disorder (4). Information about DBT training and
resources also can be found on the web site of the Behavioral Technology
Transfer Group (http://www.behavioraltech.com).
PHILOSOPHY AND ASSUMPTIONS OF

DIALECTICAL BEHAVIOR THERAPY
A dialectical world view is at the core of DBT. In its broadest sense, dialecticism
refers to the philosophy that reality comprises opposing forces, “thesis” and
“antithesis,” and when these polarities are balanced or integrated, the resulting
“synthesis” immediately creates a new set of opposing forces. Thus, reality is by
its nature complex and dynamic with the important corollary that there is not
one correct point of view or understanding. The implications of a dialectical
world view are manifold and permeate DBT. They include a responsibility on
the part of the therapist to identify the “truth” in the patient’s behavior, validate
it, and accept responsibility to make use of persuasion to encourage the
development of synthesis and a new and more adaptive reality for patients. A
dialectical world view has important implications for patients as well. For
example, patients are helped to accept that truth is not an absolute and that
apparently opposing beliefs or emotions can exist simultaneously in the same
individual.
The adoption of a dialectical world view has profound advantages in the
management of refractory eating disorder behaviors. Symptoms of eating
disorders are perplexing, confounding, and often repugnant to patients’ friends,
family members, some health care professionals, and even the patients
themselves. Thus, the conscious and deliberate attempt on the part of clinicians
DIALECTIC BEHAVIOR THERAPY 477
who work with these individuals to understand the validity of eating disorder
symptoms as efforts to cope with aversive circumstances is a critical tool in the
establishment and maintenance of an effective working relationship. Similarly, a
dialectical world view recognizes and accepts the difficulty of change and the
ambivalence that eating disorder patients have about modifying or relinquishing
their symptoms. The deliberate balance between acceptance and pulling for
change also serves to reduce ambivalence about the therapist, who may be
perceived as intrusive or controlling.
ETIOLOGY OF BORDERLINE PERSONALITY

DISORDER
Linehan’s model of the etiology of BPD also is dialectical in nature. The
biosocial model (3) posits that the development of a borderline personality is
the result of a transaction between a biological vulnerability to emotional
dysfunction and an invalidating environment that creates and maintains
borderline behavior patterns over time. Emotional dysfunction is defined as
difficulty in modulating emotions and increased vulnerability to intense
emotional experiences, which can be summarized in three ways. First, an
individual may be highly sensitive to emotional experience, having a low
threshold for emotions and developing immediate, intense reactions. Second,
dysregulated emotional processing may mean that an individual has a heightened
reactivity, or a stronger than average response, to some emotional cues. Finally,
emotion dysregulation may take the form of a slow return to a baseline level of
emotion. This slow return can be problematic because new events are more
likely to rekindle an intense emotion during the period in which the individual is
regaining emotional equilibrium.
The second idea in the biosocial model is the invalidating environment, which
refers to any environment that continually communicates that an individual’s
reactions are faulty, exaggerated, or inappropriate. Examples of environmental
invalidation include failing to validate an individual’s private experience,
oversimplifying the ease of problem solving, or punishing an emotional display.
In the context of an invalidating environment, an individual with a vulnerable
emotional processing system fails to learn to tolerate or trust her private
emotional and cognitive experiences. Some environments are obviously
invalidating, e.g., an environment where physical or sexual abuse occurs. It is
important to note, however, that invalidation need not be intentional or
dramatic, but occurs when the environment consistently is unresponsive to the
needs of a particular vulnerable individual.
The ongoing interaction of vulnerability and invalidation results in emotion
dysregulation, which in turn leads to the emotional, cognitive, and behavioral
symptoms that serve a regulating function. Within this framework, chronic

suicidal urges and parasuicidal behaviors are understood as legitimate, if
ultimately dysfunctional, attempts to solve problems in an individual who
experiences intense suffering in a life that often seems unbearable. Thus,
although the overarching goal of DBT is to help the patient develop a life that is
worth living, a more proximal goal of treatment is to reduce maladaptive
problem solving and enhance adaptive functioning in response to emotional
dysregulation.
The biosocial model of the etiology of BPD overlaps neatly with the affect
regulation model of eating disorder symptoms. In the affect regulation model,
eating-disordered behaviors, such as binge eating, food or calorie restriction,
overexercising, and purging, are seen as efforts to modulate strong negative or
aversive affective experiences. Because DBT is specifically designed to address
the core problem of emotion regulation, it provides a clear framework for
addressing problem behaviors in eating disorders. Furthermore, a growing body
of research evidence has documented that eating disorders aggregate in families,
and there is increasing consensus that biological vulnerability is involved in the
pathogenesis of eating disorders. Thus, a biosocial framework of etiology
provides a meaningful heuristic to understand the complex cascade of factors
that lead to the expression of eating disorder syndromes.
DIALECTICAL BEHAVIOR THERAPY

DBT makes a series of assumptions about patients that are critical to the
successful implementation of treatment and that inform all intervention
strategies. Two DBT assumptions, that “Patients are doing the best that they
can” and that “Patients want to improve,” are the core of a successful therapeutic
stance. These seemingly simple assumptions are critically important in work
with chronic or refractory patients, where therapists may feel at times that
patients do not want to get better or are not trying hard enough. However,
assuming that a patient does not want to recover or is not trying is a prototype
of invalidation for the patient who is likely to feel misunderstood, which in turn
leads to frustration, anger, or shame. Importantly, the adoption of these
assumptions by the therapists does not mean that patients consistently feel that
they want to recover or are always trying their best; rather, the therapist adopts
the assumptions until fact finding in session with the patient leads to a different
conclusion.
The next assumptions, which seem paradoxical, and thus embody a
dialectical world view, are that “Patients need to do better, try harder, and be
more motivated” and that “Patients may not have caused all of their own
problems, but must solve them anyway.” That patients need to do better to have
a worthwhile life usually is self-evident, and they require substantial help to

reduce barriers to change, increase motivation, and enhance skills. Similarly,
although the patient requires help from the therapist (and usually from others in
her life), the business of change is hard work, and the patient, by necessity,
carries the burden of changing. Finally, DBT makes the assumption that
“Patients cannot fail in therapy.” The treatment may fail, but not the patient. This
assumption reflects an imperfect knowledge regarding how to help and a
profound awareness of the difficulty of change.
The dialectic between the assumptions that patients are doing the best that
they can, and that they need to try harder and carry the responsibility for change
has clear applicability to the treatment of eating disorders, where there often is
the unfortunate assumption that patients do not want to relinquish their
symptoms. In some instances, the fear of a life without the eating disorder is
overwhelming, but patients invariably express a longing for a life free of eating
disorder-related debility.
In summary, the philosophy and assumptions of DBT reflect a basic
understanding of the nature of reality, the etiology of psychiatric disorder, and
the role and meaning of psychiatric symptoms. Like other cognitivebehavioral
treatments, DBT systematically utilizes the complete arsenal of change
strategies but balances the focus on change with respect for the truth in patient
symptom behaviors and acceptance of the difficulty of change. We have found
this framework to be eminently suitable for work with eating disorder patients
who have not responded to initial treatment or who have had multiple
recurrences of disorder.
MODES OF TREATMENT IN DIALECTICAL

BEHAVIOR THERAPY
Standard outpatient DBT includes four treatment modes, all of which are
important to the progress of therapy: (a) weekly individual therapy, (b) weekly
group-based skills sessions, (c) telephone consultations, and (d) regular
consultation meetings for the therapists. Individual treatment focuses on three
primary targets: life-threatening behaviors, therapy-interfering behaviors, and
quality-of-life-interfering behaviors. A fourth target area, increasing behavioral
skills, is addressed primarily in the context of the weekly skills group. The
sanctioned use of the telephone is designed to provide pa tients with the
opportunity to obtain coaching in real-life situations to promote the
generalization of skills outside of the therapy setting. Finally, DBT involves a
regular consultation with a team that provides support for therapists working
with difficult patients.
DBT progresses in stages. In the pretreatment phase patients are oriented to

treatment, and the patient and therapist agree on the overall goals of the work.
One cannot overemphasize that nothing in DBT is done without directly
involving, educating, and obtaining agreement from patients before the
initiation of a treatment strategy or treatment plan. The first stage of treatment,
which generally lasts about one year, focuses on reducing acute symptoms and
increasing behavioral skills. DBT research has focused predominantly on the
outcome of the first stage of treatment, and the content of the current chapter
also focuses on the initial year of DBT.
The target of the second stage of DBT is decreasing posttraumatic stress if
present. The reasoning behind deferring a focus on these symptoms is based on
the assumption that patients have to acquire the basic abilities and external
supports to tolerate the painful work involved in dealing with trauma. The final
stage of DBT focuses on increasing self-respect and achieving individual goals,
which in the DBT framework are possible only after the patient has developed
some ability to trust herself and the therapist.
Commitment Strategies in Dialectical Behavior

Therapy
Because the treatment process is acknowledged to be difficult, the notion of
commitment is critical in DBT. Commitment to participating in DBT for a
specified period of time and working on the overall goal of building a life worth
living and reducing life-threatening behaviors are regarded as necessary to
initiate effective treatment. The patient also is asked to commit to collaborating
in the treatment procedures within a given session. In addition to these general
commitments, commitment occurs at a more specific level, as commitment is
sought to carry out the behavior changes agreed on by the patient and therapist.
The therapist directly elicits a patient’s agreement to try new behaviors or work
on specific problem areas. Conversely, the therapist also highlights the patient’s
freedom to make her own choices while clearly presenting the realistic
consequences of those choices.
There are specific DBT strategies for eliciting and maintaining commitment
to treatment. These strategies include highlighting and discussing the pros and
cons of a commitment to change, playing the devil’s advocate (i.e., questioning
the wisdom of change) to strengthen commitment and heighten a sense of self-
control, or use of the “foot in the door, door in the face” technique (i.e., getting
a small commitment, then requesting a major change) to pull for patient
commitment to specific goals and procedures. Throughout treatment the
therapist highlights previous commitments made by the patient in an effort to
clarify and strengthen current commitments. The DBT conceptualization of
commitment as a dynamic process that occurs over time is helpful in working

with chronic eating disorder patients, whose commitment to change is often
uncertain and wavering.
Individual Therapy in Dialectical Behavior

Therapy
Individual therapy occurs once a week, and the individual therapist assumes
primary responsibility for patient care. Individual sessions are active and
structured, with the therapist and patient agreeing to focus work on the
hierarchy of treatment targets in the service of the overall goal of building a
worthwhile life. This hierarchy serves to organize the work of the therapist and
patient both within and across sessions. As noted above, DBT specifies three
treatment areas that are targeted in the order of importance. The first goal of
treatment is to decrease suicidal and other life-threatening behaviors. This goal
addresses self-injurious behaviors as well as suicidal ideation, threats, and plans.
Because staying alive is the sine qua non for building a life worth living, the
occurrence of parasuicidal ideation or behavior takes precedence over other
agenda items each week.
Next on the treatment hierarchy are behaviors that interfere with the
therapeutic process. Examples of therapy-interfering behaviors by a patient
include noncompliance with agreements made with the therapist or failing to
attend sessions. The importance accorded to therapy-interfering behaviors is an
acknowledgment that the continuity of treatment must be preserved if
improvements are to be made, and reflects an ongoing emphasis on maintaining
the therapeutic relationship. The third target in the treatment hierarchy
includes behaviors related to the patient’s quality of life. This domain includes
most symptoms and also focuses on areas of general dysfunction, such as
financial difficulties, interpersonal problems, or problems related to school or
employment.
For the most part, eating disorder behaviors, such as calorie restriction and
binge-purge behaviors, are targeted as quality-of-life issues. One exception is
that significant calorie restriction in a low-weight patient is considered therapy
interfering for two reasons. First, starvation and weight loss frequently lead to
an inpatient hospitalization, which interferes with outpatient DBT. Second,
patients who are actively anorexic are often cognitively unable to benefit from
psychotherapy. Similarly, there are times when symptoms can be considered
therapy interfering and thus assume a higher priority in treatment. For
example, if a patient is restricting liquids to the extent that she requires
rehydration and misses treatment, then this behavior can be construed as
therapy interfering. Specific eating disorder behavior targets are summarized in

Table 1.
The focus of any individual therapy session is determined by a review of the
patient’s diary card. The diary card is a weekly record of the patient’s daily
actions, including drug and alcohol use, parasuicidal acts, feelings of misery,
suicidal ideation, and urges to self-harm. We have modified the standard DBT
diary card to include eating disorder behaviors as shown in Figure 1. Therapy
begins each week with a review of this card, which is then used to set the session
agenda. Failure to complete the card is considered therapy-interfering behavior,
and the first order of business is to complete the card or recreate it.
Next, a specific problem behavior gleaned from a review of the diary card is
selected according to the hierarchy of treatment targets, and a detailed
functional analysis of the problem behavior is conducted. This analysis, called a
behavior chain analysis, begins with a clear definition of the problem behavior,
and a review of general and situation-specific precipitants of the behavior.
Together, the therapist and client determine and list the thoughts, feelings, and
circumstances that link a precipitating event to the problem behavior. Equal
emphasis is placed on identifying points in the behavior chain at which an
alternative, adaptive response could have been initiated. Therapy then proceeds
with a solution analysis, which includes identification of the skills needed and
analysis of any factors that prevented the client from utilizing these skills in the
specific situation. The focus on specific problems and the use of cognitive and
behavioral strategies in DBT is similar to that in other evidenced-based
treatments.
However, in contrast to other cognitive-behavioral treatments, DBT requires
that the strategies that focus on change (i.e., problem solving) and those based
on acceptance (i.e., validation) are of equal importance, and the therapist
continually seeks to balance problem solving and validation in each session.
Validation strategies are those that clearly communicate to the patient that her
behavior can be understood and makes sense in the context of her experience.
fgdgdf
TABLE 1 DBT Treatment Hierarchy in Standard DBT for Eating Disorders
FIGURE 1 DBT diary card adapted for eating disorder behaviors.

Therapists are required to actively observe patients, effectively utilize reflection

skills, and work to identify the truth (even if it is only a small part of a largely
ineffective or dysfunctional reaction) in the patient’s responses to her
environment. There are numerous emotional, behavioral, and cognitive
validation strategies that are described briefly below.
Emotional validation strategies include providing opportunities for emotional
expression, helping the patient to observe and label her emotions, and
commimicating in a nonjudgmental fashion that feelings are valid and
understandable. The DBT therapist also employs behavioral validation strategies
that involve identification of patient’s self-imposed behavioral demands and
working to modify negative judgments by helping patients to accept that all
behavior is understandable given the circumstances. There is, though, an
important distinction between understanding and approving of a given behavior,
specifically, bad behavior is understandable but not desirable, and usually
ineffective. For example, self-injurious behavior is understandable as a strategy
to avoid overwhelmingly negative affect, but it has deleterious consequences for
the health of the individual and often has negative effects on an individual and
her social milieu as well. Finally, cognitive behavior strategies focus on helping
patients to identify dysfunctional thoughts and values, and to differentiate
between facts and an individual’s interpretation of the facts.
Although a discussion of the full range of the dozens of DBT strategies is not
feasible here, we have presented several important strategies to illustrate the
deliberate and active role of the therapist in pulling for change and validating the
wisdom in the patient’s choices, and the ways in which DBT strategies are
adapted for use with eating disorder patients.
Dialectical Behavior Therapy Skills

DBT requires that patients participate in weekly skills training sessions as well as
individual psychotherapy based on the assumption that patients with chronic or
refractory difficulties lack the behavioral skills to make and sustain life changes.
Specifically, patients often display what Linehan terms apparent competence. That
is, an individual may be accomplished and intelligent, yet lack basic skills to
regulate mood, tolerate distress, and assertively negotiate interpersonal
relationships. The discrepancy between apparent competence and lack of ability
to effectively and accurately communicate feelings is clearly evident in eating
disorder patients. An effective, competent demeanor often masks the reality of
inner conflict, turmoil, and insecurity.
Four specific skill areas related to the problems of dysregulated
emotions and ineffective problem solving are targeted (4). The four skill
modules are (a) mindfulness, (b) interpersonal effectiveness, (c) emotion
regulation, and (d) distress tolerance. The practice and application of the skills
is reinforced continually during individual therapy and are a specific focus of
telephone consultations with a patient. An overview of the content in each of
these skill areas follows.
Mindfulness practice is rooted in the Zen tradition of acceptance and
nonjudgmental attitude. The skills taught in this module involve learning to
observe, identify, and participate in a range of experiences with awareness. The
module begins by describing the idea of “wise mind,” a term used to refer to the
synthesis, of “emotion mind,” which includes feelings, wishes, and impressions,
and “rational mind,” which consists of thoughts, logic, and facts. Patients
cultivate the ability to focus attention on one thing at a time, and the concept of
focusing on what is effective or doing what works in a given situation is
reinforced.
Distress tolerance skills focus on coping with unpleasant and painful
emotions and situations. DBT provides concrete crisis survival strategies to
increase the ability to tolerate distress and accept life as it is in the moment.
There are several classes of distress tolerance strategies, including distraction
strategies, self-soothing strategies (we utilize all of the self-soothing strategies
with eating disorder patients except for self-soothing with food), strategies to
improve the moment, and thinking of pros and cons. Acceptance skills include
cultivation of radical acceptance and willingness. Radical acceptance involves
acknowledging and accepting current emotions, thoughts, and environmental
situations and developing a capacity to accept painful emotions as a part of life.
Willingness describes the development of a capacity to be fully alive at every
moment.
Emotion regulation skills involve identifying and labeling emotions, and
effectively managing extreme emotional states. The module includes lessons on
the function of emotions, as well as skills to decrease emotional vulnerability
and increase positive emotional experiences. With eating disorder patients, the
emphasis on identifying and labeling emotions is particularly important because
severe calorie restriction and other eating disorders symptoms may serve to
numb feelings, with the result that patients have a significantly narrowed
affective range. Thus, it is often necessary to help the eating disorder patient to
recognize feelings so as to learn how to regulate them.
Interpersonal effectiveness skills include assertiveness and interpersonal
problem solving. Effective strategies for identifying a goal and developing a plan
to obtain the changes an individual wants without sabotaging important
relationships or losing self-respect in an interpersonal situation are taught.
Within the group setting, ways of asking for what one needs, saying no, and coping
with interpersonal conflict also are modeled and practiced.
In summary, the incorporation of skills training in DBT reflects the

assumption that patients who must change to have a life worth living need the
tools to achieve that goal. Some patients may lack even the most basic
emotional, cognitive, or behavioral skills, whereas others have the needed skills
but lack the capacity to implement them in real-life situations. Overall, the aim
in DBT is to eliminate dysfunctional or ineffective behaviors and replace them
with skillful ones.
Use of the Telephone in Dialectical Behavior

Therapy
The purposes of telephone contact in DBT are to encourage the generalization of
DBT skill use in daily life and to provide patients with a between-session
opportunity to clarify or repair relationship problems with the therapist.
Telephone contacts between a therapist and client are encouraged, but DBT
specifies that phone conversations be brief and focused. Problem resolution is
accomplished during the individual therapy session. As always is the case in
DBT, rules on the use of the telephone are discussed and agreed on prior to
their initiation.
An important aspect of telephone use in DBT is the “24-hour rule.” Based on
the idea that telephone contact with a primary therapist may be reinforcing and
that the first treatment target is the elimination of self-harm, patients commit to
the idea that they may not speak to their therapist on the telephone for a full day
following an episode of self-injury. The 24-hour rule is discussed in detail prior
to the onset of the treatment, and specific plans are made for how and where
the patient will receive medical care after self-injury should it occur. In our
experience, patients readily understand the rationale for 24-hour rule.
Specifically, since the patient and therapist have agreed to work on eliminating
this problem behavior, the time in which the therapist can be helpful is before
self-injury occurs.
The Dialectical Behavior Therapy Consultation

Team
The final DBT treatment mode is the consultation or supervision team. Difficult
patients frequently evoke strong feelings from therapists (and others who care
about them). Competent, successful treatment requires that therapists maintain
an awareness of responses that may interfere with treatment and develop
effective tools to manage them. Thus, therapists participate in a consultation
team (composed of at least one other DBT therapist) that provides ongoing
support and coaching to therapists working with difficult patients. The
consultation team meets weekly to discuss cases with the goal of encouraging
therapists’ motivation, enhancing their skills, and promoting a dialectical view of
the patient and her problems.
In summary, DBT requires a considerable commitment of time and energy
on the part of the patient and the treatment team, as individual therapy, skills
training, telephone use, and a therapist consultation team are standard. However,
in the context of the considerable morbidity of BPD and other difficult-to-treat
psychiatric disorders, such as chronic eating disorders, the investment required
for DBT is regarded as cost effective.
DIALECTICAL BEHAVIOR THERAPY EFFICACY

Because it was developed to manage self-injurious behaviors in individuals with
BPD, the initial DBT treatment outcome research was conducted with this
population and several studies have documented the efficacy of DBT.
Specifically, compared to women who received treatment as usual in the
community, those treated with DBT reported reductions in both the number
and severity of parasuidal acts (5). DBT also resulted in reductions in the
number of hospitalizations, lower rates of treatment attrition, and improved
global functioning compared to treatment as usual (5,6). Moreover, one year
posttreatment, psychological adjustment was improved and distress lower in
those who had received DBT compared to usual treatment (6).
DBT also has been adapted and tested in the management of other chronic
behavior problems, particularly substance abuse and eating disorders. In DBT for
substance abuse, the primary behavioral targets are expanded to include the
cessation of drug use and the continuity of therapy. Additional treatment
strategies designed to enhance the attachment of the patient to the therapist also
are incorporated (7,8). To date, all of the research on the use of DBT for
substance abuse has been conducted with women who have been diagnosed with
both substance dependence and BPD. In general, these studies have
demonstrated the utility of DBT. Compared to women who received treatment
as usual in the community, women who received DBT engaged in less drug use
and reported better global and social adjustment and less anger at the end of
1 year of treatment and at 4-month follow-up (8).
Recently, Linehan and colleagues (9) reported results of an investigation that
compared standard DBT to an intensive, standardized treatment program,
termed comprehensive validation therapy, which utilized only a subset of DBT
strategies, i.e., acceptance and validation. Results of this trial indicated that in
both treatment programs, women decreased drug use during treatment and
over a 4-month follow-up period, although the women who received standard
DBT maintained treatment gains better than those who received comprehensive
validation therapy. Furthermore, although both treatments were effective in

reducing levels of psychopathology, women were significantly less likely to drop
out of the comprehensive validation treatent than DBT. These data suggest that
the acceptance strategies of DBT, even in the absence of the behaviorally
focused change strategies, may be effective, as both DBT and comprehensive
validation therapy, a treatment involving acceptance-based DBT strategies,
appear to be effective in decreasing substance use in women with BPD.
Telch and colleagues have adapted and tested a group-based version of DBT
designed for individuals with binge eating disorder (BED) (10–12). This
program has been effective in decreasing binge eating behavior and maladaptive
attitudes about eating, shape, and weight. In one study (11), 89% of the women
with BED who completed the 20-week DBT group treatment were abstinent
from binge eating compared to only 13% of those in the waitlist group, and
more than half of the women who received DBT maintained their binge eating
abstinence in the 6 months following the end of treatment. Similarly, group-
based DBT was related to significant decreases in binge eating among women
who met modified criteria for bulimia nervosa (criteria were modified to
decrease the required binge/purge frequency to once per week) (13). Although
women who received DBT did not report improvements in mood or depressive
symptoms compared to those on a waiting list, none of the patients dropped out
of the DBT treatment. In summary, initial evidence has suggested that a group-
based version of DBT is useful in the management of BED and bulimia nervosa.
In summary, there is evidence that DBT is useful in the treatment of
individuals with BPD, substance abuse, and eating disorders. Longer term
randomized controlled trials are needed to document its efficacy, but initial
studies have provided encouraging results.
DIALECTICAL BEHAVIOR THERAPY ACROSS THE

EATING DISORDERS CONTINUUM OF CARE
Our Eating Disorders Program at Western Psychiatric Institute and Clinic offers
a full continuum of services, and patients move across levels of care as indicated
by clinical status. We have a dedicated 11-bed inpatient unit that provides care
primarily to low-weight patients who require medical monitoring and
refeeding. Occasionally, normal-weight patients with intractable vomiting or
other purge behaviors that have led to medical sequelae are more briefly
hospitalized, as are other eating disorder patients with psychiatric exacerbations
(e.g., suicidal patients). The average length of stay on the inpatient unit is
approximately 4 weeks. The Partial Hospital Program is provides a 5-day-per-
week program for patients who require close monitoring and meal supervision.
The schedule comprises three longer days (noon to 8 p.m.) with lunch and
dinner, and two shorter days (9 a.m. to 1 p.m.) with breakfast and lunch. The
average length of stay in Partial Hospital is 3 weeks. The Intensive Outpatient
Program provides 9 hours of program over three evenings from 5 p.m. to 8 p.m.,
and includes dinner. This provides a step-down level of care for patients who
have been in Partial Hospital, or a step up for patients who are not benefiting
from regular outpatient treatment. Standard DBT, as described above, is offered
to appropriate patients as part of our outpatient clinic that provides
psychotherapy, family therapy, nutrition counseling, and medical monitoring
for patients with the full spectrum of eating disorders.
We have incorporated DBT principles in each level of care in our program.
Some of the strengths of DBT are that it provides an overall philosophy for
understanding refractory symptoms, encourages a compassionate and
nonjudgmental stance for all treaters, and offers nonpejorative, easily
understood language to describe patient behaviors and staff responses to
patients. Although it is not feasible or appropriate to provide formal DBT
training to all of the professional and paraprofessional staff that work with
patients, we have disseminated the assumptions of DBT and provided training
on the use of consultation to help staffdeal with feelings they have about
working with patients. We also have provided training to ensure that staff
members temper cognitive-behavioral strategies for change with balancing doses
of validation and acceptance. DBT language is used across the program, and
patients as well as staff members understand the meaning of terms used in DBT
skills training. The use of common terminology promotes the development of
cohension and sense of purpose in the therapeutic milieu.
In the inpatient unit, patients are exposed to the DBT skills modules,
particularly mindfulness and distress tolerance as well as to the use of behavior
chain analyses. Exposure to DBT skills is intensified in the Partial Hospital
Program, and patients are encouraged to make use of DBT skills throughout acute
treatment. Patients in the partial program also must commit to working on
recovery and not to engage in symptom behavior on the premises. Thus, eating-
disordered behaviors that occur on the premises, such as food restriction or
purging, are considered therapy interfering regardless of the patient’s degree of
medical risk. If patients engage in problem behaviors they are asked to do
behavior chain analyses, which are then reviewed with the therapist. The
therapist and patient then agree on a plan to repair any negative consequences of
the problem behavior.
SUMMARY
We have successfully utilized DBT in the management of chronic eating
disorders as it provides a treatment context in which change not only is possible
but also is explicitly nurtured and reinforced in the context of respect for and
acceptance of patients’ struggles and ambivalence. DBT principles can be
disseminated to ancillary staff who work with patients and to families to
promote positive and constructive means for understanding difficult
behavior. Although research is needed to determine if DBT confers benefits for
chronic eating disorder patients over and above care as usual, our clinical
experience suggests that DBT benefits patients and the staff who work with
them.
REFERENCES
2. McCabe EB, Marcus MD. Is dialectical behavior therapy useful in the management
of anorexia nervosa? Eat Disorders J Treat Prevent 2002; 10:335–337.
3. Linehan MM. Cognitive Behavioral Treatment of Borderline Personality Disorder.
New York: Guilford Press, 1993.
4. Linehan MM. Skills Training Manual for Treating Borderline Personality Disorder.
5. Linehan MM, Armstrong HE, Suarez A, Allmon D, Heard HL.
Cognitivebehavioral treatment of chronically parasuicidal borderline patients. Arch
Gen Psychiatry 1991; 48:1060–1064.
6. Linehan MM, Heard HL, Armstrong HE. Naturalistic follow-up of a behavioral
treatment for chronically parasuicidal borderline patients. Arch Gen Psychiatry
1993; 50:971–974.
7. Dimeff L, Rizvi SL, Brown M, Linehan MM. Dialectical behavior therapy for
substance abuse: a pilot application to methamphetamine-dependent women with
borderline personality disorder. Cogn Behav Prac 2000; 7:457–468.
8. Linehan MM, Schmidt H, Dimeff LA, Craft JC, Kanter J, Comtois KA. Dialectical
behavior therapy for patients with borderline personality disorder and drug-
dependence. Am J Addict 1999; 8:279–292.
9. Linehan MM, Dimeff LA, Reynolds SK, Comtois KA, Welch SS, Heagerty P,
Kivlahan DR. Dialectical behavior therapy versus comprehensive validation therapy
plus 12-step for the treatment of opioid dependent women meeting criteria for
borderline personality disorder. Drug Alcohol Depend 2002; 67:13–26.
10. Telch CF, Agras WS, Linehan MM. Group dialectical behavior therapy for binge-
eating disorder: a preliminary, uncontrolled trial. Behav Ther 2000; 31: 569–582.
11. Telch CF, Agras WS, Linehan MM. Dialectical behavior therapy for binge eating
disorder. J Consult Clin Psychol 2001; 69:1061–1065.
12. Wiser S, Telch CF. Dialectical behavior therapy for binge-eating disorder. J Clin
Psychol 1999; 55:755–768.
13. Safer DL, Telch CF, Agras WS. Dialectical behavior therapy for bulimia nervosa.
Am J Psychiatry 2001; 158:632–634.
21
Psychopharmacology of Anorexia Nervosa,
Joanna E.Steinglass and B.Timothy Walsh
New York State Psychiatric Institute, Columbia University
As the fundamental causes of eating disorders remain unknown, it is no surprise

that development of successful treatments has not come easily. Nonetheless,
much progress has been made. The current mainstays of treatment of eating
disorders are psychological interventions, including cognitive therapy,
behavioral therapy, family therapy, and nutritional counseling (1). Clinicians
have generally looked to medication to augment the effects of psychological
intervention, or as a primary intervention when such treatment is unavailable or
ineffective.
The role of psychopharmacology in eating disorders has been greatly clarified
in the last decade. But, as in the case of other psychiatric disorders, the limited
understanding of the basic pathophysiology handicaps the ability to design
psychopharmacological treatments. In the absence of a specific biological
model, a range of differing perspectives has prompted attempts to identify
medication treatments for anorexia nervosa (AN), bulimia nervosa (BN), and
binge eating disorder (BED). In general, studies of medications in AN have been
disappointing. In contrast, in BN, antidepressants are clearly effective at
reducing binge eating and purging behaviors. Studies of BED are at an earlier
stage but have already yielded promising findings. A consistent observation
across the eating disorders is the need for double-blind, placebocontrolled
studies to assess efficacy, as a medication often looks promising in case studies
or open trial but fails to show superiority to placebo in a rigorously controlled
trial.
This chapter will review the current data on the use of medications in the
management of AN, BN, and BED with the intention of providing the clinician
with information needed to make decisions about pharmacological treatment.
The data will be reviewed according to what has been shown to be useful on the
basis of controlled trials. Each illness—and stage of illness in the case of AN—will
be considered separately and different outcome measures will be discussed
when available.
492 STEINGLASS AND WALSH
ANOREXIA NERVOSA
Anorexia nervosa is characterized by a relentless pursuit of thinness and fear of
becoming fat: patients starve themselves to extremes of low weight, resulting in
amenorrhea and risk of death. Treatment must target multiple aspects of the
disorder as patients need to gain weight, extinguish eating-disordered
behaviors, and alter cognitions that foster these behaviors. Current
recommendations focus on a multidisciplinary approach to treatment, including
psychotherapies with cognitive-behavioral components. Inpatient treatment for
patients at very low weight focuses on behavioral interventions and nutritional
counseling to encourage eating and weight gain in conjunction with beginning to
challenge cognitive disturbances. Nonetheless, AN has been difficult to treat and
has a high relapse rate. Thus medications are under investigation both to
facilitate initial treatment and to prevent relapse.
Study of the management of AN lends itself to multiple possible outcome
measures. The major initial concern is weight restoration, which can be readily
assessed by the amount and the rate of weight gain. In the long term, rate of
relapse is an important outcome, defined as significant weight loss or
reemergence of restrictive or binge-purge behaviors. Interwoven through both
phases of treatment (weight gain and relapse prevention) is the complex
problem of body image dissatisfaction.
Many psychopharmacological interventions have been tried, beginning with
the work of Dally and Sargant on antipsychotics in the 1960s. Due to the limited
understanding of the biological basis of AN, medication trials have been driven
by unproven theoretical models and/or by an interest in taking advantage of
medication side effects. While anecdotal reports of successful treatments have
been published, only a small number of randomized controlled trials have been
conducted, and definitive psychopharmacological treatment has not been
identified.
Antidepressants
Antidepressant treatment for AN is a reasonable notion given the common
concomitant symptoms of anxiety and depression. Many patients with AN
describe low mood, low energy, poor concentration, loss of interest, and social
isolation. The ritualized behaviors around eating and the obsessive
preoccupation with shape and weight can be conceptualized as on the spectrum
of obsessive-compulsive disorder (OCD), a syndrome that is also responsive to
antidepressant medication.
Controlled trials of several different medications to promote weight gain
have been generally discouraging. Initial studies involved tricyclic
PSYCHOPHARMACOLOGY 493
antidepressants (TCAs), with the hope that the side effect of weight gain would
add to the benefits of treating mood disturbance. There have been three
randomized controlled trials of TCAs. In one, clomipramine was associated
with a slower rate of weight gain than placebo despite increased appetite, and
there were no long-term effects at 1 and 4 years (2). In a study of amitriptyline
versus placebo, there was no significant difference in weight gain (3). A second
study of amitriptyline, which also had a third arm in which subjects received the
serotonin (5-HT) antagonist cyproheptadine, showed no major benefit of
amitriptyline (4). TCAs are known to prolong the QTc interval, which is also
affected by AN. These observations, coupled with concerns that TCAs in
children and adolescents may be linked to sudden death (5), suggest that TCAs
should be rarely used for patients with AN at low weight.
In light of their benign side effect profile and efficacy in many other disorders,
selective serotonin reuptake inhibitors (SSRIs) would appear promising in the
management of in AN. Initial anecdotal evidence that fluoxetine might be
beneficial for weight gain and mood symptoms (6) was supported in an open
trial (7). However, the single randomized placebocontrolled trial of fluoxetine
did not support these results. Attia et al. (8) conducted this trial of 33 patients
with AN at low weight. All patients received inpatient care in addition to either
fluoxetime (60 mg/day) or placebo for 7 weeks or until they reached 90% of
ideal body weight and maintained it for a week. Fluoxetine conferred no benefit
on weight gain, irrespective of subtype (restricting versus binge-purge). This
finding is consistent with an open trial of Strober et al. (9) who administered
fluoxetine to 33 inpatients.
While the studies of medication treatment in the acute phase focused on
weight gain, some studies also included measures of other dimensions of AN.
Mood symptoms have been found to improve with weight gain, with no
added benefit from medication (8,34). The open trial of Strober et al. described
above (9) examined severity of weight phobia and abnormal eating behaviors
and found no evidence that fluoxetine treatment was of benefit. Attia and
colleagues (8) assessed the effect of medication on body image dissatisfaction, a
core component of AN, and noted significant improvement with weight gain in
scores on Body Satisfaction Questionnaire (BSQ) in both placebo and fluoxetine
groups, although not to within the normal range in either group.
Somewhat more promising results have been found in the relapse prevention
phase, but there is just one randomized controlled trial. Thirtyfive women with
AN, restricting subtype, entered a double-blind, randomized, controlled trial
after inpatient weight restoration and received either fluoxetine (10–60 mg/
day) or placebo for 11 months (10). Subjects receiving medication were
significantly more likely than those who received placebo to maintain near-
normal weight for one year. Interpretation of data from the randomized trial is
limited in that dosage of fluoxetine was not controlled for, nor was
additional treatment (i.e., psychotherapy) restricted. Subjects were limited to
those with restricting subtype, and there is no information on the effect of
medication on parameters other than weight. In addition, a naturalistic study of
Strober et al. (11) comparing relapse among patients receiving open fluoxetine
treatment to relapse among a group of matched, historical controls failed to
detect evidence of a benefit from fluoxetine. Nonetheless, the study of
Kaye et al. (10) is virtually the only placebo-controlled examination of
medication in AN that found a statistically and clinically significant impact
of medication compared to placebo. Further study is needed to determine
replicate and extend this finding.
In summary, there is little reason to think that antidepressants add
substantially to the standard inpatient management of AN. Given the
widespread benefits of antidepressant medication in other, seemingly related
psychiatric disorders, the lack of impact of antidepressants is surprising.
Kaye et al. (12) have shown that patients at low weight have low levels of
5-hydroxyindoleacetic acid (5-HIAA), the major metabolite of serotonin, which
improve with weight gain. Low levels of 5-HIAA suggest that patients have low
levels of brain serotonin, which is consistent with the finding that dieting in non-
eatingdisordered women reduces tryptophan levels (the amino acid that is the
substrate for serotonin) and reduces serotonin production (13). Thus, it may be
that antidepressants are ineffective at low weight because they have insufficient
substrate (14). This is supported by the finding that tryptophan depletion has
been shown to reverse the effects of SSRIs in depressed patients (13).
Notably, virtually all studies of patients at low weight have been conducted in
an inpatient setting, where nonpharmacological interventions are effective in
producing weight gain. At least theoretically, there is potential for benefit from
medication in an outpatient setting, where weight gain tends to be slower.
Antipsychotics
Pharmacological treatment in AN began with antipsychotics. The theoretical
rationale for the use of this class of medication derives from the neardelusional
quality of beliefs about shape and weight held by some patients with AN. Dally
and Sargant (15) studied chlorpromazine (1600 mg/day) and found that while
the rate of weight gain was enhanced compared to historical controls, there
were significant negative effects including seizures and the emergence of binge-
purge behavior. Furthermore, benefits were not sustained over long-term
follow-up. Pimozide was studied subsequently in a randomized, controlled trial
of hospitalized patients (16). The authors found a trend toward slightly higher
daily weights while on pimozide, but effects on psychological symptoms were
inconsistent. In a study of sulpiride among hospitalized patients receiving either

medication or placebo for 3 weeks, there was no significant effect of medication
on weight gain (17). Due to the known long-term side effects of the older
antipsychotics and the side effects noted in some of the early studies, these
medications are not generally recommended for management of AN.
With the advent of the new generation of antipsychotics, which have a lower
incidence of tardive dyskinisia, extrapyramidal symptoms, and decreased
likelihood of seizures, the possible usefulness of this class of medications has
again been raised. Olanzapine would appear particularly promising as the
prominent side effect of weight gain might be advantageous in AN. There
have been several case reports of the use of olanzapine (5–10 mg/day)
comprising about 10 hospitalized patients who had been refractory to other
treatments (18–20). The patients described in these reports held neardelusional
beliefs about their bodies, with no other psychotic symptoms. Mehler et al. (20)
reported that while there was no dramatic improvement in the rate of weight gain
after initiation of medication, there was a marked improvement in patient’s
cognitive style. One patient with a history of restricting developed binge-purge
behavior while taking olanzapine (19). Olanzapine is a potentially promising
intervention for AN, but its efficacy has not yet been established. In light of the
research on antidepressants, where promising case reports and open studies
were not born out in randomized controlled trials, further study of olanzapine
and other second-generation antipsychotics is needed before definitive
conclusions can be drawn. In addition, an important clinical consideration is
whether patients will agree to a medication so clearly associated with
weight gain.
Other Agents
A number of other medication classes have been tried, targeting primarily the
weight gain phase of AN. Cyproheptadine, an antihistaminic agent that acts
centrally to decrease serotonin activity, has been studied in several controlled
trials after it was noted to cause weight gain in other conditions. Results have
been mixed. In the first placebo-controlled trial (21), cyproheptadine did not
improve weight gain. A second study found that cyproheptadine was associated
with improved weight gain in a subgroup of severely ill patients (22). In a trial
comparing amitriptyline, cyproheptadine, and placebo (mentioned above) (4),
the authors noted no significant weight gain in the cyproheptadine group.
However, they did note a difference between subtypes such that individuals
with the restricting subtype showed an increased rate of weight gain with
cyproheptadine whereas individuals with the binge-purge subtype showed an
increased rate of weight gain with amitriptyline.
Consistent with patients’ complaints about feelings of fullness and early

satiety, patients with AN have been found to have slowed gastric emptying (23).
Open trials of motility agents have been conducted using metoclopramide,
bethanacol, cisapride, and domperidone (14). Few agents have been subjected
to randomized controlled trials. Metaclopramide was found to decrease gastric
emptying time (24), but a randomized controlled trial could not be completed
because of the emergence of depression likely related to the CNS effects of the
drug (25). Cisapride is a motility agent with mixed agonist/antagonist
properties. It is an antagonist at the serotonin 5-HT3 receptor and an agonist at
the 5-HT4 receptor. Cisapride was shown to improve gastric emptying time in a
small, randomized, placebo-controlled study (N = 12) (26), but improvements
in weight gain were not noted. In a larger study, Szmukler et al. (27) described
improvement in gastrointestinal symptoms but no difference between
medication and placebo groups with respect to gastric emptying or weight gain.
Thus, the clinical benefits of cisapride in AN are uncertain. Furthermore, it was
recently withdrawn from the market in the United States due to cardiac
conduction effects, including prolonged QT interval and reports of sudden
death.
Patients often describe their eating disorder symptoms as overwhelmingly
strong urges to eat or to diet in a manner that bears some similarity to
descriptions of drug cravings. Several studies have been conducted to examine
the potential utility of opiate antagonists. Open trials of intravenous naloxone
and oral naltrexone in underweight patients suggested improved weight
gain (28). One placebo-controlled trial of naltrexone (200 mg daily) using a
crossover design was conducted in patients with AN and BN (29). While the
authors did not report results on weight gain in AN, they found that binge
eating and purging rates diminished.
Zinc deficiency has notable similarities to AN. It is associated with weight
loss, dysphoria, appetite and taste changes, and amenorrhea. Zinc deficiencies
have been noted in low-weight AN populations (30), an unsurprising finding
given the level of overall malnutrition. In contrast to most studies of AN, a number
of the studies of zinc supplementation were conducted in children and
adolescents. Three randomized controlled trials (30–32) and one open trial (33)
of zinc supplementation (50–100 mg elemental zinc/day) have been reported.
In a randomized controlled trial in children (31), there was no significant
weight effect in the zinc-treated group. In adolescents, Katz et al. (30) found
improvement in depression and anxiety in the adolescents who received zinc,
but no effect on weight gain. In contrast, Birmingham et al. (32) found that zinc
supplementation was associated with an increased rate of weight gain, even
without evidence of zinc deficiency. In light of these mixed results, the utility of
zinc supplementation is uncertain.
Other novel approaches to improving weight gain have included lithium, for
its weight gain and mood stabilizing properties, and tetrahydrocannabinol (THC)
for its appetite-enhancing effect. Lithium was associated with a small weight
increase in one small, short-duration placebo-controlled trial (34). THC was
compared to diazepam in a small randomized, doubleblind trial using a
crossover design (35). There was no benefit from THC with respect to food
intake or weight gain, and THC was associated with significant side effects,
including paranoia, sleep disturbance, and interpersonal sensitivity.
A major medical complication of AN is osteoporosis/osteopenia. Estrogen
replacement therapy has been used to treat osteoporosis in postmenopausal
women and therefore has been explored as an adjunctive treatment in AN.
However, a randomized controlled trial assessing the bone densities of subjects
receiving estrogen and progestin versus no medication found no significant
changes in the hormone-treated group (36). Those patients who resumed
menses showed improvement in bone density. These data suggest that, at
present, the best documented intervention to arrest bone loss in AN is weight
gain sufficient to restore regular menstruation.
BULIMIA NERVOSA
Bulimia nervosa is characterized by recurrent binge eating followed by
inappropriate compensatory behaviors, such as vomiting. Because in DSMIV,
AN has diagnostic precedence over BN, patients with AN who meet criteria for
BN are considered as having the binge-purge subtype of AN. Thus, most
patients with BN are of normal weight. Like patients with AN, those with BN
have a disturbance of body image and unduly value their shape and weight when
evaluating their self-esteem. Bulimia nervosa is more common than AN, with a
prevalence of 1–5% in adolescent and young adult women (37). BN tends to be
managed in the outpatient setting, making clinical trials less complicated and
costly than with AN and, presumably for these reasons, more numerous. In
addition, studies of medications in the management of BN have yielded more
promising results, most notably with antidepressants.
Antidepressants
The study of antidepressant medications resulted from the observation that
patients with BN, like those with AN, often describe depressive symptoms.
Over the past 20 years, many antidepressants have been found to be more
effective than placebo in reducing binge-purge episodes in normal-weight
women with BN (38,39). While TCAs (40–45), monoamine oxidase
inhibitors (46–48) and SSRIs (49–51) have all been shown to be effective, there
are no direct comparisons to suggest superiority of one drug over another.

SSRIs have come into favor due to their overall acceptable side-effect profile.
Open trials of sertraline (52) and fluvoxamine (53,54) have reported good
results, but only fluoxetine, at a dose of 60 mg daily, has been shown to be
effective in randomized, controlled trials (49,50). Antidepressants consistently
decrease eating-disordered behaviors and improve mood in patients with BN,
regardless of the presence of major depressive episode (49). In addition, two
randomized controlled trials have suggested efficacy of SSRIs in prevention of
relapse (55,56).
While studies of antidepressants have generally been favorable, the study of
bupropion must be mentioned for its significant side effects. In this trial of
55 women with BN (57), bupropion (up to 450 mg/day) was effective at
reducing binge-purge behavior. The study was terminated prematurely,
however, because four women experienced grand mal seizures. Because of this
association, bupropion is specifically not recommended in management of BN,
and the package insert indicates that bupropion is contraindicated in the
treatment of patients with a current or past diagnosis of BN or AN.
The clinical significance of the difference between antidepressant medication
and placebo is complicated by the broad range of effect between studies. The
improvement in binge frequency reported in controlled trials ranges from 31%
to 91% decrease (58,59). Remission rate (cessation of binge-purge behavior)
was often not reported and when reported, ranged from 4% to 34% (58). The
improvement in BN with antidepressant medication is clear, but the low
remission rate suggests that there are limitations to this treatment.
Medication trials have generally focused on binge-purge behaviors as the
primary outcome, but a few studies have also assessed body image
dissatisfaction. Most use the BSQ or the Eating Disorders Inventory Body
Dissatisfaction Scale (EDI-BD), which are self-report measures that address how
patients feel about body parts or their whole body. While these measures do
not address all of the dimensions of body dissatisfaction, they may serve as a
crude measure of this important variable. Interestingly, some, (45,49) but not
all, (60–62) studies suggest that it is possible to see change in body
dissatisfaction with medication treatment alone. Thus, the impact of
antidepressants on body dissatisfaction remains unclear.
Overall, the use of antidepressant medications in the management of BN is
well supported but there remain some gaps in the knowledge base. The current
data are mostly derived from short-term studies that range from 6 to 8 weeks
duration. Most trials have been conducted with normal weight women who use
self-induced vomiting to purge. Thus, it is unknown if these results can be
generalized to apply to other patients, such as men, adolescents, and those who
compensate for binge eating through other behaviors, e.g., excessive exercise.
Areas for further study include the optimal duration of treatment and the long-
term efficacy of antidepressants.
Anticonvulsants
An early clinical model conceptualized BN as a seizure disorder, with
bingepurge episodes thought to represent paroxysmal events. Small trials with
the anticonvulsants phenytoin (63) and carbamazepine (64) did not suggest a
robust response to medication. A recent case report (65) of a woman with
epilepsy and BN who was treated with topiramate and showed improvement in
binge-purge behaviors and in attitude about shape and weight raised the
possibility that topiramate may have benefit in the management of BN. Results
from a randomized, double-blind, placebocontrolled trial of topiramate
(25–400 mg/day) have been presented, showing reduction in binge and purge
duration and frequency (66). Although preliminary, these results are
encouraging.
Other Agents
Some agents have been studied based on biological models, as opposed to
clinical models, of BN. These models have focused mainly on the potential role
of serotonin, which has been shown to impact various aspects of feeding. As
increased serotonergic function tends to decrease food intake, it was
hypothesized that medications that increase serotonin would decrease binge
eating behavior (67). L-Tryptophan, the amino acid precursor of serotonin, was
examined in a randomized, placebo-controlled trial (N = 13), but no drug-
placebo difference was detected (68). Fenfluramine, a serotonergic agent that
both blocks reuptake and increases release, was studied with mixed results. In a
randomized, placebo-controlled trial using a crossover design, fenfluramine was
shown to decrease binge-purge frequency (69). However, in two subsequent
placebo-controlled trials, the medication showed no benefit (70,71).
Fenfluramine was withdrawn from the market in 1997 due to an association
with cardiac valve abnormalities.
Another model for treatment of BN focuses on the feeding behaviors of
patients, specifically their difficulty in identifying satiety. This model postulates
that binge eating and purging might lead to desensitization of the vagal nerve
afferents, which have a key role in signaling satiety. Subjects with BN were
found to have an increased somatosensory pain threshold, which may indirectly
reflect altered vagal nerve activity (72). Based on these observations, Faris and
colleagues (73) conducted a 4-week randomized, placebocontrolled trial in BN
and found that ondansetron (24 mg/day), a medication that blocks 5-HT3
receptors involved in visceral stimulation of the vagal nerve, was associated with
a significant decrease in binge-purge behaviors.
One clinical model of BN focused on the similarities between bingepurge
behaviors and addictive behaviors, drawing on the evidence that endogenous
opiates may be involved in appetite changes. In a small, open trial of
naltrexone, an opiate antagonist, 7 of 10 patients with BN improved with
complete or partial remission (74). One randomized, controlled trial including
AN and BN patients found a decrease in binge-purge behaviors with naltrexone
(100–200 mg/day) (29), whereas a randomized, controlled trial using a lower
dose (50 mg/day) showed no benefit (75).
Lithium has also been studied. Hsu et al. (60) conducted a randomized,
placebo-controlled, 8-week trial of lithium (mean lithium level = 0.62) in
patients with BN. Both placebo and medication groups improved, and there was
no significant difference between the groups.
Combination Treatment
While the above data clearly support the efficacy of antidepressants in BN,
controlled trials have also shown the efficacy of psychotherapy alone (see
Chapters 17,19, and 20). Seven randomized controlled trials have, in different
ways, compared treatments in an attempt to assess the benefits of psychotherapy
versus pharmacotherapy versus a combination of the two. Overall, the studies
suggest that cognitive-behavioral therapy (CBT) alone is probably more
effective in reducing binge eating and purging behaviors, but that the addition of
medication provides some additional benefit.
The first study, conducted by Mitchell and colleagues (62), randomized
subjects to one of four treatment arms for 10 weeks: imipramine alone, placebo
alone, intensive group therapy alone, or imipramine with intensive group
therapy. The psychotherapy intervention provided was an unusually intensive
group therapy, which included joint meals with the therapist on five occasions
during the first week of treatment. Outcome measures assessed eating
behaviors, affective symptoms, and attitudinal measures. The reduction in
eating-disordered behaviors in response to intensive group therapy was
impressive, and no added benefit from imipramine could be detected. On the
other hand, combination treatment significantly improved affective symptoms
more than either treatment alone. In a follow-up study, Keel et al. (76) found
that 10 years posttreatment, all three active treatment groups showed
significant improvement in social functioning as compared with placebo. There
were no significant differences on measures of depression, body image, or
eating disorder behavior.
A study of the combination of individual CBT with desipramine reached

broadly similar conclusions. Agras and colleagues (77) randomly assigned patients
to one of five treatment arms: desipramine for 16 weeks, desipramine for
24 weeks, CBT only for 16 weeks, CBT and desipramine for 16 weeks, and
CBT for 16 weeks and desipramine for 24 weeks. Response to CBT was clearly
superior to that for desipramine alone. There were some indications that
patients receiving CBT combined with 24 weeks of desipramine had the best
outcome, but these results were not statistically robust.
The finding of Mitchell et al. (62) that medication may add to the benefits of
psychological treatment was also noted in a study by our own group (78).
Patients were randomly assigned to one of five treatment arms for 16 weeks:
medication alone, CBT and medication, CBT and placebo, supportive
psychotherapy (SPT) and medication, SPT and placebo. The design allowed for
a change in medication from the TCA desipramine to the SSRI fluoxetine under
double-blind conditions. Patients assigned to receive active medication were
given desipramine, but if response was not satisfactory or if significant side
effects developed, patients were switched to fluoxetine. It was clear that CBT was
more effective than SPT in reducing disturbed eating behaviors. In addition,
active medication augmented the improvement in both behavioral and
attitudinal measures associated with psychological treatment.
Goldbloom et al. (79) conducted a study of combination treatment with
three treatment arms: fluoxetine alone, CBT alone, and fluoxetine and CBT
combined. This study was limited by a significant dropout rate (43%), which
contributed to an inability to detect statistically significant differences between
treatments on most measures. There was a significant difference in subjective
reports of binge episodes, which were most improved with combination
treatment.
In another study of fluoxetine, medication or placebo was combined
with 8 weeks of nutritional counseling (80). The nutritional intervention
had components similar to CBT, but with a psychoeducational focus replacing
cognitive restructuring. While the authors report on this as an assessment of
combined treatment, the psychotherapy intervention is sufficiently different
from CBT that it is difficult to compare findings. Nonetheless, there was a rapid
and significant improvement in both groups in binge eating and purging. There
were few differences between fluoxetine and placebo groups, but some
indications that fluoxetine augmented improvement in some psychological
spheres, such as concerns with shape and weight. Bacaltchuk et al. (81)
performed a meta-analysis of the studies of psychotherapy plus pharmacotherapy
in BN. While their conclusions are subject to the usual limitations of meta-
analyses, in the short term remission was more likely with combination of
medication and psychotherapy than with either treatment alone.
Conclusions drawn from the above studies must be considered with caution.
One major problem is that the largest studies were conducted before the
widespread use of SSRIs, making it difficult to extrapolate from these data to
current clinical practice. The limited information available suggests that CBT is
likely to be more effective and more acceptable to patients than is a course of
antidepressant medication. However, the data are reasonably consistent in
indicating that the addition of an antidepressant to psychotherapy modestly
augments improvement in psychological symptoms and, perhaps, in disturbance
of eating behavior. The effectiveness of medication in reducing relapse is
uncertain.

The more recently recognized syndrome of binge eating disorder is
characterized by episodes of binge eating without compensatory behaviors.
Although not required for the diagnosis, BED is usually associated with obesity.
As is the case with AN and BN, several clinical features of BED can be
appropriately viewed as outcome measures. The ideal intervention would
reduce the binge eating behavior, improve psychological disturbances such as
depression and overconcern with body image, and promote weight loss. To
date, most treatment studies have focused on the behavioral and psychological
components, leaving weight loss as a secondary goal. As this disorder is
relatively newly codified in the DSM-IV, only a small number of controlled
medication studies have been published.
Antidepressants
Based on the efficacy of antidepressants in the management of BN and the
similarities between these two disorders, most research has focused on
antidepressants. An early placebo-controlled trial of “nonpurging BN”
found that desipramine was effective in the short-term reduction of binge
eating (82). However, symptoms reemerged 4 weeks after medication
discontinuation. A randomized, controlled trial has also shown imipramine to
be effective (83). More recently, two studies have found benefits from SSRIs. In
one randomized controlled trial of fluvoxamine (50–300 mg/day) (84) and one
randomized controlled trial of sertraline (50–200 mg/day) (85), the authors
reported a significant reduction in binge eating behavior as well as a decrease in
BMI in the medication groups as compared with placebo. As with most of the
medication trials in eating disorders, these short-term results do not provide
information as to sustained benefit of medications.
Other Agents
As with AN and BN, a number of other classes of medication have received
some attention as being of possible use in the management of BED. Topiramate
is the latest to show promise. In one open label-study of 13 patients with BED,
topiramate was associated with weight loss (86). Appolinario et al. (87)
reported a case study of a patient who responded to topiramate after other
treatments had been unsuccessful. McElroy et al. (88) conducted a placebo-
controlled, double-blind trial of topiramate in 61 patients with BED.
Topiramate-treated subjects showed significantly greater reductions in binge
frequency, binge day frequency, and other measures of symptom severity, as
well as significant reduction in BMI.
Trials of dexfenfluramine yielded some promising results. Dexfenfluramine
was associated with a decrease in binge eating, and when used in combination
with phentermine, with weight loss, as well (89,90). However,
dexfenfluramine has since been withdrawn from the market due to its
association with cardiac valve abnormalities. An open trial of sibutramine, an
appetite suppressant approved for the management of obesity, suggested that
use of this agent was associated with improvement in both binge eating and
weight loss (91). A controlled trial has been conducted, but results are not yet
available. The opiate antagonist naltrexone was studied in a randomized placebo-
controlled trial which also included an imipramine arm (89). While both
medication groups showed improvement, there was no difference from
placebo.
Combination Treatment
Antidepressant medications appear to provide short-term benefit in the
management of BED, but the benefits do not appear to be sustained beyond the
discontinuation of the medication. Several studies have examined the benefits of
combined treatment, but, at present, the data are insufficient to support clear
conclusions (83,92–95).
CONCLUSIONS AND TREATMENT

RECOMMENDATIONS
Anorexia Nervosa
Psychopharmacological interventions have not been shown to provide
significant benefit to underweight patients with AN. The mainstays of treatment
are nonpharmacological, and focus on nutritional rehabilitation and relapse
prevention. There is preliminary evidence that fluoxetine may be of benefit for

relapse prevention after patients have regained to a normal or near-normal
weight.
Bulimia Nervosa
The data on the use of antidepressants in the treatment of BN are convincing in
indicating that that fluoxetine is safe and beneficial. While it is likely that other
SSRIs would be effective, only fluoxetine has been examined in
placebocontrolled trials, and should be used in a dose of 60 mg/day. Most
patients can be rapidly titrated to this dose over the course of a week. There are
no available data to guide treatment for relapse prevention or to suggest
recommended length of treatment. Bupropion is not recommended in the
management of BN because of the risk of seizure. There are consistent
indications that medications modestly enhance the benefits of psychological
treatment.

While emerging data suggest that SSRIs and, perhaps, antiobesity agents may
provide some benefits, at present there is insufficient information to make firm
treatment recommendations regarding the use of medication for BED.
REFERENCES
1. American Psychiatric Association. Practice guidelines for the treatment of patients
with eating disorders (Revision). Am J Psychiatry (Supplement) 2000; 157(l):l-39.
2. Lacey J, Crisp AH. Hunger, food intake and weight: the impact of clomipramine
on a refeeding anorexia nervosa population. Postgrad J Med 1980; 56(suppl 1):
79–95.
3. Biederman J, Herzog D, Rivinus T, et al. Amitriptyline in the treatment of
anorexia nervosa: a double-blind, placebo-controlled study. J Clin
Psychopharmacol 1985; 5, 10–16.
4. Halmi K, Eckert E, LaDu T, Cohen J. Anorexia nervosa: treatment efficacy of
cyproheptadine and amitriptyline. Arch Gen Psychiatry 1986; 43:177–181.
5. Wilens T, Biederman J, Baldessarini R, et al. Cardiovascular effects of therapeutic
doses of tricyclic antidepressants in children and adolescents. J Am Acad Child
Adolesc Psychiatry 1996; 35:1491–1501.
6. Ferguson J. Treatment of an anorexia nervosa patient with flouxetine. Am J
Psychiatry 1987; 144:1239.
7. Gwirtsman H, Guze B, Yager J, Gainsley B. Fluoxetine treatment of anorexia
nervosa: an open clinical trial. J Clin Psychiatry 1990; 51:1378–1382.
8. Attia E, Haiman C, Walsh T, Flater S. Does fluoxetine augment the inpatient

treatment of anorexia nervosa? Am J Psychiatry April 1998; 155(4):548–551.
nervosa: an historical case-control study. J Child Adol Psychopharmacol 1999; 9
(3):195–201.
10. Kaye W, Nagata T, Weltzin T, et al. Double-blind placebo-controlled
administration of fluoxetine in restricting and restricting-purging-type anorexia
11. Strober M, Freeman R, DeAntonio M, Lampert C, Diamond J. Does adjunctive
fluoxetine influence the post-hospital course of anorexia nervosa? A 24-month
prospective, longitudinal follow-up and comparison with historical controls.
12. Kaye W, Ebert M, Raleigh Mea. Abnormalities in CNS monoamine metabolism in
anorexia nervosa. Arch Gen Psychiatry 1984; 41:350–355.
13. Kaye W, Gendall K, Strober M. Serotonin neuronal function and selective
serotonin reuptake inhibitor treatment in anorexia and bulimia nervosa. Biol
Psychiatry 1998; 44:825–838.
14. Attia E, Mayer L, Killory E. Medication response in the treatment of patients with
anorexia nervosa. J Psychiat Pract 2001; 7:157–162.
15. Dally P, Sargant W. A new treatment of anorexia nervosa. British Medical Journal
1960; 1:1770–1773.
16. Vandereycken W, Pierloot R. Pimozide combined with behavior therapy in the short
term treatment of anorexia nervosa. Acta Psychiatr Scand 1982; 66:445–450.
17. Vandereycken W. Neuroleptics in the short-term treatment of anoreixia nervosa: a
double-blind placebo-controlled study with sulpiride. Br J Psychiatry 1984; 144:
288–292.
18. Jensen V, Mejlhede A. Anorexia nervosa: treatment with olanzapine. Br J
Psychiatry July 2000; 177:87.
19. La Via M, Gray N, Kaye W. Case reports of olanzapine treatment of anorexia
nervosa. Intl J Eat Disord 2000; 27:363–366.
20. Mehler C, Wewetzer C, Schulze U, Warnke A, Theisen F, Dittmann R. Olanzapine
in children and adolescents with chronic anorexia nervosa: a study of five cases. Eur
21. Vigersky R, Loriaux D. The effect of cyproheptadine in anorexia nervosa: a double-
blind trial. In: Vigersky R, ed. Anorexia Nervosa. New York: Raven Press, 1977:
349–356.
22. Goldberg 8, Halmi K, Eckert E, Casper R, Dacis J. Cyproheptadine in anorexia
23. Stacher G, Kiss A, Wiesnagrotzki S, et al. Oesophageal and gastric motility
disorders in patients categorized as having primary anorexia nervosa. Gut 1986; 27:
1120–1126.
24. Domstad P, Shis W, Humphries L, DeLand F, Digenis G. Radionuclide gastric

emptying studies in patients with anorexia nervosa. J Nucl Med 1987; 28:
816–819.
25. Modolsfsky H, Jeuniewic N. Preliminary report of metoclopramine in anorexia
nervosa. In: Vigersky R, ed. Anorexia Nervosa. New York: Raven Press, 1977:
373–375.
26. Stacher G, Abatzi-Wenzel T, Wiesnagrotzki S, Bergmann H, Schneider C,
Gaupmann G. Gastric emptying, body weight and symptoms in primary anorexia
nervosa: long term effects of cisapride. Br J Psychiatry 1993; 162:398–402.
27. Szmukler G, Young G, Miller G, Lichtenstein M, Binns D. A controlled trial of
cisapride in anorexia nervosa. Int J Eat Disord 1995; 17:345–357.
28. Kaye W. Opioid antagonist drugs in the treatment of anorexia nervosa. In:
Garfinkel P, Garner D, eds. The Role of Drug Treatments for Eating Disorders.
New York: Brunner/Mazel, 1987:150–160.
29. Marrazi M, Bacon J, Kinzie J, et al. Naltrexone use in the treatment of anorexia
nervosa and bulimia nervosa. Int J Clin Psychopharmacol 1995; 10:163–172.
30. Katz R, Keen C, Litt I, Hurley L, Kellams-Harrison K, Glader L. Zinc deficiency in
anorexia nervosa. J Adol Health Care 1987; 8:400–406.
31. Lask B, Fosson A, Rolfe U, Thomas S. Zinc deficiency and childhood-onset
anorexia nervosa. J Clin Psychiatry 1993; 54:63–66.
32. Birmingham C, Goldner E, Bakan R. Controlled trial of zinc supplementaion in
anorexia nervosa. Int J Eat Disord 1994; 15:251–255.
33. Safai-Kutti S. Oral zinc supplementation in anorexia nervosa. Acta Psychiatr Scand
1990; 361:814–817.
34. Gross H, Ebert M, Faden V, Goldberg S, Nee L, Kaye W. A double-blind
controlled trial of lithium carbonate in primary anorexia nervosa. J Clin
Psychopharmacol 1981; 1:376–381.
35. Gross H, Ebert M, Faden V, et al. A double-blind trial of delta9-
tetrahydrocannacinol in primary anorexia nervosa. J Clin Psychopharmacol 1983; 3:
165–171.
36. Kiblanski A, Biller B, Schoenfeld D, Herzog D, Saxe V. The effects of estrogen
administration on trabecular bone loss in young women with anorexia nervosa.
J Clin Endocrinol Metab 1995; 80:898–904.
37. Kotler L, Devlin M, Walsh BT. Eating disorders and related disturbances. 2002;
410–430.
38. Walsh B, Devlin M. The pharmacologic treatment of eating disorders. Psychiatr
Clin North Am 1992; 15:149–160.
39. Mitchell J, Raymond N, Specke S. A review ofthe controlled trials of
pharmacotherapy and psychotherapy in the treatment of bulimia nervosa. Int J Eat
Disord 1993; 14:229–247.
40. Agras W, Dorian B, Kirkley B, et al. Imipramine in the treatment of bulimia: a
double-blind controlled study. Int J Eat Disord 1987; 6:29–38.
41. Barlow J, Bloluin J, Blouin A, Perez E. Treatment of bulimia with desipramine: a
double-blind crossover study. Can J Psychiatry 1988; 33:129–133.
42. Hughes P, Wells L, Cunningham C, et al. Treating bulimia with desipramine: a

double-blind, placebo-controlled trial. Arch Gen Psychiatry 1986; 43:182–186.
43. Mitchell J, Groat R. A placebo-controlled, double-blind trial of amitriptyline in
bulimia. J Clin Psychopharmacol 1984; 4:186–193.
44. Pope HJ, Hudson J, Jonas Jea. Bulimia treated with imipramine: a
placebocontrolled, double-blind study. Am J Psychiatry 1983; 140:554–558.
45. Walsh B, Hadigan C, Devlin M, Gladis M, Roose S. Long-term outcome of
antidepressant treatment for bulimia nervosa. Am J Psychiatry 1991; 148:
1206–1212.
46. Kennedy S, Piran N, Warsh J, et al. A trial of isocarbozacid in the treatment of
bulimia nervosa. J Clin Psychopharmacol 1988; 8:391–396.
47. Walsh B, Stewart J, Roose S, Gladis M, Glassman A. Treatment of bulimia with
phenelzine: a double-blind placebo controlled study. Arch Gen Psychiatry 1984;
41:1105–1109.
48. Walsh B, Gladis M, Roose Sea. Phenelzine vs placebo in 50 patients with bulimia.
49. Fluoxetine Bulima Nervosa Collaborative (FBNC) Study Group. Fluoxetine in the
treatment of bulimia nervosa: a multi-center, double-blind, placebo-controlled
50. Goldstein D, Wilson M, Thompson Vea. Fluoxetine Bulimia Nervosa Collaborative
Study Group. Long term fluoxetine treatment of bulimia nervosa. Br J Psychiatry
1995; 166:660–666.
51. Pope HJ, Keck PE Jr, McElroy S, et al. A placebo-controlled study of trazodone in
bulimia nervosa. J Clin Psychopharmacol 1989; 9:159–254.
52. Roberts J, Lydiard R. Sertraline in the treament of bulimia nervosa [letter]. Am J
Psychiatry 1993; 150:1753.
53. Ayuso-Guttierrez J, Palazon M, Ayuso-Mateos J. Open trial of fluvoxamine in the
treatment of bulimia nervosa. Int J Eat Disord 1994; 15:245–249.
54. Spigset O, Pleym H. Case report of successful treatment of bulimia nervosa with
fluvoxamine [letter]. Pharmacopsychiatry 1991; 24:180.
55. Fichter M, Kruger R, Rief W, Hollan R, Dohne J. Fluvoxamine in prevention of
relapse in bulimia nervosa: effects of eating specific psychophathology. J Clin
56. Romano S, Halmi K, Sarkar N, Koke S, Lee J. A placebo-controlled study of
fluoxetine in continued treatment of bulimia nervosa after successful acute
fluoxetine treatment. Am J Psychiatry 2002; 159:96–102.
57. Horne R, Ferguson J, Pope HJ, et al. Treatment of bulimia with buproprion: a
multicenter controlled trial. J Clin Psychiatry 1988; 49:262–266.
58. Walsh B. Psychopharmacologic treatment of bulimia nervosa. J Clin Psychiatry
1991; 52(10, suppl):34–48.
59. Bacaltchuk J, Hay P, Mari J. Antidepressants versus placebo for the treatment of
bulimia nervosa: a systematic review. Aust N Z J Psychiatry 2000; 34:310–317.
60. Hsu L, Clement L, Santhuse R, Ju E. Treatment of bulimia nervosa with lithium
carbonate, a controlled study. J Nerv Ment Dis 1991; 179:351–355.
61. Leitenberg H, Rosen J, Wolf J, Vara L, Detzwwer M, Srebnik D. Comparison of

cognitive-behavior therapy and desipramine in the treatment of bulimia nervosa.
Behav Res Ther 1994; 32:37–45.
62. Mitchell J, Pyle R, Hatsukami D, Pomeroy C, Zimmerman R. A comparison study
of antidepressants and structured group therapy in the treatment of bulimia
63. Wermuth B, Davis K, Hollister L, Stunkard A. Phenytoin treatment of the
bingeeating syndrome. Am J Psychiatry 1977; 134:1249–1253.
64. Kaplan A, Garfinkel P, Darby P, Garner D. Carbamazepine in the treatment of
bulimia. Am J Psychiatry 1983; 140:1225–1226.
65. Knable M. Topiramate for bulimia nervosa in epilepsy. Am J Psychiatry 2001; 158:
322–323.
66. Hoopes S, Reimherr F, Kamin M, Karvois D, Rosenthal N, Karim R. Topiramate
treatment of bulimia nervosa. Scientific and Clinical Report Session 2: Eating
Disorders. Philadelphia, PA: American Psychiatric Association, May 2002 (Annual
meeting).
67. Liebowitz S. The role of serotonin in eating disorders. Drugs 1990; 39(suppl 3):
33–48.
68. Krahn D, Mitchell J. Use of L-tryptophan in treating bulimia (letter). Am J
Psychiatry 1985; 142, 1130.
69. Blouin A, Blouin J, Bushnik T, Zuro C, Mulder E. Treatment of bulimia with
fenfluramine and desipramine. J Clin Psychopharmacol 1988; 8:261–269.
70. Fahy T, Eisler I, Russell G. A placebo-controlled trial of d-fenfluramine in bulimia
71. Russel G, Checkley S, Feldman J, Eisler I. A controlled trial of d-fenfluramine in
bulimia nervosa. Clin Neuropharmacol 1988; 11:8146–8159.
72. Faris P, Kim S, Meller W, et al. Effect of ondansetron, a 5-HT3 receptor
antagonist, on the dynamic association between bulimic behaviors and pain
thresholds. Pain 1998; 77:297–303.
73. Faris P, Kim S, Meller W, et al. Effect of decreasing afferent vagal activity with
ondansetron on symptoms of bulimia nervosa: a randomised, double-blind trial.
Lancet 2000; 355:792–797.
74. Jonas J, Gold M. Treatment of antidepressant-resistant bulimia with naltrexone. In
J Psychiatry Med 1987; 16:305–309.
75. Mitchell J, Christenson G, Jennings J, et al. A placebo-controlled, double-blind
crossover study of naltrexone hydrochloride in outpatients with normal-weight
bulimia. J Clin Psychopharmacol 1989; 9:94–97.
76. Keel PK, Mitchell J, Davis T, Crow 8. Long-term impact of treatment in women
diagnosed with bulimia nervosa. Int J Eat Disord 2000; 31:151–158.
77. Agras W, Arnow B, Schneider J, et al. Pharmacologic and cognitive-
behavioral treatment for bulimia nervosa: a controlled comparison. Am J
Psychiatry 1992; 49:82–87.
78. Walsh B, Wilson G, Loeb K, et al. Medication and psychotherapy in the treatment
of bulimia nervosa. Am J Psychiatry 1997; 154:523–531.
79. Goldbloom D, Olmstead M, Davis R, et al. A randomized controlled trial of

fluoxetine and cognitive behavioral therapy for bulimia nervosa: short term outcome.
Behav Res Ther 1997; 35:803–811.
80. Beumont P, Russell J, Touyz S, et al. Intensive nutritional counseling in bulimia
nervosa: a role for supplementation with fluoxetine? Aust N Z J Psychiatry 1997;
31:514–524.
81. Bacaltchuk J, Trefiglio R, Oliveira I, Hay P, Lima M, Mari J. Combination of
antidepressants and psychological treatments for bulimia nervosa: a systematic
review. Acta Psychiatr Scand June 26 1999; 101:256–267.
82. McCann U, Agras W. Successful treatment of nonpurging bulimia nervosa with
desipramine: a double-blind, placebo-controlled study. Am J Psychiatry 1990; 147:
1509–1513.
83. Laederach-Hofman K, Graf C, Lippuner K, Lederer S, Michel R, Schneider M.
Imipramine and diet counseling with psychological support in the treatment of
obese binge-eaters: a randomized, placebo-controlled double-blind study. Int J Eat
Disord 1999; 26:231–244.
84. Hudson J, McElroy S, Raymond N, et al. Fluvoxamine in the treatment of bingeeating
disorder: a multicenter placebo-controlled, double-blind trial. Am J Psychiatry
1998; 155:1756–1762.
85. McElroy S, Casuto L, Nelson E, et al. Placebo-controlled trial of sertraline in the
treatment of binge eating disorder. Am J Psychiatry 2000; 157:1004–1006.
86. Shapira N, Goldsmith T, McElroy S. Treatment of binge-eating disorder with
topiramate: a clinical case series. J Clin Psychiatry 2000; 61:368–372.
87. Appolinario J, Coutinho W, Fontenelle L. Topiramate for binge-eating disorder.
Am J Psychiatry 2001; 158:967–968.
88. McElroy S, Arnold L, Shapira N, et al. Topiramate in the treatment of binge eating
disorder associated with obesity: a randomized, placebo-controlled trial. Am J
Psychiatry February 2003; 160(2):255–261.
89. Alger S, Schwalberg M, Bigaouette J, et al. Effect of a tricyclic antidepressant and
opiate antagonist on binge-eating in normal weight bulimic and obese bingeeating
subjects. Am J Clin Nutr 1991; 53:865–871.
90. Stunkard A, Berkowits R, Tanrikut C, Reiss E, Young L. d-Fenfluramine treatment
of binge eating disorder. Am J Psychiatry 1996; 153:1455–1459.
91. Appolonario JC, Godoy-Matos A, Fontenelle LF, et al. An open trial of
sibutramine in obese patients with binge eating disorder. J Clin Psychiatry 2002;
63:28–30.
92. Agras W, Telch C, Arnow B, et al. Weight loss, cognitive-behavioral, and
desipramine treatments in binge eating disorder. An additive design. Behav Ther
1994; 25:225–238.
93. Devlin M, Goldfein J, Carino J, Wolk S. Open treatment of overweight
binge eaters with phentermine and fluoxetine as an adjunct to cognitive-behavioral
therapy. Int J Eat Disord 2000; 28:325–332.
94. Marcus M, Wing R, Ewing L, Kern E, Gooding W, McDermott M. A
doubleblind, placebo controlled trial of fluoxetine plus behavior modification in the
treatment of obese binge eaters and non-binge eaters. Am J Psychiatry 1990; 147:
876–881.
95. Grilo C, Masheb R, Heninger G, Wilson G. Controlled comparison of cognitive
behavorial therapy and fluoxetine for binge eating disorder. In: Mitchell J, ed.
Scientific II Session. International Conference on Eating Disorders. BED and
Obesity. Boston: Academy for Eating Disorders, April 28, 2002.
22
Eating Disorders, Victimization, and
Comorbidity: Principles of Treatment
Timothy D.Brewerton
Medical University of South Carolina
Charleston, South Carolina, U.S.A.
Comorbidity is the rule rather than the exception when it comes to eating
disorders (EDs), particularly bulimia nervosa (BN). The common types of
comorbid psychiatric disorders (on axes I and II) and comorbid medical
disorders (on axis III) are reviewed in Chapters 8, 9, and 10, respectively.
Comorbidity is simply the coexistence of one disorder with another disorder in
the same person. When one disorder, e.g., major depressive disorder (MDD),
occurs with another in the present time this constitutes current prevalence or
history, whereas when another disorder occurs at any point during the lifetime
of the individual it constitutes lifetime prevalence or previous history of that
comorbid disorder. While current comorbidity is more strongly the purview of
clinicians managing acute EDs, particularly in an inpatient setting, obtaining the
lifetime history of all forms of comorbidity is extremely important in order for
the clinician to see the “big picture,” the “forest” and the “trees,” and the
wholistic, developmental, and biopsychosocial perspective.
The themes of this chapter are that (a) a cluster of comorbid disorders and
their symptoms co-occur together more often than chance would dictate, and
that this link is highly associated with a history of victimization and subsequent
PTSD; and (b) the victimization and PTSD must be specifically and adequately
addressed in order to optimize full recovery from not only the ED but all
associated comorbid disorders. Victimization usually refers specifically to major
physical boundary violations, such as rape, molestation, and aggravated assault.
All of these common events are crimes in all developed countries, and there
may be important legal ramifications for these patients to process in therapy. Of
course, victimization may also involve other traumatic experiences, such as
witnessing a homicide, severe emotional abuse, and neglect (physical and
emotional).
Just as there is a spectrum of trauma and victimization, there is also a
spectrum of trauma-related disorders (1–4). The disorders that group together
within this trauma-related spectrum include eating (especially BN), affective
(especially MDD), anxiety (especially post-traumatic stress disorder, PTSD),
512 BREWERTON
substance use, dissociative, somatoform, impulse control, and disruptive

disorders on axis I and cluster B disorders, especially borderline personality
disorder (BPD), on axis II. On axis III, most of the major leading illnesses that
cause significant mortality and morbidity in the United States, including obesity,
have been linked to major childhood abuse and adversity (5,6). Mounting
evidence attests to the powerful psychobiological underpinnings of trauma on
multiple developing and evolving systems, notably the central nervous system
(CNS), the autonomic nervous system (ANS), as well as the immune,
endocrine, and cardiovascular systems (4,7).
Axis II (personality) disorders are not reliably diagnosed quickly, or in the
face of significant axis I psychopathology, or prolonged semistarvation. This has
been evident since the famous Minnesota semistarvation experiments in the
1940s by Keys and colleagues (see Chapter 8) (8). Taken together, the scientific
literature strongly supports an aggressive focus on axis I disorders first, which
then often leads to improvement or even disappearance in apparent axis II
disorders. A more prolonged evaluation allows the clinician to collect and
clarify further information not only from the patient but from family members,
previous treatment providers, and other potential sources, e.g., teachers,
coaches, friends, extended family. It is only after an extended evaluation has
been completed and after improvement in axis I psychopathology has occurred
that a reliable and accurate axis II diagnosis can be made with any certainty.
Some of the best data regarding prevalence rates of common comorbid
psychiatric disorders in the United States come from the National Comorbidity
Study (9). In a large representative sample of more than 8000 U.S. residents
interviewed, lifetime prevalence of any axis I psychiatric disorder (not including
EDs) was 48%. The lifetime prevalence for having any type of anxiety disorder
was 24.9%, for having any affective disorder was 19.3%, and for having any
substance abuse and/or dependence was 26.6%. The lifetime prevalence of
PTSD was 7.8% (10% in women, 5% in men).
In the National Women’s Study (NWS), there were similar findings in a
random sample of more than 4000 women who agreed to take part in a highly
structured interview. The lifetime prevalence of PTSD was found to be 12.5%
in this representative sample of U.S. women (10). The anonymity of this
telephone survey may have actually allowed for more open and honest
disclosure than a “face-to-face” interview.
A central and common finding of both the National Comorbidity Study and
the NWS was that psychiatric comorbidity was highly associated with a history
of serious victimization and especially a lifetime history of PTSD. In the National
Comorbidity Study the odds ratios for major axis I disorders ranged from 2.4 to
4.5, and similarly increased rates were found in the NWS (1,11). In another
large, well-controlled study of 1411 female twins, childhood sexual abuse,
EATING DISORDERS, VICTIMIZATION, COMORBIDITY 513
especially genital rape, predicted an odds ratio of 5.62 for the development of
BN and 5.05 for the development of two or more psychiatric disorders (12). This
was an especially important and powerful study given that it controlled for
genetic influences.
RELATIONSHIP OF EATING DISORDERS TO

TRAUMA HISTORY AND PTSD
It is well established that the role of trauma or victimization is a significant risk
factor for BN in association with comorbid axis I psychopathology (13). In this
study, two centers (University of North Dakota and Medical University of South
Carolina) reviewed all available studies to date using strict a priori inclusion and
exclusion criteria. Six hypotheses were tested, four of which were supported by
the data. These included the following:
1. Childhood sexual abuse (CSA) is associated with BN, which was supported
by 8 of 12 studies.
2. CSA is more common in BN than AN, which was supported by 4 of 6
studies.
3. CSA is a specific risk factor for EDs, which was not at all supported.
4. CSA is associated with greater severity, which was not supported by the
data (but only three studies were included).
5. Particular features of CSA are associated with ED symptoms, which was
supported by 4 of 5 studies; these included decreased social competence;
poor maternal relationship; unreliable parenting; severity of CSA, and
presence of lifetime PTSD, which was a finding from the NWS (14).
6. CSA is associated with comorbidity in ED subjects, which was supported
by 5 of 6 studies.
One of the studies included in this review was the NWS (14), which remains
the most comprehensive study of the relationship of trauma history and PTSD to
EDs and comorbidity in a large national sample. Structured telephone interviews
of a representative sample of over 4000 U.S. women from four stratified
geographical areas assessed detailed histories of crime victimization experiences
(rape, molestation, attempted sexual assault, and aggravated assault), PTSD,
MDD, EDs and substance abuse/dependence using DSM-IIIR and DSM-IV
(binge eating disorder, BED) criteria. The lifetime prevalence rates for the EDs
were as follows: BN = 2.4%, BED = 1.0%; AN = 0.23%. In comparing BN
and non-BN/BED subjects who completed the survey, the following lifetime
prevalence rates were obtained: completed rape: 26.6% v. 3.3% (p < 0.01);
contact sexual molestation: 22.0% v. 12.0% (p < 0.05); attempted assault:
514 BREWERTON
26.8% v. 8.4%, p < 0.001; any direct crime victimization (one or more of the
above): 54.4% v. 31.0% (p < 0.001); lifetime PTSD: 37% v. 12% (p < 0.001);
current PTSD: 21% v. 4% (p < 0.001). In comparing BED versus non-BN/BED
subjects, there were no significant differences except for lifetime PTSD:
22% v. 12% (p < 0.01). In the AN group there were no reports of rape,
molestation, aggravated assault, or PTSD at all. The age of first rape occurred
before the age of first binge in 84% of all BN cases. The corresponding numbers
for only adolescent rapes (12–17 years) and only child rapes (≤11 years) were
96% and 100%, respectively. These data provide substantial validity to the
notion that victimization is a causative risk factor for BN, albeit a nonspecific
one. In addition, the age of first binge for BN subjects was significantly earlier in
cases of rape resulting in PTSD compared to those with rape without PTSD or
no rape. Interestingly, the prevalence rates for BN were significantly higher in
those subjects with histories of rape with PTSD (10.4%) compared to those
with rape without PTSD (2.0%) or those with no rape (2.0%), which strongly
suggests that PTSD rather than prior abuse per se best predicts the development
of BN. In the NWS links between trauma, dissociative symptoms, and BN were
also found (15). BN subjects had significantly more “forgetting” or psychogenic
amnesia of traumatic events (27%) than BED subjects (12%) or non-ED
subjects. This was defined as an endorsement by subjects that they forgot all or
part of traumatic events. Multiple linear regression using psychogenic amnesia
as the dependent variable identified the following significant variables (in
decreasing order of significance): lifetime PTSD, childhood rape, lifetime major
depression, molestation, emotional problems in the family, laxative abuse, total
number of victimization experiences, age, and vomiting (15). Based on these
and other data, it was hypothesized that purging behaviors, such as vomiting and
laxative abuse, rather than binge eating per se, are maladaptive behaviors linked
to PTSD and MDD that facilitate avoiding, numbing, and forgetting traumatic
memories.
Since the publication of the review by Wonderlich and colleagues, there have
been two large reviews on sexual abuse and EDs, both of which support the
importance of this link in at least a subgroup of ED patients (16,17).
Of course, not all patients with EDs have been victimized. However, the
index of suspicion rises substantially as the number of comorbid psychiatric
disorders rises. In the NWS there were clear links between trauma, PTSD, and
comorbidity with affective, anxiety, and substance use disorders (SUDs). A very
robust linear relationship exists between the number of comorbid axis I
diagnoses and the percentage of subjects with histories of child rape, rape at any
age, and any direct victimization experience (unpublished data). 100% of the
subjects with all four diagnoses (BN, MDD, PTSD, SUD) had experienced at
least one major direct victimization experience in their life (rape, molestation,
aggravated assault). Fifty percent of these same subjects with all four diagnoses
reported a childhood rape (≤ 17 years old).
There have been a number of more recent studies on EDs, victimization, and
comorbidity, all of which lend further support to this relationship. Some of
these studies will be reviewed below.
Kenardy showed links between past sexual and/or physical abuse and weight
dissatisfaction and/or disordered eating in young and middle-aged women (18),
while Romans reported higher rates of EDs in women who have experienced
CSA (19). In this study, the factors that increased the risk of developing an ED
in women who had experienced CSA were belonging to a younger age cohort,
experiencing menarche at an early age, and experiencing high paternal
overcontrol. Low maternal care was specifically associated with the
development of AN, whereas early age of menarche differentiated women with
BN. Younger age and early age of menarche also differentiated the CSA + ED
women from the psychiatric comparison group.
In a clinical study that supports the link between PTSD and EDs, Gleaves and
colleagues stated that 74% of 293 women admitted to a residential treatment
center who completed a PTSD symptom scale reported a traumatic experience
and 52% reported symptoms consistent with a diagnosis of PTSD (20). Of
112 AN patients, 47% met PTSD criteria, and of 103 BN patients 62% met
PTSD criteria. In a clinical sample of recovered BN patients, abused subjects
showed a trend toward more frequent lifetime diagnoses of PTSD and substance
dependence compared with nonabused subjects (21). These results suggest that
abusive experiences may be associated with some of the psychopathology of BN,
particularly that which is related to anxiety, substance abuse, and more severe
core ED pathology. In another study, women comorbid for BN and substance
dependence were found to have the highest frequency and the most severe
histories of sexual abuse compared to bulimic women without substance
dependence (22).
In an Australian sample of hospitalized patients, nearly one-half of ED
patients reported a history of child sexual abuse and one-quarter reported child
physical abuse, and these rates were significantly higher than those for the
control group (23). In addition, dissociative experiences were found to be
common and strongly associated with history of abuse and self-mutilation (24).
In a Japanese study of ED patients and controls, physical punishment histories
tended to be more prevalent among patients with AN-BP or BN than among
AN-R or controls. AN-BP and BN patients with physical punishment histories
had twofold higher scores on the Dissociative Experiences Scale (DES) and
significantly more frequent histories of self-mutilation (67% v. 33%) compared
with patients without such histories (25). Multi-impulsivity in Japanese ED
patients was associated with suicide attempts, self-mutilation, BPD, and parental
516 BREWERTON
loss (26). In a study of Italian ED patients and controls, self-destructive behavior

appeared to be the most important predictor of a history of sexual and/or
physical abuse in ED patients (27).
Some interesting data have emerged that support the association between
EDs, victimization, and comorbidity in males, both men and boys. In a sample of
more than 24,000 U.S. veterans, Striegel-Moore and colleagues found that
women and men with EDs had significant comorbidity with substance abuse and
mood disorders (28). In addition, women with EDs had higher rates of anxiety
disorders and BPD. Lipschitz and colleagues reported that hospitalized
adolescent males with PTSD were more likely to have comorbid EDs as well as
other anxiety disorders and somatization (29). In a survey of a statewide
representative sample of 9943 students in grades 7, 9, and 11 in Connecticut,
Neumark-Sztainer and coworkers reported increased disordered eating among
those who reported sexual or physical abuse, as well as low levels of family
communication and parental caring and expectations (30). The links between
abuse and disordered eating persisted even after controlling for differences in
familial and psychosocial factors. The odds ratio for the development of
disordered eating following sexual abuse was 1.99 for girls and 4.88 for boys,
whereas the odds ratio following physical abuse was 2.0 for girls and 1.95 for
boys.
Other data in children and adolescents lend further strength to these
interrelationships (31–35). In one well-controlled study, Wonderlich and
colleagues compared 20 sexually abused girls with 20 age-matched nonabused
control girls aged 10–15 years who completed psychometric instruments,
including the Kids’ Eating Disorders Survey, the McKnight Risk Factor Survey,
and the Body Rating Scale for Adolescents (31). Abused children, in comparison
with control children: (a) had significantly higher rates of weight dissatisfaction,
purging, and dieting behavior; (b) ate significantly less when emotionally upset;
and (c) were less likely to exhibit perfectionistic tendencies but more likely to
desire thinner body types. This was the first controlled study to support similar
findings in adults.
In a study of three samples of ninth through twelfth graders including over
5000 students, sexual victimization (independent of physical victimization) was
associated with weight regulation in adolescent girls, and it was also associated
with more extreme forms as well as multiple forms of weight regulation. Urban
girls with sexual victimization had significantly higher rates of purging than
urban girls without such experiences (32).
Other studies in children and adolescents have been published demonstrating
significant associations between binge-purge behavior and sexual or physical
abuse. This has been shown in large national samples of adolescents in both the
United States (33) and Sweden (34). These links have also been shown to
persist in sexually abused children long past the time of abuse (35).
Data supporting a role for various forms of abuse as a risk factor for BED
have also emerged in the literature over the last several years. Grilo and
colleagues showed higher rates of childhood psychological, physical, and sexual
maltreatment in outpatients with BED (36). A total of 83% of BED patients
reported some form of childhood maltreatment, including 59% reporting
emotional abuse, 36% physical abuse, 30% sexual abuse, 69% emotional
neglect, and 49% physical neglect. There were no differences in the distribution
of any form of childhood maltreatment by gender, obesity status, body mass
index, binge eating, or attitudinal features. Only physical neglect was associated
with dietary restraint in women, whereas emotional abuse was associated with
greater body dissatisfaction, higher depression, and lower self-esteem in men
and women, and sexual abuse was associated with greater body dissatisfaction in
men. Another study also identified other forms of victimization besides abuse,
like bullying and discrimination, as risk factors for BED (37). Unfortunately,
these studies did not report on rates of PTSD in these patients.
Other investigators have also expanded the spectrum of abuse that may
contribute to the development of EDs, including neglect (38), childhood
emotional abuse (39,40), adverse family background (41), and extreme food
deprivation (42).
In terms of studies of mediating variables between prior child abuse and BN,
research have shown that impulsivity and core beliefs involving shame, self-
esteem, and perceived control are significant mediating variables to consider in
terms not only of etiology but of treatment planning as well (43–48).
RELATIONSHIP OF COMORBID DISORDERS TO

TRAUMA HISTORY AND PTSD
The most common axis I disorders that are comorbid with each of the major EDs
have been extensively reviewed in Chapter 8. These include affective and
anxiety disorders for AN, BN, and BED, as well as SUDs for BN, AN-BP,
perhaps BED, but not AN-R. Other data support associations between BN and
dissociative disorders (DDs) or symptoms, somatoform disorders (SDs),
impulse control disorders (ICDs), and disruptive behavior disorders (DBDs).
Each of these types of disorder has been strongly linked to history of
victimization and PTSD in the scientific literature, which has been reviewed
elsewhere (46).
Trauma may therefore serve as an “organizing principle” when thinking about
etiology from a biopsychosocial perspective in many (but not all) comorbid ED
patients. However, the greater the degree of comorbidity, the greater are the
518 BREWERTON
chances of a trauma history having played a significant role in precipitating the

onset of the overall course of mental illness. Trauma-related disorders may
share common underlying factors that account for such interrelationship, e.g.,
dysregulation in neuropsychobiological mechanisms, such as serotonin and other
neurotransmitter disturbances (see Chapter 11), as well as common cognitive
schema involving issues of self-esteem, control, guilt, and shame.
Affective Disorders
A lifetime prevalence of major depressive disorder (MDD) has been found in
association with AN and BN in 24–88% of reported cases, depending on the
study (see Chapter 8). To a much lesser extent, this pattern of higher affective
disorders in ED patients runs true for dysthymia (19–20%) and bipolar disorder
(4–13%) as well, particularly BPD type II with a seasonal pattern. Depressive
symptoms that do not meet full criteria for an affective disorder are also
common in ED patients (depression NOS).
Each of these affective disorders has been reported to have higher than
expected rates of a history of trauma or victimization, and more often than not
the trauma occurred during childhood or adolescence. It is important for the
clinician to realize that fully two-thirds of all rapes of women in the United
States occur before the age of 18 (NWS). The frequency of rape in decreasing
order is in girls aged 12–17 years, then in girls at or below 11 years old, then in
women ages 18–24 years. After this age period, the frequency of rape decreases
with age precipitously. Childhood sexual and physical abuse has been found to
be highly associated with MDD and affective disturbances in general (7,49–55),
and Gladstone reported that depressed patients with childhood abuse histories
were more likely associated with BPD (see “Axis II Comorbidity” below) (54).
In addition, a history of physical or sexual abuse in childhood has been
found to be particularly associated with MDD with reversed neurovegetative
features, e.g., hyperphagia and hypersomnia, whether or not manic subjects are
included (55). A significant relationship between mania and childhood physical
abuse was also found. Across analyses there was a significant main effect of
female gender on risk of early sexual abuse, which has been a consistent finding
in the literature. These results suggest an association between early traumatic
experiences and particular symptom clusters of depression, mania, or both in
adults.
Anxiety Disorders
Just about all of the anxiety disorders, with the possible exception of panic
disorder, also occur in ED patients (23–75%) at higher than expected rates
when compared to rates in the general population. These data have been much
more extensively reviewed in Chapter 8 and will not be discussed at length.
However, the general ranges reported are as follows: PTSD (37–62%),
obsessive-compulsive disorder (OCD) (3–66%), social phobia (16–55%), panic
disorder (5–10%), generalized anxiety disorder (GND) (12%), simple phobia
(13%). Taken together, it is evident from this research that one of the most
common anxiety disorders presenting in the ED population is PTSD, a finding
from the NWS (14). Although there are many studies of the rates of child abuse
and other victimization experiences in ED samples (13,14,16,17,56), a paucity
of reliable data exists on the prevalence of PTSD in subjects with EDs.
An important and consistent finding is that anxiety disorders are usually
primary and the ED is usually secondary. Much has been published on this
particular comorbid dyad in the last few years, as noted in Chapters 6 and 8. It
appears that anxiety symptoms, particularly those involving behavioral
inhibition and obsessionality, are important risk factors for the development of
EDs. Anxiety typically persists and often worsens following the onset of an ED.
In the NWS, sexual assault and PTSD symptoms began before the first binge
ever occurred in 84% of all rapes and nearly 100% of rapes occurring during
childhood (≤17 years old) (in subjects with BN only).
A diagnostic sine qua non for PTSD is of course the occurrence of a traumatic
event. However, all of the other anxiety disorders have been reported to have
high rates of victimization, including OCD, panic disorder, GAD, social phobia,
simple phobia, as well as separation anxiety disorder and overanxious disorder
in kids (29,51–53,57–62).
Substance Use Disorders

Affective and anxiety disorders are seen at higher frequencies in all of the EDs,
but when it comes to SUDs there is a striking difference between AN and BN,
with BED falling somewhere in the middle. Rates in the EDs range from 0% to
37%, with restricting AN having less than expected rates of SUDs and BN
patients having much greater rates than expected (see Chapter 8). Some form of
substance abuse may be a compensatory strategy inherent to BN, e.g., abuse of
laxatives, diuretics, CNS stimulants (ranging from cocaine to ephedra-
containing over-the-counter preparations) (63). Some BN patients use alcohol,
opiates, and/or nicotine as appetite suppressants and anabolic steroids to build
muscle. Alcohol and other CNS depressants may also serve to depress inhibitory
influences so as to facilitate vomiting. The use of marijuana and other
psychedelics may complicate the clinical picture, causing further body image
distortion, anxiety, and mood alterations (64). SUDs have been linked not only
to bulimic disorders but to previous traumas as well (49,65,66). Dansky and
520 BREWERTON
colleagues demonstrated in the NWS that the major reason alcohol abuse so
commonly coexists with BN is because they share comorbid PTSD (66).
Dissociative Disorders
Next to PTSD, which requires an identified traumatic event, the DDs as a group
are most closely linked to trauma. This link between trauma and dissociative
phenomena has been well described in the literature (4,67,68). DDs and their
symptoms have been described all over the world and are commonly related to
severely overwhelming traumatic experiences, particularly when occurring
during childhood, and include derealization, depersonalization, time
distortions, cognitive and memory alterations (including amnesia), identity
alterations, and somatic sensations. Dissociative disorders (particularly
dissociative identity disorder, DID) are very real disorders of memory,
consciousness, and identity that represent the devastating effects of unusually
severe and often chronic violent abuse during childhood. Axis I comorbidity is
quite common, and several investigators have reported high frequencies of EDs
and behaviors in DID patients (69–71). Conversely, high frequencies of
dissociative symptoms have been reported in ED patients, particularly BN
(72–85), and in nonclinical samples with ED pathology (86,87). Previously,
there were no detailed studies of both dissociative symptoms, such as psychogenic
amnesia (PA), and victimization experiences, such as rape, in a representative
group of women with and without EDs. Data generated as part of the NWS
provided an opportunity for a more controlled examination of these
relationships.
As part of the PTSD screening, respondents were designated as having PA if
they endorsed ever “forgetting” all or part of a significant traumatic event, a
question that was part of the PTSD module. Respondents with BN endorsed PA
2.5 times more often than non-BN/non-BED respondents (27% v. 11%,
p < 0.000033, χ2square), and there was a trend for a similar difference between
BN and BED (11%, p < 0.09). These results imply that purging, rather than
bingeing per se, is more closely associated with an endorsement of PA. It is
likely that bulimic behaviors are maladaptive mechanisms with psychobiological
underpinnings linked to PTSD and depression that facilitate avoiding, numbing
and forgetting traumatic memories (15).
Somatoform Disorders
Disturbances in body perception and overt body distortion are somatoform
symptoms inherent to the EDs, especially AN. Links between somatoform
disorders (SDs) and EDs have also been reported in the literature (88–91). A
recent family study showed clustering of OCD with EDs and SDs (89),
particularly body dysmorphic disorder (BDD). BDD and EDs are similarly
characterized by obsessive and compulsive phenomenology targeted on the body
despicable. Other investigators have noted the connection between EDs and
BDD (90,91).
Links between SDs and prior victimization, especially CSA, are well
known (92–97). Among adult females, child abuse contributes not only to
general somatic preoccupation but to specific somatic symptoms in the chest and
throat areas as well (97). Interestingly, these body areas are those that are
involved in vomiting behavior. SDs are intimately linked to DDs as well as to
other axis I comorbidity. Somatoform dissociation is a unique construct that is
highly characteristic of DD patients, a core feature in many patients with SDs,
and an important symptom cluster in a subgroup of patients with EDs (88). Other
studies report that measures of somatization are significantly positively
correlated with measures of dissociation.
Impulse Control Disorders

The ICDs, such as trichotillomania, kleptomania, compulsive buying, as
well as self-mutilation, have been associated with the EDs, particularly
BN (45,98–104) and to some extent BED (94). As noted in Chapter 9, impulsivity
is a common comorbid personality trait in the EDs, particularly BN. Patients
with ICD and other impulsive behaviors have high rates of traumatic
histories (105,106).
ICDs have been hypothesized to exist on an obsessive-compulsive spectrum
(107–109). In a study designed to characterize impulsive versus compulsive self-
injurious behavior in patients with EDs, CSA and anxiety were found to
significantly predict impulsive self-injury, whereas obsessionality and age
predicted compulsive self-injury (102). Wonderlich and colleagues tested a
number of mediational pathways between CSA and subsequent eating
disturbance and found that behavioral impulsivity was the most significant
mediating factor (44).
From a clinical and preventive perspective, it is important to note that
individuals who have been sexually abused as children are at a higher risk for
being revictimized as adults, possibly in part due to the increased rates of
impulsivity that follow abuse (110,111) and perhaps due to repetition
compulsion. Revictimization serves to amplify the impact of childhood sexual
trauma on selective attention to trauma-related stimuli, contributing to
increased defenses and hence psychopathology, as well as more self-blame and
refractoriness to treatment (112,113).
522 BREWERTON
Attention Deficit-Hyperactivity Disorder and

Disruptive Behavior Disorders
Attention deficit-hyperactivity disorder (ADHD) and the EDs do co-occur with
each other although not at an increased rate. Stimulants not only can be used in
patients with BN or BED to manage the ADHD, they may also be helpful for the
binge eating as well (114–116). However, a major danger in this area is
misdiagnosis. Common comorbid disorders that can be associated with
alterations in attention, concentration, and/or arousal, such as MDD, PTSD,
and DDs, can be misdiagnosed as ADHD, particularly in children who were
abused at an early age. Dykman has demonstrated higher rates of externalizing
symptoms in sexually and physically abused children (62), and Gold and Teicher
reported higher activity levels in abused prepubertal children as measured by
computerized activity monitors (117). Nevertheless, the combined type of
ADHD has been found to be associated with child abuse in girls meeting ADHD
criteria compared to controls (118). Physical abuse in particular has been found
to drastically increase the diagnosis of ADHD (119). In contrast, ADHD, and
perhaps DBDs, may also increase the risk of being physically abused by
caretakers, probably as a result of increased activity and impulsivity (120).
In terms of DBDs, Thompson and colleagues reported higher rates of
aggressive behavior in adolescent girls with binge eating, purging, and
dietary restriction, and opine that this is a neglected area in the field (121).
Geist and coworkers reported that oppositional defiant disorder (ODD) occurs
more commonly in adolescents with EDs associated with binge-purge
symptoms (122). ODD has been reported to be associated with a history of
physical or sexual maltreatment as well as PTSD symptoms (123,124). DBDs
are often precursors to the development of antisocial personality disorder,
which has also been linked to child abuse and neglect as discussed below.
Axis II Comorbidity
The most common personality disorders (PDs) associated with the EDs have
been extensively reviewed in Chapter 9. In short, cluster C disorders [obsessive-
compulsive personality disorder (OCPD), avoidant personality disorder (APD),
and dependent personality disorder (DPD)] are common to both AN and BN. In
particular, OCPD and its traits are the best studied of this cluster and have been
shown to be important risk factors in the development of EDs (125).
Although there are no clear links between cluster C disorders and
victimization or PTSD, childhood emotional neglect has been associated with an
increased risk for APD during adolescence and early adulthood (126). It is
important to note that the presence of APD may contribute more avoidance to
that already inherent to PTSD and therefore make it even harder for the
traumatized ED patient to disclose abuse and face the exposure work required.
Childhood emotional neglect has also been associated with increased risk for
paranoid and cluster A PD symptom levels during adolescence and early
adulthood, and childhood physical neglect has been associated with an increased
risk for schizotypal PD and with cluster A PD symptom levels (126).
It is also noted in Chapter 9 that cluster B disorders, and their associated trait
of impulsivity, decidedly occur more frequently in the BN and AN bingepurge
type than in the AN restricting type of ED. Data show strong links not only
between BPD and patients with binge-purge symptomatology, but also between
BPD and histories of victimization, PTSD, affective disorders, other anxiety
disorders, SUDs, DDs, SDs, ICDs, DBDs, and ADHD (127–130). BPD is also
highly associated with multiple comorbid axis I disorders (131–133) as well as
with prior abuse and neglect. Other cluster B personality disorders, such as
narcissistic personality disorder (NPD) and antisocial personality disorder
(APD), have also been associated with child abuse and neglect(134).
Course and Outcome

The effects of trauma history on the clinical course and treatment outcome of
EDs and their comorbidity is unknown. Few or no data exist on this important
topic because it has not been a commonly identified variable in treatment
studies. However, clinical experience suggests that trauma history and
especially subsequent PTSD history carries a worse prognosis. PTSD itself is a
chronic disorder with one-third of cases followed up 10 years later still
meeting full criteria (1). Bell examined the question of whether concurrent
psychopathology at presentation influences response to treatment for BN and
found no consistent relationship between outcome and any axis I
disorder (135). However, Bell also noted that most of the studies assessing axis
II dysfunction show that borderline symptom severity or cluster B personality
disorder, which is linked to prior victimization, can impair outcome. Similarly,
multi-impulsivity, which is also linked to abusive histories, is associated with
poor long-term prognosis following treatment (136,137). Anderson reported
that abused BN patients treated in an inpatient setting were more likely to be
rehospitalized in the 3-month postdischarge period than nonabused BN patients,
which lends support to the notion that trauma may retard the recovery process
from an ED (138).
524 BREWERTON
PRINCIPLES OF TREATMENT FOR THE

COMORBID ED PATIENT WITH VICTIMIZATION
How does the clinician approach treatment of the ED patient with significant
trauma-related psychiatric comorbidity? What should the clinician do first
besides be respectful of the patient’s struggle? What are the general principles in
the management of comorbidity with EDs? As was noted by Brown, “there is a
dearth of treatment suggestions to deal with the comorbid problems of the
patient with an ED and a history of abuse” (56). The general principles in the
management of comorbidity with EDs are outlined in Table 1, but will be
elaborated here. The most tried-and-true initial approach for any patient with
or without complex symptomatology is to perform a complete psychiatric
evaluation, which substantially reduces the chances of misdiagnosis. Such a
comprehensive evaluation must include histories of all forms of abuse and
neglect, major life events or stressors, as well as natural disasters and witnessing
of violence. Many patients are not comfortable revealing major traumatic events
during an initial interview, so evaluation must be seen as an ongoing process
that blends into the treatment process and continues to shape it. Any number of
screening and diagnostic instruments may be used in the course of a complete
evaluation, and this is extensively discussed in Chapter 2 as well as in available
practice guidelines, which will be discussed below.
Of course, the formation and maintenance of a therapeutic alliance is not
only the starting point for but also the subsequent foundation of future work
and progress. Comorbidity has been reported to be a significant factor related to
ED treatment dropout (139). All involved staff and clinicians, especially the
primary therapist and/or psychiatrist, must work extra hard to convey a sense of
trustworthiness, honesty, straightforwardness, knowledge, and nonjudgmental
positive regard, which are essential ingredients for any successful course of
therapy, but especially for comorbid ED patients with trauma histories and
related disorders. In addition, it is important that countertransference feelings be
monitored closely and kept in check. In my experience, sometimes traumatized
patients do not even realize that they have been victimized until someone
explains to them the detailed definitions of the abuse in question, e.g., rape,
assault, aggravated assault, bullying, etc. This is especially true for patients who
were abused during early childhood. Only later, after a significant degree of
cognitive development has occurred, do some victims of CSA realize that they
were victims. In this sense the therapist serves as someone who offers a more
objective perspective and who assists the traumatized patient in reframing her
experiences in the context of consensual reality. All too often denial, avoidance,
self-blame, and shame cloud the view of the meanings of prior traumatic events.
It must also be remembered that ED patients are as a group highly suggestible or
TABLE 1 General Principles in the Management of Comorbidity with Eating Disorders

526 BREWERTON
hypnotizable (140). The clinician treating traumatized ED patients must take

every precaution to elicit his or her patients’ histories without being suggestive
that these events did or did not occur. Whenever traumatic events that occurred
during childhood or adolescence are reported, the clinician should be sure to
understand and obey all reporting laws mandated by the local state and/or
country government.
As a complete list of disorders is being determined it is essential for
comprehensive treatment planning that clinicians determine the chronological
order in which the disorders first occurred or appeared. Which disorder came
first in relation to the other identified comorbid disorders? What disorders are
concurrent? What are the traumas and when did they happen in relation to the
disorders identified? It is useful over time to work with the patient to construct
a phenomenological, cause-and-effect-based time line and schema in which the
development and evolution of their problems makes overall sense to them. It is
likely that this information will evolve gradually and be referred to in therapy
again and again. Identifying the functional links of meaning between comorbid
disorders and life events can be very informative and instructive for patient,
family, and therapist in terms of identifying and challenging cognitive
distortions. One very common theme underlying the link between EDs and
trauma-related comorbidity is that maladaptive coping mechanisms, such as
bingeing, purging, dieting, substance abuse, dissociation, self-mutilation,
compulsive behavior, etc., are ways to avoid trauma-related states, i.e.,
memories, affects, behaviors, and thoughts, which are often triggered by any
number of cues. This phenomenon is another type of “compensatory behavior”
and is essentially another variation of the selfmedication hypothesis (141). Self-
punishment also has a lot of cognitive explanatory power for many traumatized
ED patients.
What disorder is primary and what disorder is secondary, tertiary, etc.? If the
onset of the ED came before the onset of the MDD, then the ED is the primary
disorder and MDD is the secondary disorder, which may in part be due to the
ED. If the ED came after the MDD, then the ED is secondary and may in part
be due to the MDD. In other words, as discussed in Chapter 6, primary
disorders may be risk factors for the development of secondary disorders. There
has been a long tradition in medicine to ascertain this type of information, to
develop a chronology of events and their subsequent sequelae, as a way to guide
treatment and “go to the source” of a patient’s problems. A secondary
depression may be more likely to respond to weight restoration and successful
management of the ED, whereas a primary depression may more likely linger
on or be an impediment to the ED recovery process unless it is managed more
aggressively. Interestingly, research shows that most episodes of MDD are
secondary, and they tend to be more persistent and severe than primary MDD
episodes (142). Similar to the above example, to the extent that an ED is
secondary to victimization and the posttraumatic stress process, it is important
to eventually and adequately address this primary problem.
Whenever possible, comorbid disorders should be managed concurrently, but
again, it is a matter of emphasis and prioritization. Which disorder or problem
should the clinician focus on first? A useful guideline that should determine the
clinician’s initial emphasis, for both ethical and medicolegal reasons, is the
current level of danger, risk, and/or brain/body impairment. It is necessary to
normalize or stabilize brain function in order for the psychotherapeutic
528 BREWERTON
enterprise to “take hold” and succeed. The idea that the nutritional instability
common to all ED patients must be addressed before more intensive
psychotherapy can be effectively utilized is common knowledge among experts
in the field. Garner and Garfinkel noted that the typical stages or phases of
treatment start with refeeding, nutritional stabilization, and rehabilitation,
which sometimes require a structured setting and program (143). Walter
Vandereycken put it succinctly when he said, “First we eat, then we talk” (144).
In short, all patients should receive nutritional education and counseling as
reviewed in Chapter 16, and these messages must be in sync with the primary
psychotherapeutic process.
The body of knowledge that must be conveyed includes several key points,
including (a) the powerful semistarvation effects in conscientious objectors
reported by Keys (8) and discussed in Chapter 8, and which lead to the
conclusion that dieting induces mental and physical symptoms de novo, and
(b) the implications of the tryptophan depletion studies to EDs as noted
in Chapters 11 and 16, which basically indicate that dieting depletes
neurotransmitters and this in turn has significant and rapid clinical effects (145).
It is also important to explain to patients and families that refeeding alone may
alleviate mood and anxiety dysregulation to some degree, but that it is also not
going to “fix everything.” In some ways certain symptoms of the traumatized ED
comorbid patient may get worse before they get better, and this is useful to
predict. Refeeding may allow patients to start “feeling their feelings” again and
to therefore begin to process them, usually with increasing intensity. Their
eating-disordered brains slowly but surely start to function better and better as
neurotransmitter and neurohormone levels move toward normalization (see
Chapters 11, 12, and 13). For this reason, nutritional rehabilitation, especially
weight gain in AN, not only allows antidepressants to work but also allows
psychotherapy to work. In effect, once patients begin to think and feel more clearly
as a result of refeeding, the real psychological issues can “rise” to awareness and
become clarified, processed, and “digested.” Nevertheless, the alexithymia that
is characteristic of ED patients and other psychosomatic conditions may be an
important clinical issue to address (146).
When substance abuse complicates the picture, one should always manage
withdrawal and dependence aggressively and as much as possible achieve
abstinence (“Say no, then go”). Just as starved brains can’t learn very well,
intoxicated brains can’t either. Basic requirements for effective psychotherapy
include grossly intact brain function, including the ability to attend and the
ability to learn-unlearn. In addition, sufficient motivation, the willingness to
change, and the presence of supportive relationship(s) are important ingredients
for progress. If necessary, motivational enhancement therapy (MET) may be
beneficial in highly resistant patients (147).
Once brain function has been stabilized and the process of normalization is
well on its way, more intensive psychotherapy can begin or intensify. The
phases of nutritional rehabilitation and intensive psychotherapy are not
absolutely distinct, but instead tend to overlap and merge into one another. The
major forms of psychotherapy that have been shown to have empirical evidence
for their effectiveness in the EDs are cognitive-behavioral therapy (CBT),
interpersonal therapy (IPT), family therapy (FT), and dialectical behavior
therapy (DBT), all of which are reviewed in Chapters 17, 18, 19, and 20,
respectively. Experiential psychotherapies have not been rigorously studied, but
may have important roles in the recovery of comorbid ED patients, particularly
in dealing with body image concerns, accessing emotional states, and
being exposed to one’s own self, so to speak. Likewise, elements of
psychodynamically oriented psychotherapy may be useful adjuncts in treatment,
but this should never be the only or primary modality used.
It is imperative that the clinician who wants to effectively manage EDs and
their comorbid psychiatric disorders familiarize themselves not only with the
American Psychiatric Association’s Practice Guidelines for the Treatment of
Eating Disorders (www.psych.org/clin_res/guide.bk.cfm), but with other
practice guidelines and treatment manuals as well on an as-needed
basis (www.psychguides.com; www.issd.org; www.psych.org) (148,149).
Therapeutic techniques adequate for the non-comorbid ED are not likely to
address other problems and disorders, so the clinician must rely on his or her
own resources, creativity, and ability to integrate a multitude of approaches in
order to tailor the best treatment to match the uniqueness of each patient.
Specific aspects and approaches regarding specific comorbid combinations are
outlined in Table 2.
Lest it seem too overwhelming or daunting, it is important to realize that
there are a number of commonalities inherent to the treatment of ED patients with
comorbid trauma-related disorders, including psychoeducation, CBT, and
pharmacotherapy. When using general relaxation techniques it is, in my
opinion, better to avoid progressive muscle relaxation in ED patients. This
inadvertently focuses attention on specific body parts that are “emotionally
loaded” for ED patients, which often serves to heighten anxiety rather than
lower it. Techniques that focus on the breath with or without the use of
visualization appear to work best in my view. When anxiety disorders
enter the clinical picture, the addition of prolonged exposure to CBT is clearly
indicated (149–154). Adequate time needs to be spent educating the patient
about the rationale for in vivo exposure using lay explanations for extinction,
conditioned cues, experiential learning, and why anxiety is a necessary
component for exposure to work. Hence, medicating away anxiety with
sddfdfgfgfjn
530 BREWERTON
TABLE 2 Treatment Caveats for Specific Comorbid Combinations

532 BREWERTON
benzodiazepines should be avoided and can be countertherapeutic to the

management of PTSD. In explaining the rationale for prolonged exposure, most
patients can understand the concepts of classical conditioning, or perhaps the
notion of finishing “unfinished business,” decreasing the “power” of the
memories, or “tying up loose ends.” They can usually embrace the goal of being
able to remember traumatic events and even talk about them without feeling
the intense negative feelings that have become so familiar, and which have
driven them in part to ED “solutions.” In addition, eye movement
desensitization and reprocessing is an effective strategy for PTSD and may be a
useful adjunct to trauma work (www.emdr.org) (155). In my experience, it
complements CBT and exposure work quite well.
IPT techniques can be added for comorbid patients who do not respond to
CBT or who reach an impasse in treatment (see Chapter 19). DBT may be useful
for a variety of self-destructive behaviors, particularly in the patient with
comorbid ICD, BPD, and/or multi-impulsivity (see Chapter 20). Appropriate
family therapy consisting of specific goals jointly identified by the patient and
the therapist should be carefully considered for all patients (see Chapter 18).
Psychopharmacological interventions tend to be the rule rather than the
exception when dealing with comorbid problems associated with the EDs,
particularly comorbid affective and anxiety disorders, and this is generally
supported by most practice guidelines; drug treatment is usually ineffective
when used alone in ED patients, especially in those with trauma-related
comorbidity. The typical psychopharmacological agents used in the
management of EDs are extensively reviewed in Chapter 21, but specific
aspects related to how these common agents are used to manage comorbid
disorders are briefly reviewed in Table 3.
Overall, these principles and recommended guidelines might serve as a
useful blueprint for building a more specific treatment plan geared to the needs
of the individual patient and her particular display of comorbidity within a
developmental and biopsychosocial context.
534 BREWERTON
TABLE 3 Specific Agents in the Management of EDs with Comorbidity

536 BREWERTON
REFERENCES
1. Kessler RC, Sonnega A, Bromet E, Hughes M, Nelson CB. Posttraumatic stress
disorder in the National Comorbidity Survey. Arch Gen Psychiatry 1995; 52:
1048–1060.
2. Kessler RC, Davis CG, Kendler KS. Childhood adversity and adult psychiatric
disorder in the US National Comorbidity Survey. Psychol Med 1997; 27:
1101–1119.
3. Kessler RC. Posttraumatic stress disorder: the burden to the individual and to
society. J Clin Psychiatry 2000; 61(Suppl 5):4–12; discussion 13–14.
4. Bremner JD. Does Stress Damage the Brain? Understanding Trauma-Related

Disorders from a Neurological Perspective. New York: W.W. Norton &
Company, 2002.
5. Felitti VJ, Anda RF, Nordenberg D, Williamson DF, Spitz AM, Edwards V, Koss
MP, Marks JS. Relationship of childhood abuse and household dysfunction to many
of the leading causes of death in adults. The Adverse Childhood Experiences (ACE)
Study. Am J Prev Med 1998; 14:245–258.
6. Kendall-Tackett K. The health effects of childhood abuse: four pathways by which
abuse can influence health. Child Abuse Neglect 2002; 26:715–729.
7. Kaufman J, Charney D. Effects of early stress on brain structure and function:
implications for understanding the relationship between child maltreatment and
depression. Develop Psychopathol 2001; 13:451–471.
9. Kessler RC, McGonagle KA, Zhao S, Nelson CB, Hughes M, Eshleman S,
Wittchen HU, Kendler KS. Lifetime and 12-month prevalence of DSM-III-R
psychiatric disorders in the United States. Results from the National Comorbidity
Survey. Arch Gen Psychiatry 1994; 51:8–19.
10. Resnick HS, Kilpatrick DG, Dansky BS, Saunders BE, Best CL. Prevalence of
civilian trauma and posttraumatic stress disorder in a representative national sample
of women. J Consult Clin Psychol 1993; 61:984–991.
11. Duncan RD, Saunders BE, Kilpatrick DG, Hanson RF, Resnick HS. Childhood
physical assault as a risk factor for PTSD, depression, and substance abuse: findings
from a national survey. Am J Orthopsychiatry 1996; 66:437–448.
12. Kendler K, Bulik C. Childhood sexual abuse and adult psychiatric and substance
abuse disorders in women. Arch Gen Psychiatry 2000; 57:953–959.
13. Wonderlich SA, Brewerton TD, Jocic Z, Dansky BS, Abbott DW. The relationship
of childhood sexual abuse and eating disorders: a review. J Am Acad Child Adol
Psychiatry 1997; 36:1107–1115.
14. Dansky BS, Brewerton TD, O’Neil PM, Kilpatrick DG. The National Women’s
Study: relationship of crime victimization and PTSD to bulimia nervosa. Int J Eat
Disord 1997; 21:213–228.
15. Brewerton TD, Dansky BS, Kilpatrick DG, O’Neil PM. Bulimia nervosa, PTSD
and “forgetting”: results from the National Women’s Study. In: Williams LM,
Banyard VL, eds. Trauma and Memory. Durham: Sage Publications, 1999:
127–138.
16. Molinari E. Eating disorders and sexual abuse. Eating and Weight Disorders: EWD
2001; 6:68–80.
17. Smolak L, Murnen SK. A meta-analytic examination of the relationship between
child sexual abuse and eating disorders. Int J Eat Disord 2002; 31:136–150.
18. Kenardy J, Ball K. Disordered eating, weight dissatisfaction and dieting in relation
to unwanted childhood sexual experiences in a community sample. J Psychosom
Res 1998; 44:327–337.
538 BREWERTON
19. Romans SE, Gendall KA, Martin JL, Mullen PE. Child sexual abuse and later
disordered eating: a New Zealand epidemiological study. Int J Eat Disord 2001; 29:
380–392.
20. Gleaves DH, Eberenz KP, May MC. Scope and significance of posttraumatic
symptomatology among women hospitalized for an eating disorder. Int J Eat Disord
1998; 24:147–156.
21. Matsunaga H, Kaye WH, McConaha C, Plotnicov K, Pollice C, Rao R, Stein D.
Psychopathological characteristics of recovered bulimics who have a history of
physical or sexual abuse. J Nerv Ment Dis 1999; 187:472–477.
22. Deep AL, Lilenfeld LR, Plotnicov KH, Pollice C, Kaye WH. Sexual abuse in eating
disorder subtypes and control women: the role of comorbid substance dependence
in bulimia nervosa. Int J Eat Disord 1999; 25:1–10.
23. Brown L, Russell J, Thornton C, Dunn S. Experiences of physical and sexual abuse
in Australian general practice attenders and an eating disordered population. Aust N
Z J Psychiatry 1997; 31:398–404.
24. Brown L, Russell J, Thornton C, Dunn S. Dissociation, abuse and the eating
disorders: evidence from an Australian population. Aust N Z J Psychiatry 1999; 33:
521–528.
dissociation and impulsive behaviours. Psychol Med 1999; 29: 935–942.
26. Nagata T, Kawarada Y, Kiriike N, Iketani T. Multi-impulsivity of Japanese patients
with eating disorders: primary and secondary impulsivity. Psychiatry Res 2000; 94:
239–250.
27. Favaro A, Dalle Grave R, Santonastaso P. Impact of a history of physical and sexual
abuse in eating disordered and asymptomatic subjects. Acta Psychiat Scand 1998;
97:358–363.
28. Striegel-Moore RH, Garvin V, Dohm FA, Rosenheck RA. Eating disorders in a
national sample of hospitalized female and male veterans: detection rates and
psychiatric comorbidity. Int J Eat Disord 1999; 25:405–414.
29. Lipschitz DS, Winegar RK, Hartnick E, Foote B, Southwick SM. Posttraumatic
stress disorder in hospitalized adolescents: psychiatric comorbidity and clinical
correlates. J Am Acad Child Adol Psychiatry 1999; 38:385–392.
30. Neumark-Sztainer D, Story M, Hannan PJ, Beuhring T, Resnick MD. Disordered
eating among adolescents: associations with sexual/physical abuse and other familial/
psychosocial factors. Int J Eat Disord 2000; 28:249–258.
31. Wonderlich SA, Crosby RD, Mitchell JE, Roberts JA, Haseltine B, DeMuth G,
Thompson KM. Relationship of childhood sexual abuse and eating disturbance in
children. J Am Acad Child Adol Psychiatry 2000; 39:1277–1283.
32. Thompson KM, Wonderlich SA, Crosby RD, Mitchell JE. Sexual victimization and
adolescent weight regulation practices: a test across three community based
samples. Child Abuse Neglect 2001; 25:291–305.
33. Ackard DM, Neumark-Sztainer D, Hannan PJ, French S, Story M. Binge and purge
behavior among adolescents: associations with sexual and physical abuse in a
nationally representative sample: the Commonwealth Fund survey. Child Abuse

Neglect 2001; 25:771–785.
34. Edgardh K, Ormstad K. Prevalence and characteristics of sexual abuse in a national
sample of Swedish seventeen-year-old boys and girls. Acta Paediatrica 2000; 89:
310–319.
35. Swanston HY, Tebbutt JS, O’Toole BI, Oates RK. Sexually abused children 5 years
after presentation: a case-control study. Pediatrics 1997; 100:600–608.
36. Grilo CM, Masheb RM. Childhood psychological, physical, and sexual
maltreatment in outpatients with binge eating disorder: frequency and associations
with gender, obesity, and eating-related psychopathology. Obesity Res 2001; 9:
320–325.
37. Striegel-Moore RH, Dohm FA, Pike KM, Wilfley DE, Fairburn CG. Abuse,
bullying, and discrimination as risk factors for binge eating disorder. Am J
Psychiatry 2002; 159:1902–1907.
38. Johnson JG, Cohen P, Kasen S, Brook JS. Childhood adversities associated with
risk for eating disorders or weight problems during adolescence or early adulthood.
Am J Psychiatry 2002; 159:394–400.
39. Kent A, Waller G, Dagnan D. A greater role of emotional than physical or sexual
abuse in predicting disordered eating attitudes: the role of mediating variables. Int J
Eat Disord 1999; 25:159–167.
40. Kent A, Waller G. Childhood emotional abuse and eating psychopathology. Clin
Psychol Rev 2000; 20:887–903.
41. Kinzl JF, Mangweth B, Traweger CM, Biebl W. Eating-disordered behavior
in males: the impact of adverse childhood experiences. Int J Eat Dis 1997; 22:
131–138.
42. Favaro A, Rodella FC, Santonastaso P. Binge eating and eating attitudes among Nazi
concentration camp survivors. Psychol Med 2000; 30:463–466.
43. Andrews B. Bodily shame in relation to abuse in childhood and bulimia: a
preliminary investigation. Br J Clin Psychol 1997; 36(Pt 1):41–49.
44. Wonderlich S, Crosby R, Mitchell J, Thompson K, Redlin J, Demuth G, Smyth J.
Pathways mediating sexual abuse and eating disturbance in children. Int J Eat
Disord 2001; 29:270–279.
45. Wonderlich SA, Crosby RD, Mitchell JE, Thompson KM, Redlin J, Demuth G,
Smyth J, Haseltine B. Eating disturbance and sexual trauma in childhood and
adulthood. Int J Eat Disord 2001; 30:401–412.
46. Murray C, Waller G. Reported sexual abuse and bulimic psychopathology among
nonclinical women: the mediating role of shame. Int J Eat Disord 2002; 32:
186–191.
47. Waller G, Meyer C, Ohanian V, Elliott P, Dickson C, Sellings J. The
psychopathology of bulimic women who report childhood sexual abuse: the
mediating role of core beliefs. J Nerv Ment Dis 2001; 189:700–708.
48. Waller G. Perceived control in eating disorders: relationship with reported sexual
abuse. Int J Eat Disord 1998; 23:213–216.
540 BREWERTON
49. Brady KT, Killeen TK, Brewerton T, Lucerini S. Comorbidity of psychiatric

disorders and posttraumatic stress disorder. J Clin Psychiatry 2000; 61(suppl 7):
22–32.
50. Kaplan MJ, Klinetob NA. Childhood emotional trauma and chronic posttraumatic
stress disorder in adult outpatients with treatment-resistant depression. J Nerv
Ment Dis 2000; 188:596–601.
51. Heim C, Nemeroff CB. The role of childhood trauma in the neurobiology of mood
and anxiety disorders: preclinical and clinical studies. Biol Psychiatry 2001; 49:
1023–1039.
52. Penza KM, Heim C, Nemeroff CB. Neurobiological effects of childhood abuse:
implications for the pathophysiology of depression and anxiety. Arch Women’s
Ment Health 2003; 6:15–22.
53. Hexel M, Sonneck G. Somatoform symptoms, anxiety, and depression in the
context of traumatic life experiences by comparing participants with and without
psychiatric diagnoses. Psychopathology 2002; 35:303–312.
54. Gladstone G, Parker G, Wilhelm K, Mitchell P, Austin MP. Characteristics of
depressed patients who report childhood sexual abuse. Am J Psychiatry 1999; 156:
431–437.
55. Levitan RD, Parikh SV, Lesage AD, Hegadoren KM, Adams M, Kennedy SH,
Goering PN. Major depression in individuals with a history of childhood physical or
sexual abuse: relationship to neurovegetative features, mania, and gender. Am J
Psychiatry 1998; 155:1746–1752.
56. Brown L. Child physical and sexual abuse and eating disorders: a review of the links
and personal comments on the treatment process. Aust N Z J Psychiatry 1997; 31:
194–199.
57. Safren SA, Gershuny BS, Marzol P, Otto MW, Pollack MH. History of childhood
abuse in panic disorder, social phobia, and generalized anxiety disorder. J Nerv
Ment Dis 2002; 190:453–456.
58. Flisher AJ, Kramer RA, Hoven CW, Greenwald S, Alegria M, Bird HR, Canino G,
Connell R, Moore RE. Psychosocial characteristics of physically abused children
and adolescents. J Am Acad Child Adol Psychiatry 1997; 36:123–131.
59. Stein MB, Walker JR, Anderson G, Hazen AL, Ross CA, Eldridge G, Forde DR.
Childhood physical and sexual abuse in patients with anxiety disorders and in a
community sample. Am J Psychiatry 1996; 153:275–277.
60. Engel CC Jr, Walker EA, Katon WJ. Factors related to dissociation among patients
with gastrointestinal complaints. J Psychosom Res 1996; 40:643–653.
61. Trowell J, Ugarte B, Kolvin I, Berelowitz M, Sadowski H, Le Couteur A.
Behavioural psychopathology of child sexual abuse in schoolgirls referred to a
tertiary centre: a North London study. Europ Child Adol Psychiatry 1999; 8:
107–116.
62. Dykman RA, McPherson B, Ackerman PT, Newton JE, Mooney DM, Wherry J,
Chaffin M. Internalizing and externalizing characteristics of sexually and/or
physically abused children. Integr Physiol Behav Sci 1997; 32:62–74.
63. Cochrane CE, Malcolm R, Brewerton TD. Eating disorders in cocaine abusers.
AddictiveDisl998;23:l-7.
64. Gross H, Ebert MH, Faden VB, Goldberg SC, Kaye WH, Caine ED, Hawks R,
Zinberg N. A double-blind trial of delta 9-tetrahydrocannabinol in primary
anorexia nervosa. J Clin Psychopharmacol 1983; 3:165–171.
65. Kessler RC, Crum RM, Warner LA, Nelson CB, Schulenberg J, Anthony JC.
Lifetime co-occurrence of DSM-III-R alcohol abuse and dependence with other
psychiatric disorders in the National Comorbidity Survey. Arch Gen Psychiatry
1997; 54:313–321.
66. Dansky BS, Brewerton TD, Kilpatrick DG. Comorbidity of bulimia nervosa and
alcohol use disorders: results from the National Women’s Study. Int J Eat Disord
2000; 27:180–190.
67. Putnam F, Guroff JJ, Silberman EK, Barban L, Post RM. The clinical
phenomenology of multiple personality disorder: a review of 100 cases. J Clin
Psychiatry 1986; 47:285–293.
68. Putnam F. Dissociation in Children and Adolescents: A Developmental Perspective.
69. Torem MS. Dissociative states presenting as an eating disorder. Am J Clin
Hypnosis 1986; 29:137–142.
70. Torem MS. Covert multiple personality underlying eating disorders. Am J
Psychotherapy 1990; 44:357–368.
71. Torem MS. Eating disorders in patients with multiple personality disorder. In:
Kluft RP, Fine CG, eds. Clinical Perspectives on Multiple Personality Disorder.
Washington, DC: American Psychiatric Press, 1993:343–353.
72. Demitrack MA, Putnam FW, Brewerton TD, Brandt HA, Gold PW. Dis sociative
phenomena in eating disorders: relationship to clinical variables. J Psychaitry 1990;
147:1184–1188.
73. Goodwin JM, Attias R. Eating disorders in survivors of multimodal childhood
abuse. In: Kluft RP, Fine CG, eds. Clinical Perspectives on Multiple Personality
Disorder. Washington, D.C.: American Psychiatric Press, 1993:327–341.
74. Abraham SF, Beaumont PJV. How patients describe bulimia or binge eating.
Psychol Med 1984; 12:625–635.
75. Everill J, Waller G, Macdonald W. Dissociation in bulimic and non-
eatingdisordered women. Int J Eat Disord 1995; 17:127–134.
76. Tobin DL, Molteni AL, Elin MR. Early trauma, dissociation, and late onset in the
eating disorders. Int J Eat Disord 1995; 17:305–308.
77. Valdiserri S, Kihlstrom JF. Abnormal eating and dissociative experiences. Int J Eat
Disord 1995; 17:373–380.
78. Gleaves DH, Eberenz KP. Correlates of dissociative symptoms among women with
eating disorders. J Psychiatr Res 1995; 29:417–4426.
79. Vanderlinden J, Spinhoven P, Vandereycken W, van Dyck R. Dissociative and
hypnotic experiences in eating disorder patients: an exploratory study. Am J Clin
Hypnosis 1995; 38:97–108.
542 BREWERTON

dissociation and impulsive behaviours. Psychol Med 1999; 29:935–942.
81. Farrington A, Waller G, Neiderman M, Sutton V, Chopping J, Lask B. Dissociation
in adolescent girls with anorexia: relationship to comorbid psychopathology. J Nerv
Ment Dis 2002; 190:746–751.
82. Zebre KJ. Selves that starve and suffocate: the continuum of eating disorders and
dissociative phenomena. Bull Menninger Clin 1993; 57:319–327.
83. Levin AP, Kahan M, Lamm JB, Spauster E. Multiple personality in eating disorder
patients. Int J Eat Disord 1993; 13:235–239.
84. Greenes D, Fava M, Cioffi J, Herzog DB. The relationship of depression to
dissociation in patients with bulimia nervosa. J Psychiatr Res 1993; 27:133–137.
85. Vanderlinden J, Vandereycken W, van Dyck R, Vertommen H. Dissociative
experiences and trauma in eating disorders. Int J Eat Disord 1993; 13:187–193.
86. Santonastaso P, Favaro A, Olivotto MC, Friederici S. Dissociative experiences and
eating disorders in a female college sample. Psychopathology 1997; 30:170–176.
87. Meyer C, Waller G. Dissociation and eating psychopathology: gender differences in
a nonclinical population. Int J Eat Disord 1998; 23:217–221.
88. Nijenhuis ER, van Dyck R, Spinhoven P, van der Hart O, Chatrou M,
Vanderlinden J, Moene F. Somatoform dissociation discriminates among diagnostic
categories over and above general psychopathology. Aust N Z J Psychiatry 1999;
33:511–520.
89. Bienvenu OJ, Samuels JF, Riddle MA, Hoehn-Saric R, Liang KY, Cullen BA,
Grados MA, Nestadt G. The relationship of obsessive-compulsive disorder to
possible spectmm disorders: results from a family study. Biol Psychiatry 2000; 48:
287–293.
90. Rabe-Jablonska Jolanta J, Sobow Tomasz M. The links between body dysmorphic
disorder and eating disorders. Eur Psychiatry. J Assoc Eur Psychiatrists 2000; 15:
302–305.
91. Grant JE, Kim SW, Eckert ED. Body dysmorphic disorder in patients with
anorexia nervosa: prevalence, clinical features, and delusionality of body image. Int
J Eat Disord 2002; 32:291–300.
92. Lieb R, Zimmermann P, Friis RH, Hofler M, Tholen S, Wittchen HU. The natural
course of DSM-IV somatoform disorders and syndromes among adolescents and
young adults: a prospective-longitudinal community study. Eur Psychiatry J Assoc
Eur Psychiatrists 2002; 17:321–331.
93. Nierenberg AA, Phillips KA, Petersen TJ, Kelly KE, Alpert JE, Worthington JJ,
Tedlow JR, Rosenbaum JF, Fava M. Body dysmorphic disorder in outpatients with
major depression. J Affect Dis 2002; 69:141–148.
94. Imbierowicz K, Egle UT. Childhood adversities in patients with fibromyalgia and
somatoform pain disorder. Eur Journal Pain 2003; 7:113–119.
95. Roelofs K, Keijsers GP, Hoogduin KA, Naring GW, Moene FC. Childhood abuse
in patients with conversion disorder. Am J Psychiatry 2002; 159:1908–1913.
96. Farley M, Patsalides BM. Physical symptoms, posttraumatic stress disorder, and
healthcare utilization of women with and without childhood physical and sexual
abuse. Psychol Rep 2001; 89:595–606.
97. Sansone RA, Gaither GA, Sansone LA. Childhood trauma and adult somatic
preoccupation by body area among women in an internal medicine setting: a pilot
study. Int J Psychol Med 2001; 31:147–154.
98. George MS, Brewerton TD, Cochrane CE. Trichotillomania, trichophagy and
bulimia nervosa. N Engl J Med 1990; 322:470–471.
99. Goldman MJ. Kleptomania: making sense of the nonsensical. Am J Psychiatry 1991;
148:986–996.
100. Baum A, Goldner EM. The relationship between stealing and eating disorders: a
review. Harvard Review of Psychiatry 1995; 3:210–221.
101. McElroy SL, Hudson JI, Pope HG, Keck PE. Kleptomania: clinical characteristics
and associated psychopathology. Psychol Med 1991; 21:93–108.
102. Favaro A, Santonastaso P. Self-injurious behavior in anorexia nervosa. J Nerv Ment
Dis 2000; 188:537–542.
104. McElroy SL, Keck PE Jr, Phillips KA. Kleptomania, compulsive buying, and binge-
eating disorder. J Clin Psychiatry 1995; 56(suppl 4): 14–26; discussion 27.
105. Zlotnick C, Mattia JI, Zimmerman M. Clinical correlates of self-mutilation in a
sample of general psychiatric patients. J Nerv Ment Dis 1999; 187:296–301.
106. Lowenstein LF. The etiology, diagnosis and treatment of the fire-setting behaviour
of children. Child Psychiatry Hum Dev 1999–1999; 19:186–194.
107. McElroy SL, Phillips KA, Keck PE Jr. Obsessive compulsive spectrum disorder. J
Clin Psychiatry 1994; 55(suppl):33–51; discussion 52–53.
108. Stein DJ. Neurobiology of the obsessive-compulsive spectrum disorders. Biol
Psychiatry 2000; 47:296–304.
109. Hollander E, Rosen J. Impulsivity. J Psychopharmacol 2000; 14(suppl 1, 2):
S39-S44.
110. Messman-Moore TL, Long PJ, Siegfried NJ. The revictimization of child sexual
abuse survivors: an examination of the adjustment of college women with child
sexual abuse, adult sexual assault, and adult physical abuse. Child Maltreatment
2000; 5:18–27.
111. Humphrey JA, White JW. Women’s vulnerability to sexual assault from
adolescence to young adulthood. J Adol Health 2000; 27:419–424.
112. Field NP, Classen C, Butler LD, Koopman C, Zarcone J, Spiegel D.
Revictimization and information processing in women survivors of childhood
sexual abuse. J Anxiety Disord 2001; 15:459–469.
113. Kellogg ND, Hoffman TJ. Child sexual revictimization by multiple perpetrators.
Child Abuse Neglect 1997; 21:953–964.
114. Drimmer EJ. Stimulant treatment of bulimia nervosa with and without attention-
deficit disorder: three case reports. Nutrition 2003; 19:76–77.
544 BREWERTON
115. Sokol MS, Gray NS, Goldstein A, Kaye WH. Methylphenidate treatment for bulimia
nervosa associated with a cluster B personality disorder. Int J Eat Disord 1999; 25:
233–237.
116. Schweickert LA, Strober M, Moskowitz A. Efficacy of methylphenidate in bulimia
nervosa comorbid with attention-deficit hyperactivity disorder: a case report. Int J
Eat Disord 1997; 21:299–301.
117. Glod CA, Teicher MH. Relationship between early abuse, posttraumatic stress
disorder, and activity levels in prepubertal children. J Am Acad Child Adol
Psychiatry 1996; 35:1384–1393.
118. Hinshaw SP. Preadolescent girls with attention-deficit/hyperactivity disorder: I.
Background characteristics, comorbidity, cognitive and social functioning, and
parenting practices. J Consult Clin Psychol 2002; 70:1086–1098.
119. Heffron WM, Martin CA, Welsh RJ, Perry P, Moore CK. Hyperactivity and child
abuse. Can J Psychiatry 1987; 32:384–386.
120. Cohen AJ, Adler N, Kaplan SJ, Pelcovitz D, Mandel FS. Interactional effects of
marital status and physical abuse on adolescent psychopathology. Child Abuse
Neglect. 2002; 26:277–288.
121. Thompson KM, Wonderlich SA, Crosby RD, Mitchell JE. The neglected link
between eating disturbances and aggressive behavior in girls. J Am Acad Child
122. Geist R, Davis R, Heinmaa M. Binge/purge symptoms and comorbidity in
adolescents with eating disorders. Can J Psychiatry 1998; 43:507–512.
123. Ford JD, Racusin R, Ellis CG, Daviss WB, Reiser J, Fleischer A, Thomas J. Child
maltreatment, other trauma exposure, and posttraumatic symptomatology among
children with oppositional defiant and attention deficit hyperactivity disorders.
Child Maltreatment 2000; 5:205–217.
124. Ford JD, Racusin R, Daviss WB, Ellis CG, Thomas J, Rogers K, Reiser J,
Schiffman J, Sengupta A. Trauma exposure among children with
oppositional defiant disorder and attention deficit-hyperactivity disorder. J Consult
Clin Psychol 1999; 67:786–789.
125. Zaider TI, Johnson JG, Cockell SJ. Psychiatric comorbidity associated with eating
disorder symptomatology among adolescents in the community. Intl J Eat Disord
2000; 28:58–67.
126. Johnson JG, Cohen P, Kotler L, Kasen S, Brook JS. Psychiatric disorders associated
with risk for the development of eating disorders during adolescence and early
adulthood. J Consult Clin Psychol 2002; 70:1119–1128.
127. Zanarini MC, Frankenburg FR, Dubo ED, Sickel AE, Trikha A, Levin A, Reynolds
V. Axis I comorbidity of borderline personality disorder. Am J Psychiatry 1998;
155:1733–1739.
128. Yen S, Shea MT, Battle CL, Johnson DM, Zlotnick C, Dolan-Sewell R, Skodol AE,
Grilo CM, Gunderson JG, Sanislow CA, Zanarini MC, Bender DS, Rettew JB,
McGlashan TH. Traumatic exposure and posttraumatic stress disorder in
borderline, schizotypal, avoidant, and obsessive-compulsive personality disorders:
findings from the collaborative longitudinal personality disorders study. J Nerv

Ment Dis 2002; 190:510–518.
129. Ruchkin VV, Schwab-Stone M, Koposov R, Vermeiren R, Steiner H. Violence
exposure, posttraumatic stress, and personality in juvenile delinquents. J Am Acad
130. Shea MT, Zlotnick C, Dolan R, Warshaw MG, Phillips KA, Brown P, Keller MB.
Personality disorders, history of trauma, and posttraumatic stress disorder in
subjects with anxiety disorders. Compr Psychiatry 2000; 41:315–325.
131. Kaplan AS, Garfinkel PE. Difficulties in treating patients with eating disorders: a
review of patient and clinician variables. Can J Psychiatry 1999; 44:665–670.
132. Zimmerman M, Mattia JI. Axis I diagnostic comorbidity and borderline personality
disorder. Compr Psychiatry 1999; 40:245–252.
133. Zanarini MC, Gunderson JG, Frankenburg FR. Axis I phenomenology of
borderline personality disorder. Compr Psychiatry 1989; 30:149–156.
134. Sullivan PF, Joyce PR, Mulder RT. Borderline personality disorder in major
depression. J Nerv Ment Dis 1994; 182:508–516.
135. Johnson JG, Cohen P, Brown J, Smailes EM, Bernstein DP. Childhood
maltreatment increases risk for personality disorders during early adulthood. Arch
136. Bell L. Does concurrent psychopathology at presentation influence response to
treatment for bulimia nervosa? Eating Weight Disord 2002; 7:168–181.
137. Agras WS, Crow SJ, Halmi KA, Mitchell JE, Wilson GT, Kraemer HC. Outcome
multi-site study. Am J Psychiatry 2000; 157:1302–1308.
138. Sohlberg S, Norring C, Holmgren S, Rosmark B. Impulsivity and long-term
prognosis of psychiatric patients with anorexia nervosa/bulimia nervosa. J Nerv
MentDisl989; 177:249–258.
139. Anderson KP, LaPorte DJ, Brandt H, Crawford S. Sexual abuse and bulimia:
response to inpatient treatment and preliminary outcome. J Psychiatr Res 1997; 31:
621–633.
140. Kessler RC, Nelson CB, McGonagle KA, Liu J, Swartz M, Blazer DG.
Comorbidity of DSM-III-R major depressive disorder in the general population:
results from the US National Comorbidity Survey. Br J Psychiatry Suppl 1996; 30:
17–30.
141. Pettinati HM, Horne RL, Staats JM. Hypnotizability in patients with anorexia
nervosa and bulimia. Arch Gen Psychiatry 1985; 42:1014–1016.
142. Schoemaker C, Smit F, Bijl RV, Vollebergh WA. Bulimia nervosa following
psychological and multiple child abuse: support for the self-medication hypothesis
in a population-based cohort study. Intl J Eat Disord 2002; 32:381–388.
143. Garner DM, Garfinkel PE, eds. Handbook of Psychotherapy for Anorexia Nervosa
and Bulimia. New York: Guilford Press, 1985.
144. Vandereycken W, Kog E, Vanderlinden J, eds. The Family Approach to Eating
Disorders: Assessment and Treatment of Anorexia Nervosa and Bulimia. New
York: PMA Publishing Corporation, 1989.
546 BREWERTON
146. Cochrane CE, Brewerton TD, Hodges EL, Wilson D. Alexithymia in the eating
147. Feld R, Woodside DB, Kaplan AS, Olmsted MP, Carter JC. Pretreatment
motivational enhancement therapy for eating disorders: a pilot study. Int J Eat Disord
2001; 29:393–400.
with borderline personality disorder. Am J Psychiatry 2001; 158(suppl 10): 1–52.
149. Expert consensus guidelines for the treatment of posttraumatic stress disorder.
J Clin Psych 1999; 60(suppl 16):6–76.
150. Ballenger JC, Davidson JR, Lecmbier Y, Nutt DJ, Foa EB, Kessler RC, McFarlane
AC, Shalev AY. Consensus statement on posttraumatic stress disorder from the
International Consensus Group on Depression and Anxiety. J Clin Psychiatry 2000;
61(Suppl 5):60–66.
151. Schnicke M, Resick PA. Cognitive Processing Therapy for Rape Victims: A
Treatment Manual. London: Sage Publications, 1993.
152. Foa E, Rothbaum B. Treating the Trauma of Rape: Cognitive-Behavioral Therapy
for PTSD. New York: Guilford Press, 2001.
153. Keane T, Foa E, Friedman M. Effective Treatments for PTSD: Practice Guidelines
from the International Society for Traumatic Stress Studies. New York: Guilford
Press, 2000.
154. Heflin AH, Deblinger E. Treating Sexually Abused Children and Their
Nonoffending Parents: A Cognitive Behavioral Approach. London: Sage
Publications, 1996.
155. Shapiro F. Eye Movement Desensitization and Reprocessing (EMDR). 2d ed. Basic
Principles, Protocols, and Procedures. New York: Guilford Press, 2001.
156. Brewerton TD, Shannon M. Possible clozapine exacerbation of bulimia nervosa.
Am J Psychiatry 1992; 149:1408–1409.
157. McElroy SL, Casuto LS, Nelson EB, Lake KA, Soutullo CA, Keck PE Jr, Hudson
JI. Placebo-controlled trial of sertraline in the treatment of binge eating disorder.
Am J Psychiatry 2000; 157:1004–1006.
158. Ricca V, Mannucci E, Mezzani B, Moretti S, Di Bernardo M, Bertelli M, Rotella
CM, Faravelli C. Fluoxetine and fluvoxamine combined with individual cognitive-
behaviour therapy in binge eating disorder: a one-year follow-up study. Psychother
Psychosom 2001; 70:298–306.
159. Albert U, Aguglia E, Maina G, Bogetto F. Venlafaxine versus clomipramine in the
treatment of obsessive-compulsive disorder: a preliminary single-blind, 12week,
controlled study. J Clin Psychiatry 2002; 63:1004–1009.
160. Anderson JW, Greenway FL, Fujioka K, Gadde KM, McKenney J, O’Neil PM.
Bupropion SR enhances weight loss: a 48-week double-blind, placebocontrolled
trial. Obesity Res 2002; 10:633–641.
161. El-Giamal N, de Zwaan M, Bailer U, Lennkh C, Schussler P, Strnad A, Kasper S.

Reboxetine in the treatment of bulimia nervosa: a report of seven cases. Int Clin
162. Lam RW, Lee SK, Tam EM, Grewal A, Yatham LN. An open trial of light therapy
for women with seasonal affective disorder and comorbid bulimia nervosa. J Clin
Psychiatry 2001; 62:164–168.
163. McElroy SL, Arnold LM, Shapira NA, Keck PE Jr, Rosenthal NR, Karim MR, Kamin
M, Hudson JI. Topiramate in the treatment of binge eating disorder associated with
obesity: a randomized, placebo-controlled trial. Am J Psychiatry2003; 160:
255–261.
164. Marrazzi MA, Bacon JP, Kinzie J, Luby ED. Naltrexone use in the treatment of
anorexia nervosa and bulimia nervosa. Int Clin Psychopharmacol 1995; 10:
163–172.
165. Neumeister A, Winkler A, Wober-Bingol C. Addition of naltrexone to fluoxetine
in the treatment of binge eating disorder. Am J Psychiatry 1999; 156:797.
166. Wilens TE, Faraone SV, Biederman J, Gunawardene S. Does stimulant therapy of
attention-deficit/hyperactivity disorder beget later substance abuse? A metaanalytic
review of the literature. Pediatrics 2003; 11:179–185.
167. Saunders BE, Berliner L, Hanson RF, Eds. Child Physical and Sexual Abuse:
Guidelines for Treatment Final Report, January 15, 2003. Charleston, SC: National
Crime Victims Research and Treatment Center, 2003.
548 BREWERTON
23
Future Directions in the Management of
Eating Disorders
Joel Yager
University of New Mexico School of Medicine
Albuquerque, New Mexico, U.S.A.
Making predictions about the future is fraught with uncertainty, and humility is
always in order for those foolish enough to offer more than the most general of
prognostications. Professional futurologists employ a variety of specific
technical methods, including trend extrapolation, genius forecasting, consensus
methods, simulations, cross-impact matrix methods, scenario building, decision
trees, and creative disorder (a method that relies on innovations coming from
grass-roots levels). None of these methods may be capable of capturing
unanticipated events, known as “wild cards,” or successfully anticipating their
impacts. Lacking genius, a group capable of achieving consensus, and suitable
simulations, the safest statements one can offer are based on trend analyses, and
some imaginative scenario building and cross-matrix thinking(1,2).
A few lessons learned from futurist studies can guide our thinking about
eating disorders treatment from the start: First, technical changes always occur
more rapidly than social changes. Second, futurists usually tend to be more
optimistic about the occurrence and rates of change, particularly social change,
than is usually born out by history. Third, futurists tend to underestimate the
expenses involved in producing the changes they envision. Fourth, to help put
things into perspective, Arthur Clarke’s three laws of the future should always
be born in mind, to wit: (a) when a distinguished but elderly scientist says that
something is possible, he is almost certainly right. When he states that
something is impossible, he is very probably wrong. (b) The only way to
discover the limits of the possible is to go beyond them to the impossible.
(c) Any sufficiently advanced technology is indistinguishable from magic (3).
With respect to medicine and psychiatry, the rapid advances in neuroimaging
and genomics of the past few decades might seen like magic to those working in
this field a few decades ago. Where “magical” technological advances will appear
and how they will manifest in the next few decades can only be imagined.
Where the funds and necessary public will come from to pay for the services
that are likely to be possible, even demanded, is anyone’s guess.
550 YAGER
Thinking about future treatments requires that we consider the future of

diagnosis, epidemiology, etiology, and pathogenesis, and how these shifts may
translate to new treatments that take account of nutritional, medical, and
psychobehavioral interventions, the settings in which care will be delivered, and
the systems of care in which treatments are likely to occur. With the obvious
caveat that anything I say about the future may turn out to be entirely wrong,
the following trends seem to me to be most likely to influence how the future
management of eating disorders will shape up.
THE FUTURE OF EATING DISORDERS DIAGNOSIS

Clearly, our current diagnostic schemes are, at best, works in progress, social
constructions that attempt to connect sparse data points into functional
categories. The DSM-IV and ICD-10 languages now employed are very recent,
and there are many dissenters whose displeasure with contemporary
nomenclature is freely voiced. Controversies about how eating disorders
phenomena should be correctly lumped and split into meaningful syndromes
and disorders, let alone diseases, are rampant and assure that the shape of eating
disorders in DSM-V, probably a decade away, and ICD-10 will differ at least to
some extent from what exists today. Since evidence-based treatments are
supposed to be linked to valid diagnostic entities, the shape of treatment will be
strongly influenced by the evolution (if not revolution) we can anticipate in
diagnostics. Furthermore, since reimbursement for treatment is linked to
diagnosis, definitions of where to set the bar are of great practical significance for
patients, families, and clinicians.
In any era, psychological and biological treatments closely follow
contemporary conceptual models of etiology and pathogenesis. Historically,
psychological models and the psychotherapies they spawn have been noted to
parallel the physical sciences of their times. Thus, Freudian and early behavioral
models and their associated therapy metaphors loosely paralleled those of
Newtonian mechanics and physics, including hydraulic and electrical theories of
the late 19th century—how libidinal energy flowed from place to place. The
middle to late 20th century saw psychological models loosely based on cognitive
science and information theory. Future metaphors concerning the
psychopathology and management of eating disorders are likely to be informed
by models based on chaos and complexity theory, involving how complex
adaptive subsystems of the brain organize, communicate, and change with
various stressors and genetic unfoldings. We are likely to hear of psychological
“strange attractors,” phase transitions, and neural nets. At present, a diagnostic
system for eating disorders based on fundamental understanding at the
biological or bioecological level is not in sight, nor is it certain how long the
FUTURE DIRECTIONS 551
field will have to wait for such a system to be achieved. However, we might
anticipate that advances in functional as well as structural neuroimaging, genetic
analyses, complex systems, information processing, and related cognitive
sciences may lead to more discriminating diagnostic groupings based on
meaning biological differences within similar phenotypes. For the foreseeable
future, however, we will still rely on diagnostic systems based on clinical and
laboratory observations—phenomenology.
Already, the process of reviewing current diagnostic thinking for DSMV has
begun. Particular attention will be paid to what has been called “bioecological”
perspective, focusing on the first two decades of life, when rapid changes occur
in behavior, emotion, and cognition. Considerable attention will be paid to
developmental neuroscience, genetics, brain imaging, postmortem studies, and
animal models (4).
Well-recognized gaps exist in current categorical methods for diagnosing
personality disorders and their relationship with axis I disorders, and in the
limited provisions for the diagnosis of relational disorders in the current
nomenclature. These gaps have led to suggestions for dimensional
classifications, at least for personality disorders. Relationships between mental
disorders and measures of disability and impairment are likely to be revisited.
Particularly salient for eating disorders, the relationships of cultural factors to
psychopathology and the main cultural variables that operate in diagnostic
processes will be carefully reconsidered.
Crossing this DSM-V matrix with emerging discussions in the literature,
simple trend analysis suggests how several themes of the last several years may
color future discussions about eating disorders diagnoses. First, with regard to
nosological criteria, controversial questions have been raised regarding whether
amenorrhea, included as a diagnostic criterion for anorexia nervosa from DSM-
III on, should continue to be required as a criterion for anorexia nervosa (5).
Since experienced clinicians have seen many cases of otherwise frank anorexia
nervosa in which some degree of menstrual bleeding manages to persist, and
since this finding appears to have little if any significance regarding course or
outcome, amenorrhea may not be categorically required for distinctions
between “caseness” and “noncaseness” of anorexia nervosa. In addition,
controversies regarding whether the syndrome of binge eating disorder merits
elevation from its current lowly position, lumped in the EDNOS category, to
official diagnostic status as a distinct entity will be better informed by studies in
the next decade (6).
Regarding the relationship between axis II personality disorders and axis I
disorders, the recent research of Westen and Harnden-Fischer is informative.
Their studies support previous work suggesting that the course of anorexia
nervosa is strongly related to comorbid personality configuration. To
552 YAGER
summarize, patients with perfectionistic personalities and few other personality

quirks do best, whereas those with constricted and rigid personalities associated
with cluster A and C disorders, and those with chaotic patterns usually
associated with cluster B disorders, have much more difficult courses and are
less likely to recover (7). So long as we rely on phenomenologicalbased
nosology, careful delineation of diagnostic prototypes melding these sorts of
characteristics linked to data on treatment response and outcomes may provide
better diagnostic labels in the future.
Recent literature notes other clinical features that may work their way into
future diagnostic schemes. One is the distinction accorded to so-called “typical”
versus “atypical” anorexia nervosa. In the former, patients maintain the idea that
they are too fat, holding fast to this cognitive distortion, almost to a delusional
degree. In the latter, patients acknowledge that they are too thin but feel
helpless to combat their ego-alien thoughts and behaviors. These distinctions, in
some fashion parallel to the “delusional” versus “nondelusional” forms of body
dysmorphic disorder, have been shown by Strober et al. to have prognostic
significance (8). The “atypical” patients do somewhat better over the longer
term, perhaps in part due to the fact that they are more willing to accept and
participate in treatment.
Related to personality dimensions, typicality versus atypicality, and to
assessment of disability and impairment, diagnostic thinking in the future may
more routinely incorporate information regarding patients’ stages of motivation
for change (9), a paradigm extensively utilized in the field of addictions.
Although much additional work is required to develop reliable and valid
methods for assessing these stages, incorporating such concerns into diagnostic
systems should improve treatment planning and prognosticating.
Finally, several murky diagnostic entities and relationships are likely to be
clarified. The relationship of binge eating disorder to obesity, particularly
medically serious class II and class III obesity, will be better delineated. Various
types of night-eating syndrome, nocturnal eating disorders, and related
conditions will be more carefully described and a better classification system for
them will be developed.
THE FUTURE OF EATING DISORDERS

EPIDEMIOLOGY
How will the prevalence of eating disorders and their distribution in
communities change in the future? What sorts of associations with social,
cultural, and community-based factors are likely to emerge? Although clear-cut
predictions are no less risky than stock market predictions, several factors are
likely to contribute to trends. First, in the United States, rates of obesity have
been increasing substantially in spite of huge outcries from public health officials
about the importance of proper diet and exercise. To the extent that eating
disorders partly result from reactions against obesity-oriented tendencies,
pressures to develop eating disorders will remain strong in the community at
large and continue to affect those who are most vulnerable. Second, we can
think about how shifting social pressures may affect vulnerability factors such as
perfectionism, anxiety disorders, and depressive disorders. Might there be
relationships between the extent to which Western society continues to provide
increasing opportunities for self-fulfillment and individuation for women and
the extent to which perfectionistic tendencies are unleashed in those who
believe that excelling is of paramount importance? To what extent does today’s
society select for perfectionistic traits? Among the hypotheses recently advanced
to account for a seemingly significant increase in the numbers of children with
Asperger’s disorder in California’s Silicon Valley area is one based on the
observation that many of these children emerge from families in which both
parents were “computer nerds” (10). The question raised is whether the
selective breeding of “nerds” is more likely to breed Asperger’s children. In
parallel fashion, to what extent do contemporary societal pressures increase the
selective and assortive mating of perfectionistic men with perfectionistic
women, thereby producing higher likelihoods of perfectionistic offspring
vulnerable to anorexia nervosa? Are children with anorexia nervosa more likely
to emerge from homes parented by hard-driving, status-seeking couples than
from others?
In another vein, Sapolsky points to factors he believes are likely to contribute
to his prediction of increased rates of major depression in society. He posits that
ongoing rapid rates of social change and reductions in those social structures
that probably counter depressogenic tendencies, signaling breakdown of
community, add considerably to the stresses likely to produce depression (and,
from our perspective, potentially eating disorders as well) (11). Factors include
the increasing fragmentation of families into nuclear and one-parent families
that lack easily accessible grandparents, aunts, uncles, and cousins as
overstressed youth raised by overtaxed parents grow up. Absent the immediate
availability of extended families, youth may miss the benefits afforded by the
mediating, moderating, and supportive influences these other relatives may
provide regarding how to live. Studies concerning the relative risk for eating
disorders among adolescents raised in proximity to large extended families
versus those reared in more isolated environments may illuminate this intriguing
possibility.
Finally, it seems reasonable to predict that as globalization extends the ready
availability of fat-rich foods, opportunities for individuation and personal
choice, and the trend-setting influences of MTV and the Style Channel, the
554 YAGER
prevalence of eating disorders is likely to increase considerably in countries

where these problems are now relatively uncommon. Such trends are evident
from Becker et al.’s studies of the impact of introducing television in Fiji, where
negative self-evaluations concerning weight and shape emerged among
adolescent girls for whom these issues were previously unimportant (12).
Future implications for prevention and developing intervention services are clear.
MOLECULAR GENETIC RESEARCH IN EATING

DISORDERS: FUTURE IMPLICATION FOR
TREATMENT
As detailed in other chapters, several lines of evidence suggest genetic
vulnerability for eating disorders. These include high rates of familial
transmission. Twin studies show rates of concordance approximately three times
higher in identical twins than dizygotic twins, implying that both genetic and
psychosocial factors are active. Monozygotic twins have roughly a 50%
concordance rate for eating disorders (13). Exactly what is being inherited
remains controversial and, clearly, genetics alone is not fully responsible.
Rather, genetic factors appear to predispose, probably through effects on
temperament, cognitive style, personality, mood regulating tendencies, set
points for weight, and predispositions toward physical activity. Furthermore,
still elusive, studies imply that variations in genetically mediated serotonin
regulation may be important. Another hint pointing to genetic influence is the
identification of genetically derived animal models of anorexia nervosa (such as
“thin sow” disease—hyperactive sows that starve themselves, emerging during
breeding experiments to produce thinner hogs) (14). As a result, active searches
are underway to identify specific genes that may contribute to the appearance of
eating disorders. Molecular genetic research on these disorders is in its infant
stages. However, promising areas for future research have already been
identified. These include genes involved with 5-hydroxytryptophan (5-HT,
serotonin) receptors, specifically 5-HT2A, 5-HT2C, and 5-HT1B receptors,
uncoupling protein receptors (UCP-2/UCP-3), and estrogen β receptors.
Several large-scale linkage and association studies are underway (15–19).
Recent advances in unraveling the mechanisms of weight control point to a
crucial role of the melanocortin-4 receptor (MC4-r) systm in regulating body
weight. The orexigenic neuropeptide agouti-related protein (AGRP), an MC4-r
antagonist, has a crucial role in maintaining body weight by inducing food
intake (20). But overall, the percentage of occurrence explained by currently
studied genetic association is quite small. Furthermore, twin studies suggest that
approximately 17–46% of the variance in both anorexia nervosa and bulimia
nervosa can be accounted for by nonshared environmental factors (21). Future
questions will concern subtle genenurture interactions, which vary with

different environmental conditions throughout a person’s lifelong development.
These genetic studies and the ones that will follow may ultimately lead to a
sophisticated form of genetic profiling—gene scans—based on singlenucleotide
polymorphisms, or SNPs (“snips”) (22). Such scans will be able to easily run
through the hundreds of genes now known to be involved in regulating aspects
of intake, satiety, physical exercise, and emotional regulation, and the
thousands still to be associated with these functions. The implications for
prevention and treatment are many. Conceivably, SNP profiling may facilitate
more accurate predictions about risk for diseases, including eating disorders,
which may lead to sophisticated psychobehavioral-social and biological
preventions and interventions. A cadre of health professionals (and, no doubt,
many Web-based software programs) will be developed that specialize in
genetic counseling based on SNP profiling. Respect for SNP-related
temperamental vulnerabilities may encourage particularly susceptible
individuals toward—or away from—risky pursuits, behaviors, and
environments before onset, and toward modifications of diet, pharmafoods,
sleep, and other more biologically based functions, all designed to stave off
illnesses such as eating disorders. Conceivably, athletic and dance coaches,
employers, and others may use such profiling to select and/or deselect
individuals for participation in specific sports, vocations, and avocations.
Furthermore, inventive genetic engineering may produce animal models of
eating disorders that more closely resemble clinical disorders and that may
further illuminate issues in pathogenesis and treatment.
OTHER FUTURE BIOLOGICAL INVESTIGATIONS

IN EATING DISORDERS
To date, imaging studies have offered important leads in understanding and
corroborating clinical observations, but as yet no findings have firmly
established preexisting functional or structural abnormalities of the brain in
patients with eating disorders. Studies in adolescents with anorexia
nervosa showing that gray matter deficits (23) and unusual temporal lobe
vascular flow asymmetries (24) are found during semistarvation and may persist
following weight restoration await extension, replication, and confirmation. As
yet, brain imaging studies of pre-disordered vulnerable populations have not
been conducted. Future research utilizing emerging technologies including new
iterations of nuclear magnetic resonance, diffusion tensor magnetic resonance
imaging (DT-MRI), and functional MRI will provide increasingly detailed
information about processes in the living brain. Findings from such imaging
studies may help improve diagnostic subtyping and also enable clinicians to
556 YAGER
more precisely prognosticate the time and extent of recovery in individual

cases.
Other emerging clinical laboratory assessment based on proteonomics,
neurosteroids, and kinins will yield improved possibilities for understanding
vulnerabilities and for offering prognoses regarding brain, bone, and
reproductive functions and other critical aspects of eating disorder-related
physiology.
THE FUTURE OF BIOLOGICAL INTERVENTIONS

FOR EATING DISORDERS
It is likely that for anorexia nervosa and bulimia nervosa the principal biological
intervention will continue to be good nutrition. How nutrients can best be
delivered acceptably, effectively, and efficiently will drive research and
practice. For example, controversies over the potential value of supplemental
nasogastric feeding should be settled by controlled studies, not argued out as a
matter of religious belief (25). Conceivably, technical devices that assist patients
to eat by reminding them to do so in a timely fashion every few hours and that
record and transmit eating and associated behaviors to a clinician may be
employed via personal digital assistants (PDAs).
Until much more is known about biological contributions to the etiology of
eating disorders, speculations about specific advances in pharmacological
approaches must be circumspect. Genetic profiling may enable clinicians to
tailor medication types and doses to individual patients in order to maximize
effectiveness, minimize side effects, and avoid deleterious drug interactions.
With eating disorder patients, depending on the actual clinical added value of such
studies, it should be possible to more accurately assess gastrointestinal
absorption, protein binding, and shifting volume distributions to better inform
medication prescribing. New forms of medication delivery, such as the use of
transdermal patches now being successfully employed with the monoamine
oxidase inhibitor selegiline and other agents (26), may avert tyramine reactions,
dietary restrictions, and other complications related to alterations in intestinal
absorption in eating disorder patients. Questions remain to be answered
regarding the potential utility of atypical antipsychotic medications (including
those that do not specifically have weight gain as a metabolic side effect, such as
ziprazodone) on their own, and as adjunctive treatments for mood and
obsessive-compulsive disorders associated with eating disorders. In addition to
assessing the role of newer thymoleptics, exciting case reports appearing
in the literature suggest that other agents to be investigated should include
inositol (used for depression and obsessive-compulsive disorder) (27) and
tramadol (28), among others. New classes of medications being developed for
mood and anxiety disorders in particular are likely to be used for patients with
eating disorders, initially “off label,” and then, where interesting findings are
obtained, in systematic clinical trials. The pipeline for new medications is large.
Emerging new medications for mood disorders that may merit investigation for
eating disorders include selective noradrenergic uptake inhibitors such as
reboxetine; antiglucocorticoids such as mifepristone (RU486); corticotropin-
releasing factor (CRF) antagonists and other neurosteroid-modulating agents;
5-adenosylmethionine; the serotonin enhancer tianeptine; substance P
antagonists and other agents affecting neurokinins; and reversible monoamine
oxidase inhibitors such as moclobemide. Also of interest are new anxiolytics
such as pregabalin, and talipexole, a dopamine D2 and α2-adrenergic agonist
currently used for Parkinson’s disease, now being studied for managing
symptoms of dissociation and hyperarousal (29).
In addition to pharmacological treatments, other biological treatments using
light therapy, vagus nerve stimulation and transcranial magnetic stimulation for
associated depression, and other physiologically active procedures remain to be
explored for specific types of complex eating disorders (30).
THE FUTURE OF PSYCHOSOCIAL

INTERVENTIONS FOR EATING DISORDERS
Psychotherapies
Several trends can be identified. It is likely that research will continue on
delineating evidence-based effective individual, family, and group
psychotherapies for eating disorders, and on deconstructing and reassembling the
underlying effective elements to give better guidance to practitioners facing
complex clinical situations. Although these studies are expensive and difficult to
carry out, a number of developments may facilitate such studies in the future.
These include the development of multisite and, indeed, multinational clinical
trials, permitting larger studies. In addition, one can envision the organization
of intraprofession or multiprofession “practice research net works,” essentially
aggregations of up to several thousand practitioners organized to answer specific
clinical questions in their practices. Such networks have already proven effective
for answering clinical and health services research questions in pediatrics, family
medicine, and psychiatry (31). While psychotherapies based on cognitive-
behavioral and interpersonal psychotherapy models have been dominant, fusions
of these approaches as in cognitive-analytical therapies are exciting.
Motivational enhancement techniques, dialectic behavior therapy for complex,
multi-impulsive patients, and experimental manual-based psychotherapies based
558 YAGER
on self-psychological theories are all being studied. Of particular interest are

home-based treatments in which families are carefully trained and supervised in
managing their child’s eating behavior (so-called Maudsley model studies by
Lock and colleagues) (32). The future will likely provide better guidance as to
which therapies best suit each type of patients. As previously noted, new
metapsychologies and accompanying psychotherapies are likely to emerge,
paralleling discoveries in cognitive neuroscience, complexity theories, and other
yet-to-be-delineated scientific paradigms.
THE IMPACT OF COMPUTERS AND

INFORMATION TECHNOLOGY ON EATING
DISORDERS MANAGEMENT
Exciting innovations in the management of eating disorders patients are likely to
emerge from the application of computer and information technology to the
management of eating disorders, particularly with the increasing availability of
broad-band and wireless technologies. These technologies offer substantial
capacity to extend care to eating disorder patients by enhancing the “four As” of
contemporary health care—accessibility, affordability, affability, and
accountability. Access is enhanced by providing convenience compatible with
busy lifestyles, increasing contact time between patients and providers. Services
to rural areas and home-bound populations are getting better with increasing
broad-band technology. Affordable business plans are now being developed.
Computers are clearly “affable”—increasingly user friendly, and often increasing
ease of communication. Treatment programs delivered by computer are
increasingly accountable, often based on evidencebased, focused therapies.
Information age health care has been conceived as operating on six separate
tiers: individual self-care, friends and family, self-help networks, health
professionals as facilitators, health professionals as partners, and health
professionals as authorities (33). Information-age care for eating disorders is
already operating on each of these levels and will undoubtedly expand in the
future.
Individuals, Families and Friends, and Self-Help

On-Line
For seeking information and care, these technologies seem well suited to eating
disorders, especially as large proportions of youth are already on-line.
According to recent figures, among all 15- to 24-year-olds, 90% have gone
online, and one in four (24%) has gotten “a lot” of health information from the
Internet. Among the 90% of all 15- to 24-year-olds who have ever gone online,
75% have used the Internet at least once to find health information and about
one in four has looked up information on weight issues, mental health, drugs
and alcohol, and violence (34).
A huge amount of information is currently available for the “on-line seeker,”
and information will predictably increase. In November 2002, one of the most
popular search engines, Google, listed about 1,040,000 sites pertinent to eating
disorders. Numerous organizations devoted to eating disorders are on-line. The
Center for Counseling and Health Resources, an on-line clearinghouse, posted
about 330 links related to eating disorders. Fortunately, the “top hits” on search
engines include a large number of wellregarded sites. Easily accessed, too, are
many commercial sites and personal web pages, some of which are very helpful,
but among which are many of uncertain quality. In mid-2002 the top eating
disorder sites listed on Google included Eating Disorders Awareness and
Prevention (EDAP; now becoming National Eating Disorders Association),
Something Fishy, Eating Disorders Mirror Mirror (Canada), ANRED,
EDReferral.com, the Eating Disorders Association (England), the Harvard
Eating Disorders Center, and the Academy for Eating Disorders. Although data
to estimate the extent to which these sites are accessed are hard to obtain, one
site, Something Fishy, indicated more than 2.9 million hits since 1995.
How good are these sites? Many questions remain concerning what criteria
should be used to evaluate them. In the future, well-respected eating disorders
organizations, such as the Academy for Eating Disorders, may take on the task
of defining criteria by which to judge eating disorders sites using parameters
such as their use of standard diagnostic criteria, the quality of information they
provide about these conditions, the range of reputable treatment options they
present, and the links they offer to reputable sources and facilities.
What do these sites provide? First, they provide considerable information
about diagnostic criteria, permitting individuals to do reasonably good
screenings for the existence of active or subclinical disorders in themselves or
others. Second, they provide considerable information about the causes
and courses of eating disorders, treatment, treatment resources, research, and
clinical trials. Of importance, on-line sites offer considerable social and
emotional support, personal narratives, chats, stories, poems, and memorials.
One year long qualitative study of an on-line support group assessed the
emotional and social support provided (alt.support.eating-disord, [ASED]). The
group developed its own rules, etiquette, and norms to create safe
communications and support and to explicitly reduce competitiveness. The
group provided a mix of problem solving and emotional support. In contrast to
face-to-face support groups, the lack of physical cues and presence seemed to
reduce competition regarding who is thinnest (35). Many other general and
specific eating disorders groups appear on-line. For example, the site
560 YAGER
http://www.healthy-place.com/Communities/Eating_Disorders/Site/comm_
calender.htm offers 12 different types of eating disorder on-line support groups
with different themes, including groups for eating disorder sufferers who also self-
injure and groups for parents of children with eating disorders. Clearly such
groups are not always available locally.
But some sites provide misinformation and misguidance. Of note, thanks to a
vigorous campaign initiated by ANAD, many former “proanorexia” sites that
appeared in abundance have been eliminated. However, some pro-anorexia
messages may still be encountered, and may deleteriously influence and
encourage young browsers who are intrigued and enchanted by the idea of
becoming anorexic.
Given the extent to which patients, families, and friends are likely to be using
the internet to acquire information about eating disorders, guidelines have been
developed to assist clinicians in dealing with their “wired” patients. Clinicians
are advised to be open minded and diplomatic about information patients bring
in from their net searches, understanding that the patient wants more
knowledge. Take the information seriously, discuss accurate and questionable
issues, suggest other web sites that might help, and suggest that patients not act
on information without consulting the provider (36).
Health Professionals as Facilitators, Partners, and

Authorities
The evidence in favor of “distance medicine” is increasing. In a recent
metaanalysis, 7 of 7 controlled clinical trials in general medicine showed
positive outcomes, improved performance, or significant benefits from
providerinitiated computerized communication. Problems concerned
Alzheimer’s caregivers, cardiac rehabilitation, and diabetes management (37).
Available research suggests that distance treatments may be applicable for eating
disorders as well.
The instruments available for offering computer-assisted services are
constantly improving and increasing. Currently, they include stand-alone
computers and PDAs, computer-based interviews, CD-ROM-based programs,
PDA-based tracking and reminding programs, virtual reality pro grams,
electronic communication via email, web-based programming, chats, audio- and
video-phones, and telemedicine setups using telephone line-based videophones,
Tl lines, and, soon, PC-based videophones.
Several forms of clinician-patient e-interactions have been described. They
include e-therapy in which ongoing helping relationships take place solely via
internet communication; mental health advice in which psychotherapists
respond to one question in depth with communication taking place solely via
internet; adjunct services in which internet or e-mail communications are used

to supplement traditional, in-person treatment; and behavioral telepsychiatry
which includes sophisticated videoconferencing for remote locations, as
extensions of traditional clinic or hospital care. Currently, the extent to which
these methods is employed varies considerably. E-mail (regular or encrypted) is
used most commonly. Some clinicians now use secure web-based messaging. A
few use real-time chats. Very few now use videoconferencing and voice-over-IP
(Internet phone) (International Society for Online Mental Health,
www.ismho.org; www.metanoia.org, May 2002). Future shifts in cost,
availability, and dissemination of these various technologies will influence
patterns of use in treating eating disorder patients. What follows are some
illustrative examples of innovative works in progress that are likely to influence
future practice and research.
Perhaps the most developed program using technology in relation to eating
disorder-related issues is “Student Bodies,” an e-intervention for weight- and
shape-concerned college students (who do not have frank eating disorders)
organized at Stanford University and San Diego State University, involving a
large number of collaborating investigators. With the aim of reducing body
dissatisfaction and eating disorder-related attitudes, these investigators have
conducted a number of studies using Internet-delivered, computer-assisted
programs (38–40). They have explored various combinations of therapeutic
elements in 8-week psychoeducational programs that have included face-to-face
group sessions for orientation, interactive text, audio, video, self-monitoring
journals and behavior change exercises, weekly readings (CBT based), progress
notes completed at each week’s log-in, encouragement for the posting of
weekly entries in on-line body image journals. They have explored the benefit of
anonymity in groups and have also used moderated weekly asynchronous and
synchronous (real-time) online groups in which discussions are linked to photos
and personal statements of the other participants, fostering a sense of personal
familiarity and connection. Studies using random assignments have compared
on-line results with those of comparison groups participating in classroom
forms of the program, “Body Traps,” and wait-list controls. In general, students
participating in the on-line groups have improved scores regarding eating
concerns, eating restraint, and drive for thinness in student bodies. Results have
been modestly effective for all participants and more useful for higher risk
participants. Of importance, no harm occurred to participants. Not surprising,
results corresponded with adherence to the program. Results for the classroom
program were comparable to wait-list controls.
The lessons derived in these studies on factors that increase participant
compliance, including ease of use, structure, reminders and prompts, will be
useful to others establishing distance treatment in the future. Of importance are
562 YAGER
the complex clinical safety, legal, and ethical issues emerging in the context of
computer-based treatments and distant medicine. These derive in part from the
fact that clinicians do not have access to patients’ nonverbal/behavioral cues,
lack control over the participant’s environment, and cannot assure the validity or
accuracy of information, among other factors. These issues resemble problems
of telephone encounters. Of particular concern, the potential for
miscommunication and delayed response to clinical emergencies, such as
postings of suicidal ideation, requires attention (41). In part, these concerns
have been addressed by reminding and reiterating that on-line communications
are not effective for crises, and by assuring that clinicians have accurate home
and work phone numbers for patients.
Adjunctive Use of e-Mail in Ambulatory Treatment

of Eating Disorder Patients
I have explored the potential use of employing frequent contact by e-mail as
adjunctive treatment for ambulatory adolescents with anorexia nervosa. My
experience with several dozen patients suggests that this intervention is
particularly useful for restricting anorexia nervosa patients who have obsessional
and perfectionistic traits. It is also helpful, but not as consistently used, by
patients with binge eating-purging types of anorexia nervosa and with bulimia
nervosa. The intervention is particularly helpful for patients living a distance
from treatment centers. Patients are usually asked to record and email details of
their daily food intake (including calorie counts) and exerciserelated caloric
expenditure. They often add other personal comments. Phone calls and face-to-
face visits are necessary for ongoing monitoring and especially when urgent
issues emerge. Health Insurance Portability and Accountability Act of 1996
(HIPAA) regulations regarding privacy must be kept in mind (42).
Pilot studies to identify and treat patients with bulimia nervosa entirely by e-
mail have been conducted by P. Robinson and M.A. Sefaty at the University of
London. A group of 23 subjects with a high morbidity for eating disorder and
depression symptoms were recruited via an e-mail solicitation, selected for a
3-month treatment trial, asked to e-mail twice weekly, and assigned for formal
CBT or eclectic approach, which included diary keeping as well as interpersonal
elements. These investigators found overall modest reductions in depression
and eating disorder symptom scores. They surmised that their average results
were less than would be anticipated from face-toface treatment. Of note, as a
result of these interventions most participants said they desired more therapy,
primarily face to face (43).
Other Emerging Treatments

At the University of North Dakota work in progress involves the treatment of
several patients with bulimia nervosa via telemedicine, using a 128-kbps ISDN
line, a relatively “low tech” method. Feasibility for treatment has been
demonstrated at 80 miles (44). Sansone has explored establishing carefully
determined patient-to-patient connections via e-mail, essentially establishing e-
mail “pen pals” to provide patients with additional emotional support. Careful
matching of patients requires consideration of such factors as age, degree of
isolation, personality traits, eating disorder diagnosis, and marital status. Thus
far, he has not connected patients with clear-cut borderline personality
disorders (45). In Germany, the University of Leipzig sponsors a professionally
led, on-line information and consultation service for persons with eating
disorders (www.ab-server.de) supported by a consortium of academic centers.
This service has received several hundred requests for information in the few
years it has been in existence.
Guided self-help programs, usually based on CBT treatment models, are
increasingly studied and appear to be helpful for a substantial minority of
patients as an important aspect of stepped care. These programs are widely
available through easily acquired published workbooks, and will be increasingly
administered by means of computer-aided self help programs using workbooks,
monitoring, and feedback on CD-ROM or Internet (46,47). Future research to
extend these models to include elements from interpersonal therapy, self-
psychology, and other approaches seems warranted.
Emerging Technologies
Wireless two-way appliances increasingly used for chronic disease management,
the so-called “personal health buddy,” may be employed for eating disorders,
facilitating reporting of eating, exercise, mood and associated features, and
responses from clinicians (http://www.healthhero.com). Research conducted
by Wonderlich et al. using personal digital assistants exemplifies these
developments (48). Other emerging clinical applications include computer-
based body image assessments (49,50) and virtual reality-based body image
assessment and treatment (51).
Computer technology may provide better ways to train larger numbers of
health professionals, families, and patients about eating disorders using well-
established distance education models that integrate the Internet, audio tapes
and videotapes, telephone supervision, etc. Such methods are already being used
internationally (A.Parker, personal communication; A.Barriguete, personal
communications).
564 YAGER
Futuristic Applications
Further into the future, a number of additional applications may be developed,
limited only by imagination, funding, and zeal. Of potential value for the
management of eating disorders are the following possibilities: PDAbased
intelligent agents linked to patient’s cognitive style; voice, face, and emotion
recognition artificial intelligence systems for computer-assisted selfguided
therapy; outsourcing distance psychotherapy internationally in a global market;
video-telemedicine for family therapy with disconnected families. Finally,
research and treatment may be aided by sophisticated simulations capable of
portraying increasingly complex models of individual and family dynamics.
Envision future versions of “Sim Family,” including the emergence of the Sims
On-Line, permitting patients and families to demonstrate and work out
important issues in family relationships (www.the sims.ea.com). This is not
science fiction.
A cautionary note: Use of electronic communications and information
technology in applications such as those described above introduces a huge set
of confidentiality, patient safety, professional boundary, liability, and related
ethical and legal concerns that must be carefully addressed through informed
consents and practice protocols. All those writing in these fields emphasize
these issues, and several offer important ethical and legal guidelines (52).
FUTURE SYSTEMS OF CARE FOR EATING

DISORDERS
Substantial concerns exist concerning what systems of care may be available to
treat and to fund the treatment of persons with eating disorders in the future.
While resources and funding will vary from jurisdiction to jurisdiction (and,
depending on insurance coverage, from family to family), overall it can be
expected that resources will rarely meet needs and that services will be lacking
in many communities. As parity legislation for severe mental disorders gradually
works its way through state and federal agendas in the United States, and as
severe eating disorders are gradually incorporated into the list of recognized
biologically based disorders, the capacity to treat patients appropriately may
increase. If funding is available, new, sorely needed treatment facilities may be
created to care for currently underserved patients. The political will and efforts
of family members of patients with severe eating disorders will be required to
keep eating disorders needs on the minds and in the programs of those
influential politicians, business persons, and health care ad ministrators whose
support will always be necessary to sustain the facilities and funding
infrastructure necessary for treating patients. Recent appellatecourt level
judicial decisions that support insurance coverage based on medical rather than
psychiatric benefits for treatment of malnutrition associated with severe eating
disorders may be precedent setting and help patients, families, and clinicians
obtain needed services (53).
Even if decent funding were to be made available, many communities lack
appropriate facilities and health providers properly trained to assess and treat
patients with severe eating disorders. Responsibility for increasing the
availability of training opportunities will fall to academic centers and
organizations with strong eating disorders capacities, and, as alluded to above,
should be enhanced by the increasing availability of distance learning. However,
distance learning will never substitute for live supervised clinical experience.
Funding and incentives for training providers will determine the extent to which
these needs are met in the future.
Regardless of the extent to which funding may improve, it should be clear
that the costs of medical care will always increase so as to consume more than
the available resources. No doubt, as technologically sophisticated methods for
assessing genes, brains, and fluids become increasingly available, costs of care
are likely to continue to increase substantially just to pay for these tests and
procedures. However, the fact that new tests and treatments become available
does not assure that they will necessarily contribute materially to better
outcomes. Therefore, excellent judgment will always be required to ascertain
the actual value of performing new procedures and treatments.
As the various health professionals involved in assessing and treating eating
disorder patients continue to carefully observe and evaluate their experiences,
ongoing attention will be required for continually revising, updating, and
optimizing not only the evidence-based treatments and practices that command
respect but also the judgment-infused future editions of practice guidelines.
One hopes that tomorrow’s practice guidelines will better delineate how
collaborating health providers can best employ potential tests and treatments to
assure that outcomes become progressively better for patients and their
families. One also hopes that tomorrow’s research, on which such
recommendations will be based, will be guided by an unending supply of
curiosity and increasingly sophisticated questions.
REFERENCES
1. Yager J. A futuristic view of psychiatry. In: Yager J, ed. The Future of Psychiatry as
a Medical Specialty. Washington, DC: American Psychiatric Press, 1989:135–159.
2. Brockman J, ed., The Next Fifty Years: Science in the First Half of the TwentyFirst
Century. New York: Vintage, 2002.
566 YAGER
3. Clarke AC. Technology and the future. In: Clarke AC, ed. Report on Planet
Three. New York: Signet, 1972:129–141.
4. Kupfer DJ, First MB, Regier DA, eds. A Research Agenda for DSM-V.
Washington, DC: American Psychiatric Association, 2002.
5. Garfinkel PE, Lin E, Goering P, Spegg C, Goldbloom D, Kennedy S, Kaplan AS,
Woodside DB. Should amenorrhoea be necessary for the diagnosis of anorexia
nervosa? Evidence from a Canadian community sample. Br J Psychiatry 1996; 168:
500–506.
6. Pincus HA, First M. Critical differences between binge eating and overeating. Arch
Gen Psychiatry 1999; 56:951.
7. Westen D, Harnden-Fischer J. Personality profiles in eating disorders: rethinking
the distinction between axis I and axis II. Am J Psychiatry 2001; 158:547–562.
8. Strober M, Freeman R, Morrell W. Atypical anorexia nervosa: separation from
typical cases in course and outcome in a long-term prospective study. Int J Eat
Disord 1999; 25:135–142.
9. Rieger E, Touyz SW, Beumont PJ. The Anorexia Nervosa Stages of Change
Questionnaire (ANSOCQ): information regarding its psychometric properties. Int
J Eat Disord 2002; 32:24–38.
10. Silberman S. The Geek Syndrome: Autism—and its milder cousin Asperger’s
syndrome—is surging among the children of Silicon Valley. Are math-and-tech
genes to blame? Wired Dec 2001; 9(12). http://www.wired.com/wired/archive/
9.12/aspergers.html.
11. Sapolsky RM. Will we still be sad fifty years from now? In: Brockman J, ed. The
Next Fifty Years: Science in the First Half of the Twenty-First Century. New York:
Vintage Press, 2002:105–113.
12. Becker AE, Burwell RA, Gilman SE, Herzog DB, Hamburg P. Eating behaviours
and attitudes following prolonged exposure to television among ethnic Fijian
adolescent girls. Br J Psychiatry 2002; 180, 509–514.
13. Gorwood P, Bouvard M, Mouren-Simeoni MC, Kipman A, Ades J. Genetics and
anorexia nervosa: a review of candidate genes. Psychiatr Genet 1998; 8(1): 1–12.
14. Treasure JL, Owen JB. Intriguing links between animal behavior and anorexia
15. Levitan RD, Kaplan AS, Masellis M, Basile VS, Walker ML, Lipson N, Siegel GI,
Woodside DB, Macciardi FM, Kennedy SH, Kennedy JL. Polymorphism of the
serotonin 5-HT1B receptor gene (HTRIB) associated with minimum lifetime body
mass index in women with bulimia nervosa. Biol Psychiatry 2001; 50:640–643.
16. Westberg L, Bah J, Rastam M, Gillberg C, Wentz E, Melke J, Hellstrand M,
Eriksson E. Association between a polymorphism of the 5-HT2c receptor and
weight loss in teenage girls. Neuropsychopharmacology 2002; 26:789–793.
17. Klump KL, Kaye WH, Strober M. The evolving genetic foundations of eating
disorders. Psychiatr Clin North Am 2001; 24:215–225.
18. Hu X, Murphy F, Karwautz A, Li T, Freeman B, Franklin D, Giotakis O, Treasure
J, Collier DA. Analysis of microsatellite markers at the UCP2/UCP3 locus on
chromosome 1 Iql3 in anorexia nervosa. Mol Psychiatry 2002; 7:276–277.
19. Eastwood H, Brown KM, Markovic D, Pieri LF. Variation in the ESR 1 and ESR2
genes and genetic susceptibility to anorexia nervosa. Mol Psychiatry 2002; 7:
86–89.
20. Vink T, Hinney A, van Elburg AA, van Goozen SH, Sandkuijl LA, Sinke RJ,
Herpertz-Dahlmann BM, Hebebrand J, Remschmidt H, van Engeland H, Adan RA.
Association between an agouti-related protein gene polymorphism and anorexia
nervosa. Mol Psychiatry 2001; 6:325–328.
21. Klump KL, Wonderlich S, Lehoux P, Lilenfeld LR, Bulik CM. Does environment
matter? A review of nonshared environment and eating disorders. Int J Eat Disord
2002; 31:118–135.
22. Schork NJ, Fallin D, Lanchbury JS. Single nucleotide polymorphisms and the future
of genetic epidemiology. Clin Genet 2000; 58:250–264.
23. Katzman DK, Zipursky RB, Lambe EK, Mikulis DJ. A longitudinal magnetic
resonance imaging study of brain changes in adolescents with anorexia nervosa.
Arch Pediatr Adol Med 1997; 151:793–797.
24. Gordon I, Lask B, Bryant-Waugh R, Christie D, Timimi S. Childhood-onset
anorexia nervosa: towards identifying a biological substrate. Int J Eat Disord 1997;
22:159–165.
25. Robb AS, Silber TJ, Orrell-Valente JK, Valadez-Meltzer A, Ellis N, Dadson MJ,
Chatoor I. Supplemental nocturnal nasogastric refeeding for better short-term
outcome in hospitalized adolescent girls with anorexia nervosa. Am J Psychiatry
2002; 159:1347–1353.
26. Bodkin JA, Amsterdam JD. Transdermal selegiline in major depression: a double-
blind, placebo-controlled, parallel-group study in outpatients. Am J Psychiatry
2002; 159(11):1869–1875.
27. Levine J. Controlled trials of inositol in psychiatry. Eur Neuropsychopharmacol
1997; 7:147–155.
28. Mendelson SD. Treatment of anorexia nervosa with tramadol. Am J Psychiatry
2001; 158:963–964.
29. Boschert S. Novel antidepressant therapies near approval. Clin Psychiatry News
2002; 30(10): 1–20.
30. Rush AJ, George MS, Sackeim HA, Marangell LB, Husain M, Giller C, Nahas Z,
Haines S, Simpson RK Jr, Goodman R. Vagus nerve stimulation (VNS) for
treatment-resistant depressions: a multicenter study. Biol Psychiatry 2000; 47:
276–286.
31. Pincus HA, Zarin DA, Tanielian TL, Johnson JL, West JC, Pettit AR, Marcus SC,
Kessler RC, McIntyre JS. Psychiatric patients and treatments in 1997: findings from
the American Psychiatric Practice Research Network. Arch Gen Psychiatry 1999;
56:441–449.
32. Lock J, Le Grange D, Agras S, Dare C. Treatment Manual for Anorexia Nervosa: A
Family-Based Approach. New York: Guilford Press, 2001.
33. Ferguson T. Health Online: How to Find Health Information, Support Groups, and
Self-Help Communities in Cyberspace. Boston: Addison-Wesley, 1996.
568 YAGER
34. Kaiser Family Foundation. How young people use the internet for health
information. GenerationX.Com.,http://www.kff.org/content/2001/2001121
la/GenerationRx.pdf.
35. Walstrom MK. “You know, who’s the thinnest?”: Combating surveillance and
creating safety in coping with eating disorders online. Cyber Psychol Behav 2000;
3:761–783.
36. Chin T. Site reading: physicians grapple with recommending websites. Am Med
News, Oct 23–30, 2000, 24 (http://www.ama-assn.org/sci-pubs/amnews/
pick_00/tesa 102 3. htm).
37. Balas EA, Jaffrey F, Kuperman GJ, Boren SA, Brown GD, Pinciroli F, Mitchell JA.
Electronic communication with patients. Evaluation of distance medicine
technology. JAMA 1997; 278:152–159.
38. Winzelberg AJ, Eppstein D, Eldredge KL, Wilfley D, Dasmahapatra R, Dev P,
Taylor CB. Effectiveness of an internet-based program for reducing risk factors for
eating disorders. J Consult Clin Psychol 2000; 68:346–350.
39. Celio AA, Winzelberg AJ, Wilfley DE, Eppstein-Herald D, Springer EA, Dev P,
Taylor CB. Reducing risk factors for eating disorders: comparison of an internetand
a classroom-delivered psychoeducational program. J Consult Clin Psychol 2000;
68:650–657.
40. Zabinski MF, Wilfley DE, Pung MA, Winzelberg AJ, Eldredge K, Taylor CB. An
interactive internet-based intervention for women at risk of eating disorders: a
pilot study. Int J Eat Disord 2001; 30:129–137.
41. Humphreys K, Winzelberg A, Klaw E. Psychologists’ ethical responsibilities in the
Internet-based groups: issues, strategies, and a call for dialogue. Prof Psychol Res
Pract 2000; 31:493–496.
42. Yager J. E-mail as a therapeutic adjunct in the outpatient treatment of anorexia
nervosa: illustrative case material and discussion of the issues. Int J Eat Disord
2001; 29:125–138.
43. Robinson PH, Sefaty MA. The use of e-mail in the identification of bulimia nervosa
and its treatment. Eur. Eat Disord Rev 2001; 9:182–193.
44. Bakke B, Mitchell J, Wonderlich S, Erickson R. Administering cognitivebehavioral
therapy for bulimia nervosa via telemedicine in rural settings. Int J Eat Disord 2001;
30:454–457.
45. Sansone R. Patient to patient e-mail support for clinical practices. Eat Disord Treat
Prev 2001; 9:373–375.
46. Kenwright M, Liness S, Marks I. Reducing demands on clinicians by offering
computer-aided self-help for phobia/panic. Feasibility study. Br J Psychiatry 2001;
179:456–459.
47. Thiels C, Schmidt U, Treasure J, Garthe R, Troop N. Guided self-change for bulimia
nervosa incorporating use of a self-care manual. Am J Psychiatry 1998; 155:
947–953.
48. Smyth J, Wonderlich S, Crosby R, Miltenberger R, Mitchell J, Rorty M. The use of
ecological momentary assessment approaches in eating disorder research. Int J Eat
Disord 2001; 30:83–95.
49. Harari D, Furst M, Kiryati N, Caspi A, Davidson M. A computer-based method for

the assessment of body-image distortions in anorexia-nervosa patients. IEEE Trans
Inform Technol Biomed 2001; 5:311–319.
50. Leit RA, Gray JJ, Pope HG Jr. The media’s representation of the ideal male body:
a cause for muscle dysmorphia ? Int J Eat Disord 2002; 31:334–338.
51. Riva G, Bacchetta M, Baruffi M, Molinari E. Virtual-reality-based
multidimensional therapy for the treatment of body image disturbances in binge
eating disorders: a preliminary controlled study. IEEE Trans Inform Technol
Biomed. 2002; 6:224–234.
52. Hsiung R. Suggested principles of professional ethics for the online provision of
mental health services. Telemed J E-Health 2001; 7:39–45.
53. Simons v. Blue Cross and Blue Shield of Greater New York, 144 A.D.2d 28, 536
N.Y.S.2d431, 1989.
570 YAGER
Index
Acrocyanosis, 235 dopamine, 260

ACTH, 283, 285 elderly, 11
Activities therapy, 372 family characteristics, 428
Adolescent Dissociative Experiences Scale family studies, 168
(A-DES), 33 family therapy, 349, 355, 432
Adolescent Drinking Index (ADI), 32 gender, 78
Affective disorders, 187, 358, 511, 517 incidence, 75
Aggravated assault, 511, 513 infantile, 51
Alexithymia, 311, 528 inpatient treatment, 350, 493
Amenorrhea, 1, 247 interpersonal psychotherapy (IPT),
American Academy of Child and 467
Adolescent Psychiatry, 526 involuntary treatment, 372
American Psychiatric Association Practice males, 10
Guidelines for eating disorders, 19, mortality, 97
350, 443, 526 neurodevelopmental aspects, 312
Americanization, 429 norepinephrine, 258
Amitriptyline, 495 nutritional assessment, 379
Animal models, 300 nutritional rehabilitation, 379, 386
Anorexia nervosa partial hospitalization, 350, 368
age, 78 peptide YY prevalence, 75
amenorrhea, 1 prognostic factors, 98
beta-endorphin, 285 psychopharmacology, 491
cognitive-behavioral therapy (CBT), recovery, 97
355, 361, 386, 406 relapse, 98
comorbidity, 184, 513 risk factors, 122, 124
course, 97 Russell’s criteria in the young, 11
course comparison with bulimia [Anorexia nervosa]
nervosa, 103 Russell’s description, 1
cross-over, 98 serotonin, 261, 326
day treatment, 350, 368 time trends, 78
[Anorexia nervosa] treatment utilization, 99, 220
diagnosis, xxiv, 19 twinstudies, 169, 173
571
572 INDEX
vasopressin, 238, 283, 286 family therapy, 440

Anorexia Nervosa Stage of Change interpersonal psychotherapy (IPT),
Questionnaire, 26 395, 466
Anorexigenic circuit, 302 males, 10
Antianxiety agents, 349, 352 mortality, 103
Anticonvulsants, 498 nutritional assessment, 394
Antidepressant agents, 349, 367, 492, nutrition intervention and education,
497, 501 395
Antiglucocorticoids, 556 prevalence, 513
Antihistamines, 495 recovery, 103
Antipsychotic agents, 258, 261, 349, relapse, 103
352, 491, 494, 556 psychopharmacology, 501
Anxiety, 31, 59, risk factors, 149
Anxiety disorders, 190, 511, 518 treatment utilization, 103
Appetite control, 302 twin studies, 173
Assessment, 15, 18, 59, 234 victimization, 513, 516
Atherosclerosis, 235 Binge Eating Scale, 21
Attention deficit hyperavtivity BMI, 2, 61, 288
disorder (ADHD), 197, 521 Body image assessment, 562
Autoantibodies, 261 Body image challenge, 307
Body perception group, 365, 371
Barrett’s esophagus, 243 Body Rating Scale for Adolescents, 515
Beck Anxiety Inventory (BAI), 31 Bombesin (BBS), 289
Beck Depression Inventory (BDI-II), Bone mineral density, 246
30 Borderline personality disorder, 212, 217,
Behavioral family systems therapy, 473, 487, 511, 523
436 Bradycardia, 245
Benzodiazepines, 533 Brain development, 298
Berkson’s bias, 183, 187 Brain function, 299
Beta-endorphin, 284 Brain imaging receptor binding, 258, 262,
Bethanocol, 495 264
Bicarbonate, 238 Brain structure
Binge eating, 4, 19, 74 anorexia nervosa, 298
Binge eating disorder bulimia nervosa, 299
cognitive behavioral therapy (CBT), Brief Symptom Inventory (BSI), 33
395, 406, 440 Bulimia nervosa
comorbidity, 186 age, 83
course, 103 attachment, 438
cross-over, 103 [Bulimia nervosa]
[Binge eating disorder] beta-endorphin, 245, 285
diagnosis, 7 cholecystokinin, 242
dialectical behavior therapy (DBT), cognitive-behavioral therapy (CBT),
488 349, 361, 405, 410
family characteristics, 430 combination therapy, 499
INDEX 573
comorbidity, 185, 186, 512 Carnosinase activity, 245

course, 100, 522 Catachol O-methyltransferase, 338
course comparison with anorexia CD-ROM-based programs, 559, 562
nervosa, 103 Central nervous system complications,
course comparison with binge eating 244
disorder, 108 Cerebral atrophy, 244
cross-over, 101 Chest pain, 235
diagnosis, 3 Children’s Eating Attitudes Test
dialectical behavior therapy (DBT), (ChEAT), 22
488 Childabuse, 214, 512
dopamine, 260 Child Sexual and Physical Abuse:
family characteristics, 429 Guidelines for Treatment, 526
family studies, 168 Chinese anorexia, 12
family therapy, 437 m-Chlorophenylpiperazine (m-CPP), 258,
galanin 263
gender, 83 Chlorpromazine, 494
incidence, 82 Cholecystokinin, 242, 284, 287
inpatient treatment, 351 Choking, 10, 51
interpersonal psychotherapy, 349, Cimetidine, 243
465 Cisapride, 495
isatin Civil commitment, 372
leptin Classification, xxiv, 71
loss, 438 Clinical diagnostic interview, 9, 15
males, 10 Clinician Administered PTSD scale
mortality, 100 (CAPS), 30
norepinephrine, 258 Cluster A Personality disorders, 211, 522,
nutritional assessment, 389 551
nutritional rehabilitation, 389 Cluster B Personality disorders, 212, 217,
prevalence, 80 511, 522, 551
prognostic factors, 101 Cluster C Personality disorders, 211, 522,
psychopharmacology, 496 551
risk factors, 123, 137 Cognitive behavioral therapy, 349, 355,
serotonin, 261 361, 366, 370, 391, 395, 403, 440,
recovery, 100 475, 499, 528, 560
relapse, 101 Cognitive restructuring, 414
time trends, 83 Comorbidity, 182, 358, 510
treatment utilization, 103, 220 axis I, 182, 358, 517
twinstudies, 171 axis II, 211, 218, 510
Bullying, 516 axis III, 231, 510, 511
Bupropion, 497 Comparison of eating disorders, 103
Buspirone, 265 Computed tomography, 296
Computers, 557
Carbamazepine, 498 Compulsivity, 215, 220, 265, 269
Cardiac arrhythmias, 234, 235
574 INDEX
Concerns About Change Questionnaire Diffusion tensor magnetic resonance

(CAC), 28 imaging (DT-MRI), 555,
Constipation, 243 Discharge process, 367
Control, 56, 516 Disordered eating, 72
Corticotropin-releasing factor antagonists, Disruptive behavior disorders, 197, 521
556 Dissociative disorders, 197, 519
Corticotropin-releasing hormone (CRH), Dissociative Experiences Scale (DES), 32,
283, 285 515
Cortisol, 239, 244, 247, 264, 266, 283, Dissociative symptoms, 513, 556
313 Distance medicine, 559
Course of illness Distress tolerance skills, 484
anorexia nervosa, 97 Diuretics, 74
binge eating disorder, 103 Domperidone, 495
bulimia nervosa, 100 Dopamine, 257, 260, 301
Countertransference, 526 Drug Abuse Screening Test (DAST), 32
Cues and consequences, 413 DSM-III, 1, 89, 550
Cultural effects on diagnosis, 11 DSM-III-R, 2
Culture-independent diagnosis, 12 DSM-IV, 2, 321, 440, 549
Cyproheptadine, 495 DSM-V, 549
Dual-energy X-ray absorptiometry
Delayed gastric emptying, 241 (DEXA), 247
Demographics, 85 Dysdiadokinesis, 312
Dental complications, 241 Dyslipidemia, 234
Depression, 30, 245 Dysregulation hypothesis, 266
Depression Anxiety and Stress Scales
(DASS), 30 Eating Attitudes Test (EAT and EAT-26),
Desipramine, 500 22, 358
Development, 61 Eating Disorders Examination (EDE), 20
Developmental characteristics, 213 Questionnaire (EDE-Q), 24
Developmental disorders, 47, 55 Eating Disorders Inventory-2 (EDI-2), 22,
Dexamethasone suppression test, 239, 358, 498
283 Eating disorder not otherwise specified, 7,
Diabetes mellitus, 239 83, 89
Diagnosis, xxiv Elderly, 11
Diagnostic Interview for Children and Electrolyte abnormalities, 238, 245
Adolescents (DICA), 29 Elite athletes, 87
Diagnostic issues, xxiv, 233 E-mail, 561
Dialectical behavior therapy (DBT), 473, Emetine, 237
528 Emotional abuse, 511, 516
Diazepam, 496 Emotion regulation skills, 484
Dieting, 72 Endocrine abnormalities, 239
Diet pills, 74 Enteral feeding, 55
Differential diagnosis, 9 Epidemiology, 69, 552
Esophageal problems, 242
INDEX 575
Estrogen, 239, 248, 496 Frisch tables, 352

Estrogen receptors, 334 Functional imaging, 244, 297, 305
E-therapy, 560 Future directions, 547
Euthyroid sick syndrome, 283
Experiential psychotherapies, 528 Gastrin-releasing peptide (GRP), 289
Expert Consensus Guidelines on the Gastroesophageal reflux, 48, 243
Treatment of PTSD, 526, 533 Gastrointestinal complications, 241, 495
Eye movement desensitization Gene polymorphisms, 267, 326
reprocessing (EMDR), 524, 529 Genetic methods, 324
Genetics, 13, 135, 147, 150, 164, 217,
Failure to thrive, 51, 54 312, 321, 553, 555
Family Adaptation and Cohesion GERD, 243
Evaluation Scale, 432 Ghrelin, 243, 288
Family Assessment Measure, 432 Glycosylated hemoglobin (HbAlc), 239
Family Assessment Measure (FAM-III), 35 Goal-weight range, 354
Family characteristics, 213 Gonadotropin-releasing hormone (GRH),
Family Environment Scale (FES), 35, 432 239
Family evaluation, 431 Gout, 238
Family functioning, 34, 61, Gray matter, 244
Family involvement, 387 Grief, 459
Family obstacles, 444 Growth charts, 354
Family therapy, 365, 426, 528 Growth hormone, 283
Feeding disorders of infancy and early Group therapy, 362, 370
childhood, 45, 312 Gynecological complications, 248
assessment, 59
clinical presentation, 51 Harm avoidance, 218, 269
criteria, 49 Homicide, 511
management, 63 Homovanillic acid (HVA), 258, 260, 268
outcome, 64 HSKCa3, 339
Fenfluramine or d-fenfluramine or Hunger and satiety, 243, 302
dlfenfluramine, 258, 263, 499, 502 Hydroxybutyric acid, 281
Fluoxetine, 492, 500, 503 5-hydroxyindoleacetic acid (5-HIAA),
Fluvoxamine, 502 258, 262, 268, 493
f-MRI, 306, 310, 314, 555 5-hydroxytryptophan, 258, 264
Follicle-stimulating hormone (FSH), 239, Hyperactivity, 285
248 Hypercortisolism, 239
Food avoidance emotional disorder, 59 Hypertension, 235
Food challenges, 305 Hypocalcemia, 238
Food deprivation, 516 Hypoglycemia, 245
Food intake, 300 Hypokalemia, 234, 237
Food record, 388, 390, 396 Hypomagnesemia, 238
Food refusal, 52 Hypophosphatemia, 238
Forgetting, 513 Hypotension, 245
Free fatty acids, 281 Hypothalamic hypogonadism, 285
576 INDEX
Hypothalamic-pituitary-adrenal (HPA) Lithium, 496, 499

axis, 239, 283, 285 LOD score, 326
Hypothalamic-pituitary-gonadotropin Long-term outcome, 96
(HPG) axis, 239, 281, 285 Lutenizing hormone (LH), 285
Hypothermia, 235, 245
Maudsley approach or model, 366, 435,
ICD-10, 549 439
Imipramine, 502 McKnight Risk Factor Survey, 515
Impulse control disorders, 195, 520 Meal plan, 382, 392, 395, 397
Impulsivity, 210, 215, 220, 265, 269 Medical comorbidity, 231
Industrialization, 85 Medications, 367
Infantile anorexia nervosa, 51 Melanocortinergic system genes, 338
Information technology, 557 Melatonin, 241
Inosotol, 556 Metabolic acidosis, 237
Insulin, 283, 302 Metabolic alkalosis, 238
Insulinlike growth factor-1 (IGF-1), 247, 3-Methoxy-4-hydroxyphenylglycol
284 (MHPG), 258, 268
Interpersonal deficits, 458 Metoclopramide, 242, 495
Interpersonal effectiveness skills, 484 Metropolitan Life Tables, 352
Interpersonal formulation, 457 Michigan Alcoholism Screening Test
Interpersonal functioning, 34, 451 (MAST), 32
Interpersonal inventory, 457 Mifepristone, 556
Interpersonal psychotherapy (IPT), 449, Migraine, 266
528 Milan systems model, 434
Interpersonal role disputes, 458 Mitral valve prolapse, 235
Inventory of Interpersonal Problems (IIP), Mindfulness, 484
34 Minnesota semi-starvation experiments,
Involuntary patients, 372 189, 511, 527
Ipecac, 237 Mizes Anorectic Cognitions Scale (MACS-
Isatin, 268 R), 24
Isoproterenol, 258, 260 Moclobemide, 556
Molecular biology, 321, 553
Kidney failure, 237 Molestation, 511, 513
Kids’ Eating Disorders Survey (KEDS), Monoamine oxidase, 215, 268
23, 73, 515 inhibitors, 497, 555
Monoamines, 257
Laboratory tests, 234 Mood disorders, 187, 358, 511, 517
Lansoprazole, 243 Mortality rate
Laxatives, 74, 243, 392 anorexia nervosa, 97, 231, 235
Leptin, 241, 247, 287, 301 binge eating disorder, 103, 233
Leukocytes, 258, 260, 263 bulimia nervosa, 100, 233
Leuven Family Questionnaire, 432 Morton, 207
Light therapy, 556 Motility agents, 495
Linkage analysis, 339 Motivation to change, 19, 25, 63
INDEX 577
Motivational Enhancement Therapy Occupational therapy, 371

(MET), 528 Olanzapine, 494
Multi-impulsivity, 515, 523 Omeprazole, 243
Muscle weakness, 245 Ondansetron, 268, 499
Musculoskeletal complications, 245 Opiate antagonists, 495
Myopathy, 245 Opioid peptides, 285
Oral complications, 241
Naloxone, 495 Oral contraceptives, 1
Naltrexone, 495, 499 Orexigenic circuit, 302
Nasogastric feeding, 555 Orthostatic hypotension, 235
National Comorbidity Study, 511 Osteopenia, 234, 246, 496
National Women’s Study, 512, 517, 518, Osteoporosis, 239, 246, 496
519, Oxytocin, 286
Neglect, 54, 511, 516
Nephrolithiasis, 238 Pain thresholds, 245
Neurobiology, 215 Palpitations, 235
Neurocognitive abnormalities, 244 PANDAS, 313
Neurodevelopmental aspects, 312, 314 Panic disorder, 518
Neuroendocrinology, 281 Parasuicidal behavior, 473
Neuroimaging, 296, 554 Parotid gland enlargement, 242
Neurokinins, 556 Paroxetine binding, 215
Neurological soft signs, 312 Pediatric eating disorders, 11
Neuropeptides, 284 Peptide YY (PYY), 286
Neuropeptide Y (NPY), 284, 286 Perfectionism, 133, 145, 210, 213, 516
Neurotransmission, 307 Peripheral nervous system complications,
Neurotransmitters, 256 245
precursors, 258, 261 Peripheral neuropathies, 245
Night eating syndrome, 233 Perseverative feeding disorder, 52
Nocturnal sleep-related eating disorder, Personal digital assistants (PDA’s), 555,
233 559, 563
Norepinephrine, 257 Personal health buddy, 562
transporter, 339 Personality, 34, 207
Nuclear magnetic resonance, 555 disorders, 34, 211, 218, 511
Nutritional assessment, 379, 389 traits, 209
Nutritional rehabilitation, 352, 380, 389 PET scans, 262, 264, 302
Nutrition counseling, 377 Phenytoin, 498
Nutrition education, 391 Pica, 47
Pickyeaters, 52, 54, 312
Obesity, 283, 312, 340, 551 Pimozide, 494
Obsessive-compulsive disorder (OCD), Platelet 4, 217
191, 492, 518 Platelets
Obsessive-compulsive symptoms, 29, 367 5-HT content, 258, 265
Obstetrical complications, 248 5-HT uptake, 258, 263
Occupation, 87 imipramine-binding, 258, 263
578 INDEX
MAO content, 258, 263, 268 Reglan, 367

paroxetine-binding, 258, 264, 266 Regurgitation, 47
receptor-mediated aggregation, 258, Relapse, 368
266 Relapse prevention, 411, 422, 493, 497,
Polycystic ovaries, 248 503
Positron emission topography (PET), 244, Renal complications, 237
300 Repetition compulsion, 521
Post-traumatic feeding disorder (PTFD), Restraint Scale, 21
51 Revised Bulimia Test (BULIT-R), 23
Post-traumatic stress disorder (PTSD), Rhabdomyolysis, 238
30, 33, 192, 510, 524, 529 Riskfactors, 116
Pregabalin, 556 anorexia nervosa, 124
Pregnancy, 136, 148, 248 basic concepts, 117
Preoccupations, 21 binge eating disorder, 149
Prilosec, 367 bulimia nervosa, 137
Progestin, 496 definition of, 117
Progressive muscle relaxation, 529 typology, 119
Prolactin, 215, 263, 283 Rituals, 21
Prolonged exposure, 529 Role transitions, 459
Prolonged QT interval, 234, 235, 352, Rosenberg Self-Esteem Scale (RSES), 31
381 Rumination, 47
Psychoeducation, 362
Psychogenic amnesia, 513 S-adenosylmethionine, 556
Psychogenic vomiting, 59 Schizophrenia, 196
Psychometrics, 18 Scoliosis, 246
Problem-solving, 417 Seasonal affective disorder, 188
Program scheduling, 356 Seizures, 245
Proton pump inhibitors, 243 Selective eating, 52
Psychopharmacology, 490 Self-esteem, 31, 516
Psychotherapies, 556 Self-help, 558
Psychotic disorders, 196 Self-injurious behaviors, 473, 515, 529
PTSD (See Post-traumatic stress disorder) Self-report instruments, 21, 30
PTSD Symptom Scale-Self-Report (PSS), Serotonin, 135, 147, 215, 257, 307, 326,
33 493
Punishment, 515 receptors, 263, 266, 310, 330
Pyramid model, 441 Sertraline, 502
Sexualabuse, 122, 127, 140
Rantidine, 243 Sexual activity, 285
Rape, 511, 513, 517 Shape- and Weight-Based Self-Esteem
Readmission, 368 Inventory (SAWBS), 25
Reboxetine, 556 Shame, 516
Refeeding syndrome, 237 Short stature, 246
Regional cerebral blood flow (rCBF), Short tandem repeats, 325
300, 305 Social class, 85
INDEX 579
Social phobia, 191, 518 Tricuspid valve prolapse, 235

Social skills training, 417 Tricyclic antidepressants, 492, 500
Somatization, 515 Tryptophan or L-tryptophan, 258, 261,
Somatoform disorders, 59, 197, 520 266, 498
Somnambulism, 233 Tryptophan depletion, 258, 261
SPECT, 300, 305, 310 Tryptophan hydroxylase, 267
SSRI’s, 492 L-TRP/LNAA ratio, 261
Stallers, 53
State v. Trait, 212 Uncoupling protein, 338
Structural abnormalities in the brain, 244, University of Rhode Island Change
296 Assessment Scale (URICA), 26
Structural family therapy, 428, 432, 439, Urbanization, 86
442
Structured interviews, 20, 28 Vagus nerve, 245, 499, 556
Structured Clinical Interview for DSMIV Vasopressin, 238, 283, 286
(SCID-IV), 28 Vegetarianism, 354
Substance abuse, 32, 193, 487, 511, 519 Ventricular enlargement, 244
Substance P antagonists, 556 Victimization, 510
Sulcal enlargement, 244 Video, 62
Sulpiride, 494 Vomiting, 48, 74
Symptom Checklist-90-R (SCL-90-R), 33
Systems family therapy, 428, 434, 439, Water intoxication, 245
442 Weight restoration, 351, 381
Systems of care, 563 White matter, 244
Talipexole, 556 Yale-Brown-Cornell Eating Disorder Scale

Telephone contact, 485 (YBC-EDS), 21
Tendon reflexes, 245 Yale-Brown Obsessive Compulsive Scale
Tetany, 238 (Y-BOCS), 29
Tetrahydrocannabinol (THC), 496
Three Factor Eating Questionnaire, 21 Zantac, 367
Thyroid, 283 Zen Buddhism, 473
Tianeptine, 556 Zinc deficiency, 496
Toddler refusal, 52 Ziprazodone, 556
Topiramate, 498, 502
Torsade de pointes, 235
Tramadol, 556
Transcranial magnetic stimulation, 556
Transcultural diagnosis, 11
Transmission disequilibrium test (TDT),
325, 331, 342
Trauma, 510
Trauma-related disorders, 511, 517
Tribulin, 268

Brewerton. Clinical Handbook of Eating Disorders. An Integrated Approach PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Brewerton. Clinical Handbook of Eating Disorders. An Integrated Approach PDF

Uploaded by

Copyright:

Available Formats

Clinical Handbook of Eating Disorders

Bennett Leventhal, M.D.

Siegfried Kasper, M.D.

Mark H.Rapaport, M.D.

1. Handbook of Depression and Anxiety: A Biological Approach, edited by Johan

24. Autism Spectrum Disorders, edited by Eric Hollander

5. Clinical Management of Anxiety, edited by Johan A.den Boer

MARCEL DEKKER, INC.

ISBN 0-203-02176-2 Master e-book ISBN

ISBN: 0-8247-4867-0 (Print Edition)

disorders, the authors rightly confine their appraisal to combinations of

Finally, a neat example of integration is provided in the very first chapter,

In Chapter 2, Drs. Jacqueline Carter, Traci McFarlane, and Marion Olmsted

eating disorders, which underlie many of their clinical manifestations, including

Foreword Gerald Russell viii

PART I: DIAGNOSIS, EPIDEMIOLOGY, AND COURSE

PART II: RISK FACTORS, ETIOLOGY, AND COMORBIDITY

7 Role of Genetics in Anorexia Nervosa, Bulimia 165

PART III: PSYCHOBIOLOGY

PART IV: TREATMENT

17 An Overview of Cognitive-Behavioral Approaches to 405

Arnold E.Andersen University of Iowa Hospital & Clinics, Iowa City,

Dorothy Grice* Center for Neurobiology and Behavior, University of

*Current affiliation: Child and Adolescent Psychiatry, Department of Psychiatry,

Lisette Morris Department of Clinical Psychology and Psychotherapy,

Susan G.Willard Department of Psychiatry & Neurology, Pediatrics,

TABLE 1 DSM-III Diagnostic Criteria for Anorexia Nervosa

Source: American Psychiatric Association, 1980.

TABLE 2 DSM-IHR Diagnostic Criteria for 307.10 Anorexia Nervosa

Source: American Psychiatric Association, 1987.

examples of how one’s body weight or shape is cognitively experienced in a

TABLE 3 DSM-IV Diagnostic Criteria for 307.1 Anorexia Nervosa

Source: America Psychiatric Association, 1994.

TABLE 4 DSM-III Diagnostic Criteria for Bulimia

Source: American Psychiatric Association, 1980.

TABLE 5 DSM-IIIR Diagnostic Criteria for 307.51 Bulimia Nervosa

Source: American Psychiatric Association, 1987.

In addition, the frequency with which binge episodes and inappropriate

TABLE 6 DSM-IV Diagnostic Criteria for 307.51 Bulimia Nervosa

Source: American Psychiatric Association, 1994.

In the DSM-IV, the diagnosis of bulimia nervosa becomes more specific, as

Binge Eating Disorder

TABLE 7 DSM-IV Research Criteria for Binge Eating Disorder

Source: American Psychiatric Association, 1994.

TABLE 8 DSM-IV 307.50 Eating Disorder Not Otherwise Specified

Source: American Psychiatric Association, 1994.

Unlike those with bulimia nervosa or anorexia nervosa-B/P, individuals

CLINICAL IDENTIFICATION OF EATING

COMMON DIFFERENTIAL DIAGNOSES IN EATING

The Very Young

TABLE 9 Russell’s Criteria for Anorexia Nervosa in the Young

The Very Old

EFFECTS OF CULTURE ON THE DIAGNOSIS

From this viewpoint, the precise rationale or cognitive content would be

EFFECT OF GENETIC CAUSALITY ON DIAGNOSIS

TABLE 10 Proposed Culture-lndependent Diagnostic Criteria for Anorexia Nervosa

1. Introduction and overview

Comprehensive psychometric assessment is essential for diagnosis, case

difficulties. The interested reader is referred to Groth-Marnat (5) for more

Structured Interview for Anorexia and Bulimia (10). However, these

Eating Disorder Examination (EDE; 11)

Yale-Brown-Cornell Eating Disorder Scale (YBC-EDS; 12)

rituals. Preoccupations are defined as “ideas, thoughts, images or impulses that