Robert L. Woolfolk, Lesley a. Allen-Treating Somatization_ a Cognitive-Behavioral Approach (2006)

TREATING SOMATIZATION

TREATINGSOMATIZATION

A Cognitive-Behavioral Approach

Robert L. WoolfolkLesley A. Allen

THE GUILFORD PRESSNew York London

2007 The Guilford PressA Division of Guilford Publications, Inc.72 Spring Street, New York, NY 10012www.guilford.com

All rights reserved

Except as indicated, no part of this book may be reproduced, translated, storedin a retrieval system, or transmitted, in any form or by any means, electronic,mechanical, photocopying, microfilming, recording, or otherwise, withoutwritten permission from the Publisher.

Printed in the United States of America

This book is printed on acid-free paper.

Last digit is print number: 9 8 7 6 5 4 3 2 1

LIMITED PHOTOCOPY LICENSE

These materials are intended for use only by qualified mental health profes-sionals.

The Publisher grants to individual purchasers of this book nonassignablepermission to reproduce all materials for which photocopying permission isspecifically granted in a footnote. This license is limited to you, the individ-ual purchaser, for use with your own clients and patients. It does not extendto additional clinicians or practice settings, nor does purchase by an institu-tion constitute a site license. This license does not grant the right to reproducethese materials for resale, redistribution, or any other purposes (includingbut not limited to books, pamphlets, articles, video- or audiotapes, and hand-outs or slides for lectures or workshops). Permission to reproduce thesematerials for these and any other purposes must be obtained in writing fromthe Permissions Department of Guilford Publications.

Library of Congress Cataloging-in-Publication Data

Woolfolk, Robert L.Treating somatization : a cognitive-behavioral approach / by Robert L.

Woolfolk and Lesley A. Allen.p. ; cm.

Includes bibliographical references and index.ISBN-10: 1-59385-350-5 ISBN-13: 978-1-59385-350-1 (hardcover : alk. paper)1. Somatization disorder. 2. Medicine, Psychosomatic. 3. Cognitive

therapy. I. Allen, Lesley A. II. Title.[DNLM: 1. Somatoform Disorderstherapy. 2. Cognitive Therapy

methods. WM 170 W913t 2006]RC552.S66W662 2006616.8524dc22

2006012304

To Katie

About the Authors

Robert L. Woolfolk, PhD, is Professor of Psychology and Philosophyat Rutgers University and Visiting Professor of Psychology at Prince-ton University. He has published many papers and several books onpsychotherapy and psychopathology, including Stress, Sanity, and Sur-vival (1978, Monarch) and The Cure of Souls: Science, Values, and Psycho-therapy (1998, Jossey-Bass). A practicing clinician for more than 30years, Dr. Woolfolk has sought in both his work with patients and hisscholarly endeavors to integrate the scientific and humanistic tradi-tions of psychotherapy.

Lesley A. Allen, PhD, is a clinical psychologist and Associate Profes-sor of Psychiatry at Robert Wood Johnson Medical School at the Uni-versity of Medicine and Dentistry of New Jersey. She has authoredor coauthored numerous journal articles and book chapters on somato-form disorders. Dr. Allen is actively involved in clinical work, teach-ing, and supervising. She is currently Principal Investigator on anongoing program of clinical research funded by the National Instituteof Mental Health.

vii

Acknowledgments

Writing any book entails a variety of struggles that are successfulonly with the assistance of others. We owe thanks to several peo-ple. Mike Gara provided theoretical and methodological guidance,sound practical judgment, and occasional flashes of brilliance. PaulLehrer was there at the beginning with his unsurpassed scholarshipand estimable clinical wisdom. Javier Escobars enlightened adminis-trative leadership and impeccable intellectual sensibilities were indis-pensable to the project. Jim Nageotte of The Guilford Press providedthe optimal combination of encouragement and oversight. Thanks alsoto Seymour Weingarten for being everything a publisher should be.

We owe our deepest gratitude, however, to the many patients andtrainees who have labored along with us to alleviate the form of suffer-ing that this book addresses.

The writing of this book and the research described herein werepartially supported by grants from the National Institute of MentalHealth (Grant Nos. K08 MH01662, R21 MH066831, R01 MH60265, andP20 MH074634).

ix

Contents

Chapter 1. Introduction 1

Chapter 2. Somatization: Epidemiology, Clinical Characteristics,and Treatment

14

Chapter 3. Affective Cognitive-Behavioral Therapyfor Somatization: Rationale and Overview

46

Chapter 4. Assessment 66

Chapter 5. The Context of the Therapy 85

Chapter 6. Behavioral Interventions 95

Chapter 7. Working with Cognitions and Emotions 110

Chapter 8. Interpersonal Methods 130

Chapter 9. Assorted Clinical Topics 140

Overview of Appendices 151Appendix A. Ten-Session Treatment Manual 153

Appendix B. Clinical Trial Assessing the Efficacy of AffectiveCognitive-Behavioral Therapy

165

Appendix C. Severity of Somatic Symptoms Scale 173

Appendix D. Somatic Symptom Questionnaire 179

Appendix E. Instructions for Abbreviated ProgressiveMuscle Relaxation

187

Appendix F. Examining Thoughts 190

References 191

Index 219

xi

TREATING SOMATIZATION

C H A P T E R 1

Introduction

Somatization is among the most puzzling phenomena that health-care workers encounter. In somatization physical symptoms occur inthe absence of any identifiable causal mechanism. The causes ofsomatization that we are able to implicate are neither proximate norsomatic, seeming instead to be indirect and to reside in the patientsmind or culture. Somatization appears to be universal. We find it in allpresent societies and in all past societies for which we have relevantrecords. Forty centuries ago, the physicians of Egypt were familiarwith somatization; some years later so were those of ancient Greece.

For the contemporary clinician, the patient who somatizes is apressing practical problem. Here there is distress, dysfunction, and dis-ability of great magnitude and intransigence. Patients diagnosed withthe most severe form of somatization, somatization disorder, havebeen shown to incur healthcare expenses that are nine times theU.S. average and consume disproportionate amounts of the time andenergy of healthcare providers (Smith, Monson, & Ray, 1986a). In addi-tion to the extensive direct costs, somatization disorder creates enor-mous indirect costs to the economy in the form of lost work productiv-ity. Individuals diagnosed with somatization disorder report beingbedridden for 27 days per month (Katon et al., 1991; Smith et al.,1986a). Somatization disorder is not only costly, but also difficult totreat successfully. In a longitudinal study following patients withsomatization disorder who were receiving standard medical care,only 31% recovered after 15 years (Coryell & Norten, 1981). Typically,patients with somatization disorder are dissatisfied with the medicalservices they receive and repeatedly change physicians (Lin et al.,

1

1991). These treatment-resistant patients frustrate healthcare provid-ers with their frequent complaints and dissatisfaction with treatment(Lin et al., 1991). No controlled medication trial for somatization dis-order has been published. Anecdotal evidence suggests that manypatients diagnosed with somatization disorder refuse to take medica-tion and that those who do frequently report adverse medication sideeffects (Murphy, 1982). The story is much the same with other poly-symptomatic somatoform disorders (Fallon, 2004). As of this writing,pharmacological treatment has had minimal success with somatiza-tion.

In this book we describe our efforts to alleviate the suffering ofpatients with somatization. Over the last decade, we have developed adedicated psychosocial treatment for somatization that draws uponvarious traditions in psychotherapy, especially cognitive-behavioraltherapy and emotion-focused experiential therapies, a treatment wecall affective cognitive-behavioral therapy (ACBT). The principal aimof this book is to describe that treatment and to provide the trainingmaterial necessary for its effective use. To frame our approach to thisproblem, we first provide a brief review of the history of somatizationand of psychosomatic medicine. We then discuss philosophical andsociocultural underpinnings of somatization and conclude with anoverview of theories of somatization.

BACKGROUND: HISTORICAL AND THEORETICAL

The history of somatization begins with hysteria. Hysteria was firstdescribed 4,000 years ago by the Egyptians. Typical cases involvedpain in the absence of any injury or pathology in the location of thepain. The Egyptian theory held that a wandering uterus moved aboutthe body and produced pain from various regions. Greek physiciansdescribed a similar set of psychosomatic symptoms and essentiallyretained the Egyptian theory. The Greeks gave us the word hysteria,from the Greek hystera, meaning womb. The Greco-Egyptian formula-tion reveals two noteworthy features: that the disorder was primarilyobserved in females and that there was something thought to be essen-tially female about the disorder. Although the diagnostic category sub-sumed more than somatization, the term hysteria continued to bewidely used to label somatization patients until 30 years ago.

Medieval and Renaissance medicine preserved the ancient formu-lation of hysteria as described by the ultimate authorities, Hippocratesand Galen, until the 17th and 18th centuries, when it was first linkedwith the nervous system and the emotions. At the beginning of the

2 TREATING SOMATIZATION

17th century, the French physician Charles Le Pois opposed the uterinetheory of hysteria (he believed the spleen to be the culprit) anddeclared that hysteria could occur in men; a few years later, ThomasSydenham declared that hysteria was the result of psychological andemotional causes and that in men hysteria was manifested as hypo-chondria (Boss, 1979). Foucault (1961/1965) states that by the end ofthe 18th century hysteria and hypochondria were beginning to beviewed as diseases of the nerves akin to such recognized mental disor-ders as melancholia. By the 18th century, some authorities, such asJoseph Raulin, began to question hysterias organic basis. Raulindescribed hysteria as a disease in which women invent, exaggerate,and repeat all the various absurdities of which a disordered imagina-tion is capable (quoted in Foucault, 1961/1965, pp. 137138). Beforethe 19th century, due to the heterogeneous nature of hysterical symp-toms and the hypothesized connection with the emotions, physicianshad begun to allege that these symptoms were feigned or imagined.The unsympathetic attitudes of contemporary healthcare workers to-ward somatizers and the tendency to regard them as malingerers canbe traced to this period in the history of medicine.

Paul Briquets (1859) seminal monograph, Trait Clinique et Thra-peutique de Lhystrie, was a landmark in the descriptive psychopathol-ogy of somatization. Our current conception of somatization disorderderives directly from this paper. Briquets meticulous and exhaus-tive listing of the symptomatology of hysteria remains unsurpassed.In fact, he described three related syndromes: conversion phenom-ena, hysterical personality, and multiple chronic unexplained somaticsymptoms (Dongier, 1983; Mai & Merskey, 1980), all overlapping insymptomatology somewhat and often observed to co-occur. Briquetsperspicuous work was revived by Purtell, Robins, and Cohen (1951)and developed further by members of the illustrious Washington Uni-versity Department of Psychiatry. Perley and Guze (1962) published alist of 57 symptoms commonly reported by women diagnosed withhysteria, symptoms that were clustered in 10 different areas. Theseinvestigators were the first to suggest specific criteria for the diagnosisof hysteria: the presence of 25 symptoms from at least 9 of the 10 symp-tom areas (Guze, 1967). Later, this list of 57 symptoms was expandedto 59 symptoms and the term Briquets syndrome was adopted(Guze, Woodruff, & Clayton, 1972). The criteria for Briquets syndromewere incorporated into the Feighner criteria (Feighner et al., 1972), theprecursor to the symptom set that appeared in the third edition of theDiagnostic and Statistical Manual of Mental Disorders (DSM-III; AmericanPsychiatric Association, 1980). In that volume the theoretically neutralterm somatization was preferred over the traditional terminology.

Introduction 3

Although some of the traditional language remains in the fourth edi-tion of the DSM (e.g., conversion disorder), the word hysteria nolonger appears (DSM-IV; American Psychiatric Association, 1994). Theninth edition of the World Health Organizations (1979) InternationalClassification of Diseases (ICD-9), a more cosmopolitan nosology of so-matic and mental disorders published a year earlier than DSM-III,retains much of the perennial terminology, including not only hysteriabut also neurasthenia.1 ICD-10 (World Health Organization, 1993)has shifted in the direction of the DSM, though without banishingevery bit of the classical vocabulary.

The history of somatization also is interconnected with two im-portant and historically related developments in the history of psychi-atry: (1) psychosomatic medicine and (2) psychoanalysis. Althougheventually absorbed by psychoanalysis and subsumed within a psy-choanalytic theoretical framework, psychosomatic medicine had anestablished history before Freud. From antiquity, the interaction be-tween mind and body and its effects upon health had been alluded toby many writers. An early systematic account was William Falconers(1788) A Dissertation on the Influence of the Passions upon the Disorders ofthe Body. The term psychosomatic was used first by Heinroth in 1818as describing the interplay between mind and body in health and dis-ease (West, 1982, p. xvi). By the end of the 19th century nervous con-ditions, including psychosomatic ailments such as neurasthenia, andthe nerve doctors who treated them had proliferated, so much sothat during the Victorian era bad nerves was thought of as some-thing of an epidemic (Shorter, 1997).

It was about this time that Sigmund Freud entered upon the scene.As a young man Freud spent the winter of 188586 as a student ofJean-Martin Charcot at the Salptrire hospital in Paris. There he ob-served the worlds leading authority, Charcot, use hypnosis to removehysterical symptoms. Upon his return to Vienna, Freud began a closecollaboration with Joseph Breuer. The product of this collaborationwas the book Studies in Hysteria (1895/1974), in which Breuer andFreud developed the concept of conversion, a process whereby intra-psychic activity putatively brings about somatic symptoms. AlthoughFreud was later to break with Breuer and go on to create the substan-


1Neurasthenia is defined in ICD-10 as persistent and distressing feelings of exhaus-tion after minor mental or physical effort, accompanied by one or more of the fol-lowing symptoms: muscular aches or pains, dizziness, tension headache, sleep dis-turbance, inability to relax, and irritability. The term was coined in 1856 by RobertMayne and popularized by the American neurologist George Beard during the sec-ond half of the 19th century (Gijswijt-Hofstra & Porter, 2001).

tial edifice of psychoanalysis, his work on hysteria was a blueprint forand harbinger of later theoretical efforts. Here the ideas of early emo-tional trauma or intrapsychic conflict as the cause of physical symp-toms began to take shape. This work also introduced the notion of aphysical symptom as an unconscious form of communication, a devicefor securing secondary gain, or a means for avoiding emotional pain.

The notion of the transduction of psychological conflict into bod-ily symptoms was widely disseminated as psychoanalysis began todominate psychiatry. Stekel (1924) coined the term somatization(somatisieren) during the early 1920s and defined it as the conversionof emotional states into physical symptoms (p. 341). That is, Stekelregarded somatization as equivalent to the mechanism of conversionthat Breuer and Freud had used to explain the development of sensoryor voluntary motor symptoms in hysteria. A strident and eccentric pro-ponent of the mindbody interaction was Georg Groddeck (1977), whobelieved that psychic processes are etiological factors in all diseases(Avila & Winston, 2003). Groddeck contended that the symptoms ofany somatic disease might be interpreted as symbolic expressions ofunconscious motives and caused via the same mechanisms believed tounderlie hysteria.

The father of modern American psychosomatic medicine, FranzAlexander, attempted to minimize the excesses of indiscriminate psy-choanalytic approaches such as Groddecks. Alexander (1950) tookgreat pains to distinguish between two types of psychosomatic symp-toms: (1) those cases in which psychological conflict was convertedand communicated symbolically through physical symptoms, and (2)those cases in which the somatic symptoms resulted from the directand indirect physiological effects of emotional arousal. This secondkind of psychosomatic mechanism required few, if any, psychoanalyticassumptions and was quite compatible with mainstream scientificresearch, especially the work of Cannon, Seyle, and others on psycho-social stress. As psychoanalysis declined in influence, psychosomaticmedicine declined also. Today, the term psychosomatic, which wasfaddish in the 1950s, is no longer in vogue. Many of the problems oncetreated within the context of psychosomatic medicine now fall underthe purview of what is, in some sense, its successor discipline: behav-ioral medicine.

SOCIOCULTURAL AND PHILOSOPHICAL ISSUES

The biopsychosocial concept of illness, proposed by George Engel(1977) and to which we subscribe, suggests that illness is a complex

Introduction 5

entity involving the interplay of physical, psychological, and culturalfactors. In particular, many illnesses cannot be adequately compre-hended without taking into account the social contexts in which theydevelop and are manifested, diagnosed, and treated. What phenomenasocieties come to label illness or how human suffering is expressed andpresented to healers are complicated matters that can be conceptual-ized at several different levels, levels that involve variables that inter-act causally. The advent of such disciplines as psychoneuroimmun-ology and behavioral medicine has brought us evidence, in manydomains of medicine, of the close connections and complex concurrentinteractions among mental, behavioral, and somatic variables. Whenwe examine many somatic illnesses (e.g., hypertension) from the vari-ous standpoints of etiology, symptomatology, and treatment, theyemerge as complex entities with multifaceted interacting components,with biological, psychological, and social causes (Baum & Posluszny,1999; Cohen & Herbert, 1996). In the case of mental disorders or psy-chiatric syndromes the situation is even more complex and the levelsof explanation more deeply intertwined.

The historical record shows that Western categories of psycho-pathology have been influenced strongly by sociocultural factors andthat what gets labeled a mental illness is, to some degree, a reflection ofcultural values. Drapetomania, the desire of slaves to escape captivity,was in the early 19th century considered a mental illness (Cartwright,1851/1981; Szasz, 1987). Victorian physicians regularly performedtherapeutic clitorectomies on masturbators, who were thought to bementally ill. As recently as 1938, listed among the 40 psychiatric dis-orders in a leading textbook (Rosanoff, 1938) were moral deficiency,masturbation, misanthropy, and vagabondage. Homosexuality, whichWestern psychiatry regarded as a manifestation of mental illness, wasofficially depathologized, after a contentious political struggle, by areferendum of the American Psychiatric Association membership in1974 (Kutchins & Kirk, 1997). Not so long ago, psychiatrists in the for-mer Soviet Union performed the Orwellian maneuver of medicalizingopposition to the state when they employed the diagnosis of sluggishschizophrenia to effect the incarceration of many political dissidents(Bloch & Reddaway, 1977). Symptoms that indicate pathology in onesociety (e.g., regularly hearing the voice of a dead relative) are normaland customary in others.

Cultural variation in psychopathology results not only from dif-ferences in how psychiatric labels are applied, but from the fact thatdifferent societies seem to produce different forms of psychopathology.Many specific syndromes are unique to particular cultural contexts,


such as ataque de nervios, koro, or taijin kyofusho.2 The epidemics ofanorexia nervosa and bulimia in the contemporary West are unprece-dented but are spreading to middle and upper classes around theworld along with Westernization and its current aesthetic ideal of aslender female body (Ung & Lee, 1999). Writers such as Ian Hacking(1995, 1999) have argued persuasively that some mental disorders(e.g., multiple personality disorder) consist of roles that are created bythe theories and practices of the mental health professions and subse-quently enacted by patients. The articulation and dissemination ofinformation about psychopathology through professional activitiesand by the media provide a symptom set and patient profile that canbe assimilated by disturbed individuals who possess sufficient psychicmalleability (Woolfolk, 1998).

Historically in Western psychiatry, a mental disorder has beenposited in one of two instances. The first of these occurs when thereexists a theory of psychogenesis, such as psychoanalysis, that hypothe-sizes mental entities to be the underlying causes of the symptoms of adisorder. The second instance involves the presence of symptoms inthe absence of a physicalistic explanation. In this second instance, psy-chogenic etiology may be inferred solely from the absence of a knownunderlying physical mechanism, thus revealing a tacit dualism thatoriginated even before Paracelsus and that continues to underlie West-ern medicine: Disease entities, whether they be causes or symptoms,belong to one of two categories, either the physical or the mental, thesetwo categories being mutually exclusive (Robinson, 1996). Symptomsof almost any variety that cannot be linked to a scientifically explainedphysical pathology are assumed to be psychogenic. Individuals af-flicted with multiple sclerosis, Wilsons disease, temporal lobe epi-lepsy, and numerous other maladies currently within the purview ofsomatic medicine were once regarded as mentally ill. Through thecourse of medical progress, mental illness has served as a residual cate-gory wherein poorly understood or refractory illness has been placed,often temporarily, only to be removed when medical science estab-lished the physical mechanisms underlying the disorder (Grob, 1991).

Dualistic assumptions operate not so subtly within DSM-IV. InDSM-IV two principal classes of disease entities are posited: (1) gen-eral medical conditions and (2) primary mental disorders. Contra-distinct from the most paradigmatic mental disorders contained inDSM-IV are mental symptoms resulting from a general medical con-

Introduction 7

2Each of these culture-specific syndromes comprises somatic and psychologicalsymptoms.

dition (read: physical illness). Symptoms arising from such causes,indeed, imply the absence of a mental disorder. Of course, DSM-IVsauthors claim that the distinction between primary mental disordersand those stemming from a general medical condition should not betaken to imply that there are fundamental differences between men-tal disorders and general medical conditions (American PsychiatricAssociation, 1994, p. 165). But the volume is careful to distinguishsymptoms deriving from a general medical disorder from those thatemanate from a primary mental disorder. This distinction is drawnso sharply that symptoms of organic origin are exclusionary for thediagnosis of such paradigmatic disorders as schizophrenia. The locu-tion of the volume instantiates not only dualism, but also logical cir-cularity, in that a general medical condition is defined as a medicalcondition other than a primary mental disorder, and a primary men-tal disorder is defined as something other than the result of a generalmedical condition. The distinction drawn here, whether nominal orsubstantive, is old Cartesian wine in a new bottle, that venerable dis-tinction between the functional and the organic. In its languageDSM-IV also stipulates, as a kind of axiom, the historical role of psy-chiatry as a processor of aberrations within the category of illness.Mental illnesses are abnormal, poorly understood illnesses that nor-mal physicians do not treat.

Some have argued that the concept of somatization is unintelligi-ble in medical traditions without a dualistic ontology. In many socie-ties the concept of somatization has no meaning, since distinctionsbetween mental and physical illness are not prevalent (Fabrega, 1991).For example, within the medical traditions of China and India, illnessis conceived holistically in terms of various imbalances. The mindbody distinction is neither fundamental nor sharply drawn.

Studies of mental illness in non-Western societies reveal that so-matic, rather than psychological symptomatology, often is the primaryindication of a psychiatric disorder. For example, research in China hasfound that symptoms of psychiatric patients were predominantly so-matic. For years the most commonly diagnosed mental disorder inChina has been shenjing shuairuo, an indigenous diagnostic categorysignifying a weakness of nerves (Parker, Gladstone, & Chee, 2001).This disorder is described in the Chinese Classification of Mental Dis-eases, 2nd Edition, Revised (CCMD-2-R; Chinese Medical Association &Nanjing Medical University, 1995) and is accepted as a commonplaceand legitimate illness by both medical practitioners and the generalpublic. The disorder, oddly enough, is characterized largely by somaticsymptoms, many of the same symptoms treated by Euroamericannerve doctors in the 19th century, such as fatigue (Shorter, 1997).


Hence, the disorder is most often translated for Westerners as neuras-thenia. Medical anthropologist Arthur Kleinman (1982) evaluated asample of Chinese neurasthenics and determined that the majoritymanifested significant depressive symptomatology, albeit not suffi-cient nor in the requisite configuration to meet DSM criteria for a diag-nosis of major depression. He concluded that many Chinese given theneurasthenia diagnosis could be suffering from depression, though notthe Euroamerican form of depression that is characterized by despon-dence and patterns of thinking described by cognitive theorists such asAaron Beck (Beck, 1976; Beck, Rush, Shaw, & Emery, 1979). In fact,studies of depression in China have found that relatively few Chinesemanifest the DSM-IV syndrome of depressed mood, self-criticism,guilt, and pessimism. Chinese epidemiological research also suggeststhat patients with anxiety, like patients with depression, present agreater ratio of somatic to psychological symptoms than that found inthe West (Parker, Cheah, & Roy, 2001; Tsoi, 1985; Zhang, Shen, & Li,1998).

Research on psychopathology in China frequently is used to arguethat cultural factors are crucial in determining the manner in whichhuman suffering is experienced and, more specifically, to support theview that non-Western societies are prone to generate somatic expres-sions of distress. One distinction that is made in cross-cultural theory isthat between somatization and psychologization. The former refersto the experience of bodily aspects of distress whereas the latter refersto the experience of the psychic, social, and mental aspects of distress(Kirmayer, 1984; White, 1982). According to this formulation, either so-matization or psychologization could serve as alternative modalitiesthrough which a negative emotional reaction is experienced and asalternative idioms of distress through which emotional pain is com-municated. It has been suggested that psychologization is compatiblewith Western, Euroamerican concepts of selfhood and with an individ-ualistic, psychologically minded worldview that emphasizes causalexplanations implicating individuals and their traits as sources ofevents (Kihlstrom & Canter Kihlstrom, 1999; Kirmayer, Young, & Rob-bins, 1994). Somatization, in contrast, has been associated with themore sociocentric cultural views of selfhood where self-reflection andself-examination are deemphasized or disvalued; here, behavior ismore often viewed as caused by the external environment, rather thanby qualities of the person such as psychological traits or willpower.Other factors that might influence the ratio of psychologization to so-matization are the stigma attached to psychological symptoms and thedegree to which a desired treatment is obtained through either a psy-chological or a somatic presentation (Kirmayer, 2001).

Introduction 9

The twin, reciprocal processes of somatization and psychologi-zation are at first glance somewhat obscure and inaccessible. Theworkings are most often illustrated by examples of how emotional dis-tress might be somatized. But let us first take the example of how aphysical illness might be psychologized. Suppose you are a personwho has contracted a mild viral infection and you are beginning tobecome symptomatic. Typically, if you conduct an examination of yourinitial symptoms, you may tend to emphasize elevated core body tem-perature or gastrointestinal motility, which you or your healthcare pro-vider can readily link to the effect of a viral infection. The most familiarbodily discomforts of an infection constitute an incomplete inventoryof all of its adverse accompaniments. There may be feelings of lethargyand dysphoria and difficulty concentrating. Why emphasize the for-mer somatic symptomatology rather than the latter psychologicaleffects of an infection? One could argue that the practice depends uponthe patients frame of reference. We report our physical symptoms, itcould be argued, because we think we have an illness, which we con-ceive as a disorder stemming from physical causes, the importanteffects of and treatments for we believe to be somatic. Yet suppose youhave the aforementioned symptoms but do not know that you have aninfection and do not, therefore, privilege and adopt a physicalisticframe of reference. Suppose, also, that you are in the midst of a psycho-logical crisis, such as job loss or divorce. Perhaps you have recentlyseen a daytime television talk show on which people report the vari-ous adverse emotional effects of psychological stress. Under these cir-cumstances, with an attributional bias primed by salient psychosocialperturbations, you might be less inclined to measure your body tem-perature and, instead, might be disposed to focus on your mental stateand conclude that you are depressed or stressed out.

The theory we have been describing assumes that the backgroundassumptions or the idiom of distress is crucial to determiningwhether a disruption of homeostasis is experienced as physical ormental. An individual can learn to attend to and express physical dis-comfort, rather than psychological distress, especially if an idiom ofaffect is not available. Thus, the theory has it that somatization con-ditions can be shaped through processes of selective attention to physi-cal symptoms and by learning a vocabulary of somatic symptomatolo-gy.

Part of the allure of cross-cultural research on somatization symp-toms is the possibility that cultural differences in the experience of ill-ness might hold the key to understanding the mechanisms that under-lie somatization. At this juncture, however, cross-cultural research on


unexplained physical symptoms must be regarded as inconclusive andfraught with methodological problems, including confounds involvingsocioeconomic factors and cross-cultural differences in healthcare sys-tems.

Whether there are cultures that foster somatization is still a com-plex and controversial question. The World Health Organizationsinternational collaborative study of Psychological Problems in GeneralHealth Care (Gureje, Simon, Ustun, & Goldberg, 1997) did not find thedisparity in somatization disorders between East and West that mighthave been predicted from the early formulations of medical anthropol-ogists. Nor did the ratio of somatic to psychological symptoms ofdepression vary across cultures in a systematic or expected fashion.The data were, however, to some degree consistent with the culturalhypothesis. For example, somatization rates were significantly higherin Latin America than in the rest of the world, and rates of somatiza-tion were higher in China than in the United States (Simon, Von Korff,Piccinelli, Fullerton, & Ormel, 1999). This study failed to include indig-enous, culture-specific syndromes or to analyze single-symptom pre-sentations, nor was there adequate assessment of the ratio of somatiza-tion to psychologization in the more frequently diagnosed forms ofpsychopathology, such as depression. A study that provides probativeevidence on relative somatizing versus psychologizing tendenciesacross cultures has yet to be conducted.

MODELS OF SOMATIZATION

Not since the psychoanalytic era has somatization been viewed as awell-understood phenomenon. In some sense we have not progressedvery far beyond Breuer and Freuds psychodynamic theory of conver-sion hysteria. Currently, there is widespread admission among author-ities that no adequate theory of somatization exists. Indeed, one mightargue that, with the exception of the oft-criticized and scientificallybeleaguered psychoanalytic theory, there are no well-developed theo-ries of somatization, only some fragmentary models or speculation. InChapter 3, we describe and analyze some models of somatization andexamine the relevant empirical evidence supporting each.

In evaluating models of somatization one needs to be mindful ofthe logical pitfalls that abound in the territory of mindbody relation-ships. As we have seen in conjecture about the cause of somatization,such phrases as emotional distress expressed as physical symptomsor a somatic idiom of distress or a tendency to somatize rather than

Introduction 11

psychologize are invoked. Such locutions are problematic and be-speak the poverty of our theories as they invariably risk the possibilityof emerging as either pseudoexplanatory or tautologous. Such formu-lations often fail to explain because they leave key terms undefinedand unexplicated. We have seen this kind of fallacious logic before inpsychiatric discourse that utilizes the notion of chemical imbalancein explanations of the treatment of depressionthat is, a prior chemi-cal imbalance is inferred from a presumptive balancing of neuro-chemistry by antidepressants. Key concepts that are unexplained canresult in circular reasoning, illustrated by Molieres physician whoattributed the effect of a soporific to its dormative powers. A cogentexplanation of somatization must spell out exactly what is denoted bythe emotional distress that is putatively expressed somatically andalso include valid and reliable methods for measuring it. Can the emo-tional distress thought to underlie somatization be identified indepen-dently of its somatic expression? If not, what is the epistemological sta-tus of our model or theory? Are we simply assuming a priori that thereis some emotional basis for any unexplained physical symptom? If so,we have engaged in question begging rather than explanation.

Fortunately for scientists, healthcare providers, and patients, vari-ous diagnoses and the forms of psychotherapy and pharmacotherapyapplied are warranted, for the most part, by empirical findings thatvalidate clinical practices rather than confirm underlying theory. Thereare few cases in psychiatry in which a treatment can be shown to pro-duce clinical benefits because it affects a well-understood mechanismthat is implicated conclusively in pathogenesis. We simply do not havetheories or models of mental disorders that have been validated in themanner of our theories of, for example, infectious diseases. Our tech-nologies of healing are legitimized not by the verification of the under-lying theory, but by the efficacies of these technologies. Research ontreatments for mental disorders is much more akin to industrial prac-tices of product testing than to theory-based applied science. Fortu-nately, effective clinical interventions need not wait upon validated sci-entific theories of psychiatric disorder. We and our colleagues, as didthe empirics of old, put our money on pragmatic observable results ofinterventions, as opposed to armchair speculation about the truenature of things.

This is not to say that we practice dust-bowl empiricism or areunguided or undisciplined by a priori assumptions, models, and theo-ries. Our approach to somatization draws heavily on several sources:(1) stress research and the stress-management and self-regulationliterature; (2) the contemporary psychology of emotion and those


experiential approaches to psychotherapy that emphasize emotionalprocessing; (3) social learning theory and the cognitive-behavioralinterventions that are predicated on cognitive-appraisal and condition-ing models of behavior; (4) role theory as it derives from socioculturalanalyses of illness behavior by sociologists and medical anthropolo-gists. In Chapter 3, we elaborate the rationale for our treatment, butfirst, in Chapter 2, we define more precisely the problem we wish totreat.

Introduction 13

C H A P T E R 2

SomatizationEpidemiology, Clinical Characteristics,

and Treatment

Maria had just lifted the last bag of groceries out of her shopping cart.As she turned toward the open hatch of her SUV, a sharp, stabbing painmade her cry out. As she clutched her lower abdomen, the bag landedon the pavement and its contents spilled out onto the parking lot. Out ofthe corner of her eye, Maria saw an orange rolling rapidly away fromher. She cursed quietly as she realized that she was doubled over with asearing pain that ran between her hips and across the front of her body.Slowly, Maria lowered herself to the pavement and lay on her side. Shehad experienced this pain before and knew that if she reclined, it wouldeventually subside. The pains had begun in her late 20s, and doctorshad been of little help. Irritable bowel syndrome was the diagnosis thatthe gastroenterologist had given her. Maria had tried various treatments,without any real benefit. Her primary care physician had suggested thatstress was the likely cause, but that conjecture had not been helpful.Maria tried to reduce her stress, but the pains seemed unpredictableand unrelated to other events in her life. They would come, out of theblue, even when she felt calm.

Our treatment is designed for patients with multiple unexplainedphysical symptoms, including both patients who meet full criteria forsomatization disorder and those whose symptoms are not sufficientlynumerous or diverse to qualify for a somatization disorder diagnosis.We have adopted this broad definition of somatization because itseems to reside on a continuum, one that cannot be adequately repre-sented by a categorical scheme (Katon et al., 1991). We would suggest

14

that somatization disorder represents one extreme on the somatizationcontinuum and that subthreshold somatization (including undifferen-tiated somatoform disorder, multisomatoform disorder, abridged so-matization, irritable bowel syndrome, chronic fatigue syndrome, andfibromyalgia) lies closer to the midpoint on the same continuum.

This chapter reviews the diagnostic criteria for the aforemen-tioned somatization-spectrum disorders, as well as the epidemiology,clinical characteristics, and treatments associated with the various dis-orders.

DEFINITIONS AND EPIDEMIOLOGY

Somatization Disorder

Although medicine has long recognized the existence of a group ofpatients with multiple medically unexplained symptoms and abnor-mal illness behavior, there has been and continues to be disagreementover the precise diagnostic criteria for somatization disorder. The termsomatization disorder was introduced in the third edition of the Ameri-can Psychiatric Associations classification system, the Diagnostic andStatistical Manual of Mental Disorders (DSM-III; American PsychiatricAssociation, 1980). The criteria required for its diagnosis have beenrevised in each of the manuals subsequent versions. According toDSM-III, somatization disorder in men is characterized by a lifetimehistory of at least 12 medically unexplained physical symptoms; forwomen, the diagnosis requires a history of at least 14 medically unex-plained physical symptoms (American Psychiatric Association, 1980).Abandoning DSM-IIIs gender distinction, DSM-III-R defines somatiz-ation disorder as a lifetime history of at least 13 medically unexplainedphysical symptoms for both men and women (American PsychiatricAssociation, 1987). More recently, the authors of DSM-IV reduced thenumber of symptoms required for the somatization disorder diagnosisto eight, while introducing the requirement that those symptomsinvolve multiple bodily systems. According to DSM-IV (AmericanPsychiatric Association, 1994), somatization disorder is characterizedby at least four unexplained pain symptoms, two unexplained non-pain gastrointestinal symptoms, one unexplained sexual or menstrualsymptom, and one pseudoneurological symptom (see Table 2.1). TheWorld Health Organization (WHO) has offered yet another set of diag-nostic criteria for somatization disorder. In the International StatisticalClassification of Diseases and Related Health Problems (ICD-10) the disor-der requires a history of at least six unexplained symptoms from atleast two of the following symptom groups: gastrointestinal, cardio-

Epidemiology, Clinical Characteristics, and Treatment 15

vascular, genitourinary, and skin/pain (World Health Organization,1993). In all three versions of DSM and in ICD-10, symptoms countedtoward the diagnosis of somatization disorder must be either medi-cally unexplained or experienced in excess of what would be expectedfrom physical examination and laboratory findings. In addition, ICD-10 requires that each symptom be distressing enough to prompt therepeated seeking of medical treatment. In DSM-III and its later revi-sions, unexplained physical symptoms counted toward the somatiza-


TABLE 2.1. DSM-IV Diagnostic Criteria for Somatization Disorder

A. A history of many physical complaints beginning before age 30 years thatoccur over a period of several years and result in treatment being sought orsignificant impairment in social, occupational, or other important areas offunctioning.

B. Each of the following criteria must have been met, with individualsymptoms occurring at any time during the course of the disturbance.(1) four pain symptoms: a history of pain related to at least four different sites

or functions (e.g., head, abdomen, back, joints, extremities, chest, rectum,during menstruation, during sexual intercourse, or during urination)

(2) two gastrointestinal symptoms: a history of at least two gastrointestinalsymptoms other than pain (e.g., nausea, bloating, vomiting other thanduring pregnancy, diarrhea, or intolerance of several different foods)

(3) one sexual symptom: a history of at least one sexual or reproductivesymptom other than pain (e.g., sexual indifference, erectile or ejaculatorydysfunction, irregular menses, excessive menstrual bleeding, vomitingthroughout pregnancy)

(4) one pseudoneurological symptom: a history of at least one symptom ordeficit suggesting a neurological condition not limited to pain(conversion symptoms such as impaired coordination or balance,paralysis or localized weakness, difficulty swallowing or lump in throat,aphonia, urinary retention, hallucinations, loss of touch or painsensation, double vision, blindness, deafness, seizures; dissociativesymptoms such as amnesia; or loss of consciousness other thanfainting)

C. Either (1) or (2):(1) after appropriate investigation, each of the symptoms in Criterion B

cannot be fully explained by a known general medical condition or thedirect effects of a substance (e.g., a drug of abuse, a medication)

(2) when there is a related general medical condition, the physicalcomplaints or resulting social or occupational impairment are in excessof what would be expected from the history, physical examination, orlaboratory findings

D. The symptoms are not intentionally produced or feigned (as in FactitiousDisorder or Malingering).

Note. From American Psychiatric Association (1994). Copyright 1994 by the American Psychi-atric Association. Reprinted by permission.

tion disorder diagnosis must have either prompted the seeking ofmedical treatment or impaired the individuals functioning.

Not only do the various versions of DSM and ICD establish differ-ent symptom thresholds for the diagnosis of somatization disorder, butthey also utilize different lists of criterial symptoms. The symptom listsfor somatization disorder according to DSM-III, DSM-III-R, DSM-IV,and ICD-10, as well as the lists used for hysteria (Feighner et al., 1972;Guze, 1967; Perley & Guze, 1962) are displayed in Table 2.2. The 59-item PerleyGuze/Feighner list was shortened for DSM-III largely byeliminating its psychological symptoms (e.g., nervousness, depressedfeelings). Additional, though minimal, reductions to the original listwere made for DSM-III-R, and it was condensed further for DSM-IV.DSM-III and DSM-III-R fashion a purely somatic variety of hysteria,while DSM-IV includes hallucinations, one of the original psychologi-cal symptoms from Perley and Guzes list. This symptom is not associ-ated with somatization disorder in either DSM-III or DSM-III-R. ICD-10 somatization disorder bears little resemblance to Perley and Guzeshysteria. Symptoms listed in ICD-10 are significantly fewer in number,including none of the menstrual or pseudoneurological symptomsconsidered the cardinal features of the original syndrome (Guze &Perley, 1963). In ICD-10 a distinction is drawn between somatization/somatoform disorders and dissociative/conversion disorders. Pseudo-neurological symptoms fall into the latter category according to theWHO.

A decade since the publication of the most recent classificationsystems (ICD-10 and DSM-IV), discord continues over appropriatesymptom lists and symptom thresholds for diagnosing somatizationdisorder. All four versions of somatization disorder are predictive ofimportant clinical outcomes, such as overuse of medical services andfunctional impairment (Bass & Murphy, 1991; Gureje, Simon, et al.,1997; Smith et al., 1986a; Swartz, Blazer, George, & Landerman, 1986).There is little evidence, however, supporting DSM-IVs and ICD-10ssubgrouping of symptoms (Liu, Clark, & Eaton, 1997). Also, given thegreater likelihood that women will experience and report somatizationsymptoms than will men, some investigators have encouraged thecontinued use of different symptom thresholds for men and women, asin DSM-III (Liu et al., 1997). And, although most research suggests thatthe criteria from the various versions of DSM somatization disor-der (and from Perley and Guzes hysteria) identify roughly identicalgroups of patients, ICD-10s criteria for somatization disorder mayindeed designate a different group of patients (Rief et al., 1996; Yutzyet al., 1995).


18

TAB

LE2.2

.Sy

mpt

omLi

sts

for

Cla

ssifi

catio

nSy

stem

sof

Som

atiz

atio

nD

isor

der/

Hys

teria

Perl

eyG

uze/

Feig

hner

DSM

-III

DSM

-III

-RD

SM-I

VIC

D-1

0

Pseu

don

euro

logi

cal

sym

ptom

sPa

raly

sis

Para

lysi

sor

mus

cle

wea

knes

sPa

raly

sis

orm

uscl

ew

eakn

ess

Para

lysi

sor

loca

lized

wea

knes

sb

Trou

ble

wal

king

(ata

xia)

aTr

oubl

ew

alki

ngTr

oubl

ew

alki

ngIm

pair

edco

ord

inat

ion

orba

lanc

eb

Lum

pin

thro

atD

iffic

ulty

swal

low

ing

Diff

icul

tysw

allo

win

gD

iffic

ulty

swal

low

ing

orlu

mp

inth

roat

b

Aph

onia

Los

sof

voic

eL

oss

ofvo

ice

Aph

onia

b

Dea

fnes

sD

eafn

ess

Dea

fnes

sD

eafn

essb

Dou

ble

visi

onD

oubl

evi

sion

Dou

ble

visi

onb

Blin

dne

ssB

lind

ness

Blin

dne

ssB

lind

ness

b

Fits

orco

nvul

sion

sSe

izur

esor

conv

ulsi

ons

Seiz

ure

orco

nvul

sion

Seiz

ures

b

Uri

nary

rete

ntio

nU

rina

ryre

tent

ion

ord

iffic

ulty

urin

atin

gU

rina

ryre

tent

ion

ord

iffic

ulty

urin

atin

gU

rina

ryre

tent

ionb

Fain

ting

spel

lsFa

inti

ngor

loss

ofco

nsci

ousn

ess

Fain

ting

orlo

ssof

cons

ciou

snes

sL

oss

ofco

nsci

ousn

ess

(oth

erth

anfa

inti

ng)b

Am

nesi

aM

emor

ylo

ssA

mne

sia

Am

nesi

ab

Vis

ual

blur

ring

Blu

rred

visi

onB

lurr

edvi

sion

Ane

sthe

sia

Los

sof

touc

hor

pain

sens

atio

nbU

nple

asan

tnu

mbn

ess

orti

nglin

gc

Hal

luci

nati

ons

Hal

luci

nati

onsb

Unc

onsc

ious

ness

Wea

knes

sO

ther

conv

ersi

onsy

mpt

omsa

19

Gas

troi

ntes

tina

lsy

mpt

oms

Bad

tast

ein

mou

thor

coat

edto

ngue

d

Nau

sea

Nau

sea

Nau

sea

(oth

erth

anm

otio

nsi

ckne

ss)

Nau

seae

Nau

sead

Vom

itin

gV

omit

ing

(oth

erth

and

urin

gpr

egna

ncy)

Vom

itin

g(o

ther

than

dur

ing

preg

nanc

y)V

omit

ing

(oth

erth

and

urin

gpr

egna

ncy)

eV

omit

ing

orre

gurg

itat

iond

Abd

omin

albl

oati

ngB

loat

ing

(gas

sy)

Blo

atin

g(g

assy

)B

loat

inge

Feel

ing

bloa

ted

orfu

llof

gasd

Dia

rrhe

aD

iarr

hea

Dia

rrhe

aD

iarr

heae

Freq

uent

and

loos

ebo

wel

sor

anal

dis

char

ged

Food

into

lera

nces

Food

into

lera

nce

Food

into

lera

nce

Food

into

lera

ncee

Con

stip

atio

nA

nore

xia

Wei

ght

loss

Mar

ked

fluc

tuat

ions

inw

eigh

t

Sexu

al/

men

stru

alsy

mpt

oms

Men

stru

alir

regu

lari

tyM

enst

rual

irre

gula

rity

Irre

gula

rm

enst

rual

peri

ods

Irre

gula

rm

ense

sf

Exc

essi

vem

enst

rual

blee

din

gE

xces

sive

men

stru

albl

eed

ing

Exc

essi

vem

enst

rual

blee

din

gE

xces

sive

men

stru

albl

eed

ingf

Am

enor

rhea

Vom

itin

gth

roug

hout

preg

nanc

ySe

vere

vom

itin

gth

roug

hout

preg

nanc

y

Vom

itin

gth

roug

hout

preg

nanc

yV

omit

ing

thro

ugho

utpr

egna

ncyf

Sexu

alin

diff

eren

ceSe

xual

ind

iffer

ence

Sexu

alin

diff

eren

ceSe

xual

ind

iffer

ence

f

Sexu

alfr

igid

ity

Impo

tenc

eE

rect

ileor

ejac

ulat

ory

dys

func

tion

f

Oth

erse

xual

diff

icul

ties

Lac

kof

plea

sure

dur

ing

inte

rcou

rse

Unu

sual

orco

piou

sva

gina

ld

isch

arge

i

(con

tinu

ed)

20

TAB

LE2.2

.(c

onti

nued

)

Perl

eyG

uze/

Feig

hner

DSM

-III

DSM

-III

-RD

SM-I

VIC

D-1

0

Pain

sym

ptom

sA

bdom

inal

pain

Abd

omin

alpa

inA

bdom

inal

pain

(oth

erth

anw

hen

men

stru

atin

g)

Pain

inab

dom

eng

Abd

omin

alpa

ind

Bac

kpa

inPa

inin

back

Bac

kpa

inPa

inin

back

g

Pain

inlim

bs,

extr

emit

ies,

orjo

ints

cJo

int

pain

Pain

injo

ints

Join

tpa

inPa

inin

join

tsg

Ext

rem

ity

pain

Pain

inex

trem

itie

sPa

inin

extr

emit

ies

Pain

inex

trem

itie

sg

Bur

ning

inse

xual

orga

ns,m

outh

,or

rect

um

Pain

inge

nita

lar

ea(o

ther

than

dur

ing

sex)

Bur

ning

inse

xual

orga

nsor

rect

um(o

ther

than

dur

ing

sex)

Pain

inre

ctum

gU

nple

asan

tge

nita

lse

nsat

ions

i

Pain

dur

ing

urin

atio

nPa

inon

urin

atio

nPa

ind

urin

gur

inat

ion

Pain

dur

ing

urin

atio

ngD

ysur

iaor

freq

uent

urin

atio

ni

Dys

men

orrh

eaPa

infu

lm

enst

ruat

ion

Pain

ful

men

stru

atio

nPa

ind

urin

gm

enst

ruat

iong

Dys

pare

unia

Pain

dur

ing

inte

rcou

rse

Pain

dur

ing

inte

rcou

rse

Pain

dur

ing

inte

rcou

rseg

Oth

erbo

dily

pain

Oth

erpa

in(o

ther

than

head

ache

s)O

ther

pain

(oth

erth

anhe

adac

he)

Hea

dac

hePa

inin

head

g

Car

dio

vasc

ular

sym

ptom

sC

hest

pain

Che

stpa

inC

hest

pain

Pain

inch

estg

Che

stpa

insh

Bre

athi

ngd

iffic

ulty

Shor

tnes

sof

brea

thSh

ortn

ess

ofbr

eath

whe

nno

tex

erti

ngon

esel

fB

reat

hles

snes

sw

itho

utex

erti

onh

Palp

itat

ion

Palp

itat

ions

Palp

itat

ions

Diz

zine

ssD

izzi

ness

Diz

zine

ss

21

Oth

ersy

mpt

oms

Feel

ing

sick

lym

ost

oflif

eFe

elin

gsi

ckly

muc

hof

life

Fati

gue

Anx

iety

atta

cks

Ner

vous

ness

Fear

sD

epre

ssed

feel

ings

Qui

ttin

gw

ork

orot

her

func

tiona

llim

itatio

nsbe

caus

efe

elin

gsi

ckC

ryin

gea

sily

Feel

ing

life

isho

pele

ssT

hink

ing

ago

odd

eal

abou

td

ying

Wan

ting

tod

ieT

hink

ing

ofsu

icid

eSu

icid

eat

tem

pts

Blot

chin

ess

ordi

scol

orat

ion

ofsk

inc

a Inc

lud

edin

Feig

hner

crit

eria

only

(i.e

.,no

tin

Perl

ey&

Guz

es

[196

2]or

igin

alsy

mpt

omlis

t).

b Cat

egor

ized

inth

eps

eud

oneu

rolo

gica

lcl

uste

rof

sym

ptom

sin

DSM

-IV.

c Cat

egor

ized

inth

esk

in/

pain

clus

ter

ofsy

mpt

oms

inIC

D-1

0.d C

ateg

oriz

edin

the

gast

roin

test

inal

clus

ter

ofsy

mpt

oms

inIC

D-1

0.e C

ateg

oriz

edin

the

gast

roin

test

inal

clus

ter

ofsy

mpt

oms

inD

SM-I

V.fC

ateg

oriz

edin

the

sexu

al/

men

stru

alcl

uste

rof

sym

ptom

sin

DSM

-IV.

gC

ateg

oriz

edin

the

pain

clus

ter

ofsy

mpt

oms

inD

SM-I

V.h

Cat

egor

ized

inth

eca

rdio

vasc

ular

clus

ter

ofsy

mpt

oms

inIC

D-1

0.iC

ateg

oriz

edin

the

geni

tour

inar

ycl

uste

rof

sym

ptom

sin

ICD

-10.

Despite the differences among the specific criteria for somatiza-tion disorder, the essence of somatization disorder according to anyclassification system is chronic unexplained physical symptoms. Ac-cording to the diagnostic criteria described in DSM-III, DSM-III-R, andDSM-IV, at least some of the somatization symptoms must haveoccurred prior to the patients 30th birthday (American PsychiatricAssociation, 1980, 1987, 1994). The ICD-10 criteria require the patientto report experiencing somatization symptoms for at least 2 years butdo not specify at what age the symptoms must have first occurred(World Health Organization, 1993). The course of somatization dis-order tends to be characterized by symptoms that wax and wane,remitting only to return later and/or be replaced by new unexplainedphysical symptoms (Lieb et al., 2002). Thus, an important feature ofsomatization disorder is that it is a polysymptomatic disorder whoserequisite symptoms need not be manifested concurrently.

This persistent, polysymptomatic disorder is difficult to identifyin clinical practice. Spotting it requires the clinician to review patientspast and current symptomatology and to determine whether eachphysical complaint is medically unexplained or whether it producesdiscomfort in excess of that expected from the organic pathologyobserved. Furthermore, the clinician must establish whether the symp-tom has brought about functional impairment, the use of medication,and/or the seeking of medical treatment. Time limitations may pre-vent clinicians from investigating so many symptoms that may haveoccurred over so long an interval. For example, a patient who has ahistory of multiple unexplained physical symptoms but is presentlyexperiencing only unexplained and debilitating back pain may notreceive a diagnosis of somatization disorder if the clinician asks onlyabout the patients current symptomatology. Also, clinicians may hesi-tate to make the somatization diagnosis because of uncertainty regard-ing potential underlying organic pathology. Because of these complexrequirements for a somatization disorder diagnosis, it is likely under-diagnosed in medical and psychiatric clinics (Fabrega, Mezzich, Jacob,& Ulrich, 1988; Fink, Sorensen, Engberg, Holm, & Munk-Jorgensen,1999; Peveler, Kilkenny, & Kinmonth, 1997).

The differences in the diagnostic criteria of DSM-III, DSM-III-R,DSM-IV, and ICD-10, though subtle in some instances, may be respon-sible for some of the inconsistencies in the epidemiological findingsthat are described below. Some epidemiological research has suggestedthat somatization disorder is relatively rare. In the EpidemiologicCatchment Area (ECA) study, the largest survey of somatization disor-der (carried out in a community sample of 20,000 people across five


sites in the United States), the lifetime prevalence of DSM-III somatiza-tion disorder was 0.13% (Robins & Reiger, 1991). Prior research hadresulted in higher estimates for Perley and Guzes hysteria, rangingfrom 0.4 to 2% of the population (Weissman, Myers, & Harding, 1978;Woodruff, Clayton, & Guze, 1971). Not surprisingly, somatization dis-order appears to be more common in primary care settings than incommunity populations. In the WHO Cross-National Study of MentalDisorders in Primary Care, in which 5,438 primary care patients at 15centers in 14 countries were assessed, the prevalence of somatizationdisorder was 0.9% and 2.8%, as defined by DSM-III-R and ICD-10,respectively (Gureje, Simon, et al., 1997). Smaller studies conducted inprimary care settings have estimated the prevalence of DSM-III-R so-matization disorder to range from 1% in a sample of 685 patients(Kirmayer & Robbins, 1991) to 5% of a sample of 222 patients (Peveleret al., 1997).

For a number of reasons, several authorities suggest the actualprevalence of somatization disorder may be substantially higher thanthe literature suggests. First, autobiographical memory of past psychi-atric symptoms, including somatization symptoms, is unreliable. Indi-viduals seem to forget (or at least fail to report) previously reportedsymptoms that are no longer troublesome (Simon & Gureje, 1999;Simon & Von Korff, 1995). Given that the somatization disorder diag-nosis requires patients to describe both current and remitted symp-toms, the latter often not recalled, the true occurrence of somatizationdisorder is probably underestimated in these studies. Second, the diag-nosis of somatization disorder necessitates that a physical examina-tion and diagnostic tests be performed, or that medical records bereviewed, to determine the nature of each symptom. Such extensiveinvestigations of physical symptoms are too costly to incorporate intolarge epidemiological studies. The third argument for the underesti-mation of prevalence rates has to do with the conjecture that physi-cians are more likely to make the somatization diagnosis than arenonphysician diagnosticians (Martin, 1991; Swartz, Blazer, et al., 1986).Presumably, physicians are better able to distinguish between a medi-cally sound explanation for a symptom and the patients medical-sounding explanation for that symptom. Because nonphysicians con-ducted assessments without access to medical records in the ECA andWHO studies cited above, somatization disorder may have beenunderdiagnosed in those studies. At the one site in the ECA studywhere physicians evaluated symptom reports to determine whethersymptoms were medically explained, the prevalence of somatizationdisorder was higher than at the other sites (Swartz, Blazer, et al., 1986).


Subthreshold Somatization

Some investigators have encouraged a broadening of the somatizationconstruct to include the many patients affected by unexplained symp-toms not numerous enough to meet criteria for full somatization disor-der. Both ICD-10 and DSM-IV include residual diagnostic categoriesfor subthreshold somatization cases. In DSM-IV (American PsychiatricAssociation, 1994), undifferentiated somatoform disorder is a diag-nosis characterized by one or more medically unexplained physicalsymptoms lasting for at least 6 months (see Table 2.3). ICD-10s undif-ferentiated somatoform disorder category differs from DSM-IVs inthat ICD-10 requires multiple unexplained symptoms lasting for atleast 6 months (World Health Organization, 1993). ICD-10 provides anadditional category for subthreshold somatization disorder, somato-form autonomic dysfunction, for cases of three or more unexplainedsymptoms of autonomic arousal (World Health Organization, 1993).We are aware of no published data that establish the validity of any ofthese three diagnostic categories.

Two research teams have suggested categories for subthresholdsomatization other than those described in DSM-IV and ICD-10. Escobar,


TABLE 2.3. DSM-IV Diagnostic Criteria for UndifferentiatedSomatoform Disorder

A. One or more physical complaints (e.g., fatigue, loss of appetite,gastrointestinal or urinary complaints).

B. Either (1) or (2):(1) after appropriate investigation, the symptoms cannot be fully explained

by a known general medical condition or the direct effects of a substance(e.g., a drug of abuse, a medication)

(2) when there is a related general medical condition, the physicalcomplaints or resulting social or occupational impairment are in excessof what would be expected from the history, physical examination, orlaboratory findings

C. The symptoms cause clinically significant distress or impairment in social,occupational, or other important areas of functioning.

D. The duration of the disturbance is at least 6 months.

E. The disturbance is not better accounted for by another mental disorder (e.g.,another Somatoform Disorder, Sexual Dysfunction, Mood Disorder, AnxietyDisorder, Sleep Disorder, or Psychotic Disorder).

F. The symptom is not intentionally produced or feigned (as in FactitiousDisorder or Malingering).

Note. From American Psychiatric Association (1994). Copyright 1994 by the American Psychi-atric Association. Reprinted by permission.

Burnam, Karno, Forsythe, and Golding (1987) proposed the labelabridged somatization to be applied to men experiencing four ormore unexplained physical symptoms or to women experiencing six ormore unexplained physical symptoms. Kroenke and colleagues (1997)suggested the category of multisomatoform disorder to describemen or women currently experiencing at least three unexplained phys-ical symptoms and reporting a 2-year history of somatization. Both ofthese subthreshold somatization categories appear to be significantlymore prevalent than is full somatization disorder, described earlier.Abridged somatization has been observed in 4.4% of community sam-ples (Escobar, Burnam, et al., 1987) and 16.622% of primary caresamples (Escobar, Waitzkin, Silver, Gara, & Holman, 1998; Gureje,Simon, et al., 1997; Kirmayer & Robbins, 1991). The occurrence ofmultisomatoform disorder has been estimated at 8.2% of primary carepatients (Kroenke et al., 1997).

The demographic characteristic most often associated with soma-tization is sex. In the ECA study, women were 10 times more likelyto meet criteria for somatization disorder than were men (Swartz,Landermann, George, Blazer, & Escobar, 1991). Sex differences, thoughnot as extreme, also have been found in most studies employingsubthreshold somatization categories, such as Escobars abridged so-matization or Kroenkes multisomatoform disorder (Escobar, Rubio-Stipec, Canino, & Karno, 1989; Kroenke et al., 1997). A more complexpicture of the association between sex and somatization was suggestedby the WHOs Cross-National study in which female primary carepatients were more likely to meet ICD-10 criteria for full somatizationdisorder, but no more likely to meet Escobars abridged somatizationcriteria, than were their male counterparts (Gureje, Simon, et al., 1997).At least on the severe end of the continuum, somatization disorder isuncommon in men. Sex differences are less obvious in the varioussubthreshold syndromes.

Current thinking is that the low prevalence of somatization disor-der in men may be explained, in part, by stereotypical male traits, suchas a disinclination to admit discomfort and an unwillingness to seekmedical treatment (see Wool & Barsky, 1994, for review). Also physi-cians may be less likely to consider somatization as a possible explana-tion for a mans symptoms than for a womans symptoms (Golding,Smith, & Kashner, 1991). Gender biases may cause physicians to com-municate with and treat male patients differently from the ways inwhich they communicate with and treat female patients. At this junc-ture, we have only conjectural explanations for the different rates of so-matization in men and women.


Ethnicity, race, and education have also been associated with so-matization disorder and subthreshold somatization. Epidemiologicalresearch has shown patients with somatization were more likely to befemale, nonwhite, and less educated than nonsomatizers (Gureje,Simon, et al., 1997; Robins & Reiger, 1991). Findings on ethnicity havebeen less consistent across studies. In the ECA study, Hispanics wereno more likely to meet criteria for somatization disorder than werenon-Hispanics (Robins & Reiger, 1991). The WHO study, conducted in14 different countries, revealed a higher incidence of somatization, asdefined by either ICD-10 or Escobars abridged criteria, in Latin Amer-ican countries than in the United States (Gureje, Simon, et al., 1997).

Findings on the relationship between age and somatization havebeen somewhat consistent. Most studies indicate that somatization dis-order and abridged somatization are more common in middle-agedand older patients (over 45 years of age) than in younger patients(Gureje, Simon, et al., 1997; Swartz, Blazer, et al., 1986). Some research,however, has detected no association between somatization disorderand age (Robins & Reiger, 1991). Children and adolescents, of course,are extremely unlikely to meet criteria for somatization disorder (andcannot, if one uses DSM-IV criteria), perhaps because they have notlived long enough to acquire enough clinically significant somatizationsymptoms, especially sexual and menstrual symptoms.

Functional Somatic Syndromes

The term functional somatic syndrome is used to describe groups ofco-occurring symptoms that are medically unexplained. Polysympto-matic functional somatic syndromes frequently encountered by bothmental health and primary care practitioners include irritable bowelsyndrome (IBS), chronic fatigue syndrome (CFS), and fibromyalgia.

CFS is characterized by unexplained fatigue, lasting at least 6months, that causes substantial reduction in activities. At least four ofthe following symptoms must have co-occurred with the fatigue: sig-nificant memory impairment or concentration difficulties, sore throat,tender lymph nodes, muscle pain, joint pain, headache, nonrestorativesleep, and postexertional fatigue (Fukuda et al., 1994). CFSs preva-lence has been estimated to be 0.0020.6% in the general populationand 2.6% in primary care settings (Jason et al., 1999; Reyes et al., 2003;Wessely, Chadler, Hirsch, Wallace, & Wright, 1997). Women are morelikely than men to suffer from CFS (Jason et al., 1999; Reyes et al.,2003). Although the incidence of CFS appears to be elevated in minor-ity groups (especially Hispanics and African Americans), larger epide-


miological studies are required to confirm these findings (Jason et al.,1999).

A diagnosis of fibromyalgia is given for chronic widespread painand multiple tender points that have no known biological basis andare accompanied by nonrestorative sleep, fatigue, and malaise (Wolfeet al., 1990). Fibromyalgia has been estimated to occur in about 2% ofthe U.S. population and in 620% of general medical outpatients(Wolfe, Ross, Anderson, Russell, & Hebert, 1995). Like the other func-tional somatic syndromes, fibromyalgia is more likely to affect womenthan men (Wolfe et al., 1995).

IBS is characterized by persistent abdominal pain along withaltered bowel habits and abdominal distension that cannot be ex-plained by organic pathology (Thompson, Dotevall, Drossman, Heaton,& Kruis, 1989). It has been estimated that 820% of the U.S. population,12% of primary care patients (Drossman, Whitehead, & Camilleri,1997), and 2228% of gastroenterologists patients are afflicted with IBS(Harvey, Salih, & Read, 1983; Thompson & Heaton, 1980). The ratio offemale to male patients diagnosed with IBS has been estimated to be2:1 (Drossman et al., 1993).

Our decision to aggregate the research on somatization disorder,subthreshold somatization, CFS, fibromyalgia, and IBS may invite con-troversy. Some medical specialists, focusing on the bodily organ or so-matic system of their specialization, assume these disorders have dis-tinct pathophysiological causes and draw sharp distinctions amongthese syndromes. Other authorities suggest these syndromes should beviewed as one disorder. After all, many patients diagnosed withone functional somatic syndrome meet diagnostic criteria for one ormore of the other functional somatic syndromes, resulting in multi-system comorbid functional syndromes (Buchwald & Garrity, 1994;Goldenberg, Simms, Geiger, & Komaroff, 1990; Veale, Kavanagh,Fielding, & Fitzgerald, 1991; Yunus, Masi, & Aldag, 1989a). Also,investigators have noted similarities in the illness behaviors, illnessbeliefs, and psychological functioning of patients diagnosed with oneof these somatization syndromes (Barsky & Borus, 1999; Rief, Hiller, &Margraf, 1998; Wessely, Nimnuan, & Sharpe, 1999). As will be dis-cussed later in this chapter and in Chapter 3, patients with somatiza-tion tend to adopt a sick role (Parsons, 1951), overutilizing medical ser-vices and withdrawing from their normal activities. They tend toassume their symptoms are signs of a serious, disabling illness that islikely to worsen; they think catastrophically about their health. Theyfrequently suffer from concurrent emotional disorders. Because wehave concluded that the research suggests greater similarity than


distinction among the various categories of somatization, we haveadopted the convention of using the terms somatization and soma-tization syndromes to refer to and encompass the all the poly-symptomatic syndromes, including CFS, fibromyalgia, IBS, abridgedsomatization, multisomatoform disorder, and somatization disorder.

Differential Diagnosis

Within the context of DSMs dualistic framework that distinguishesbetween mental disorders and general medical conditions, somatiza-tion symptoms cannot be accounted for entirely by the latter. Beforethe diagnosis of somatization can be made, potential underlying medi-cal conditions must be excluded. Medical conditions that are mostlikely to account for the multisystem complaints associated with fullsomatization disorder include multiple sclerosis, hyperparathyroid-ism, and systemic lupus erythematosus. Countless other medical ill-nesses could account for single-system or single-symptom complaints,and there are numerous borderline symptoms for which morpholog-ical anomalies can be measured but fail to correspond to subjective dis-tress (e.g., disc herniation and its relation to back pain). Thus, a closecollaboration with patients medical practitioners is required to recog-nize and treat organic pathology.

This is not to say that somatization and organic illness cannot co-occur. Many patients with medically explained symptoms also experi-ence somatization symptoms (Chaturvedi, Maguire, & Somashekar,2006; Heckman et al., 2002; Sonino et al., 2004). These cases, in fact,may be the most difficult to treat (Dickinson et al., 2003). For exam-ple, in patients with cancer, comorbid somatization symptoms havebeen shown to complicate diagnosis, treatment, and cancer outcome(Chaturvedi et al., 2006). General medical conditions also may be theoriginal causes of symptoms that are judged to be excessive or dispro-portionate in relation to the organic cause and, therefore, part of a pat-tern of somatization.

A second area of distinction is between somatization and hypo-chondriasis. Hypochondriasis is defined in DSM-IV as a preoccupa-tion with fears that or the belief that one has a serious, undiagnoseddisease (American Psychiatric Association, 1994). To meet full crite-ria for hypochondriasis, the preoccupation must persist for at leastsix months despite medical assurance of no organic illness (Ameri-can Psychiatric Association, 1994). Although most patients with hy-pochondriasis experience somatic symptoms, it is not the somaticsymptoms themselves that impair functioning in hypochondriasis.Instead, in hypochondriasis, it is the fears about the meaning of the


physical symptoms that cause distress, dysfunction, and overuse ofhealthcare services. Whether hypochondriasis is more accurately rep-resented by a discrete diagnosis or by a dimension is a subject ofdebate (Creed & Barsky, 2004). Not only DSM-IV defined hypo-chondriasis, but excessive health anxiety that does not meet all ofDSM-IVs criteria for a diagnosis of hypochondriasis is associatedwith significant impairment and cost (Gureje, Ustun, & Simon, 1997;Looper & Kirmayer, 2001).

While the boundaries between DSM-IV somatization disorder andhypochondriasis can be distinguished, there remains significant over-lap between these constructs. The co-occurrence of DSM-IV somatiza-tion disorder and DSM-IV hypochondriasis is infrequent (Barsky,Wyshak, & Klerman, 1992; Escobar, Gara, et al., 1998). Only about 20%of patients with hypochondriasis meet DSM criteria for somatizationdisorder, and about 20% of patients with somatization disorder meetDSM criteria for hypochondriasis (Barsky et al., 1992; Escobar, Gara, etal., 1998). Nevertheless, many patients with somatization have hypo-chondriacal beliefs, and many patients with hypochondriasis experi-ence somatization symptoms (Creed & Barsky, 2004). As will be dis-cussed below, similar treatments have been shown to be effective forsomatization and hypochondriasis.

The presentation of an anxiety disorder may resemble that of so-matization (Smith et al., 2005). Patients experiencing panic attacksreport multiple medically unexplained physical symptoms. The differ-ence between somatization and panic disorder is that the symptoms ofthe latter occur with a sudden and acute onset, whereas the former aremore insidious (American Psychiatric Association, 1994). Likewise,generalized anxiety disorder may present itself as a largely somatic setof complaints, including muscle tension and pain, fatigue, and insom-nia. The distinguishing feature of generalized anxiety disorder is thepervasive worry that accompanies the somatic symptoms (AmericanPsychiatric Association, 1994). Although somatic symptoms occurringduring a panic attack or in the context of generalized anxiety disorderare not counted toward the diagnosis of somatization, there are manypatients who report comorbid somatization and panic disorder or gen-eralized anxiety disorder (Brown, Golding, & Smith, 1990; Fink, 1995;Katon, Von Korff, & Lin, 1992; Robins & Reiger, 1991).

So, too, patients experiencing depression may present somaticsymptoms rather than a dysphoric mood or anhedonia to healthcareprofessionals (Smith et al., 2005). Somatic presentations of depressionoften involve symptoms such as fatigue, weakness, gastrointestinaldistress, headaches, other pain symptoms, or hypochondriacal beliefs(Lipowski, 1988), symptoms common in somatoform disorders. Upon


questioning, however, patients with depression acknowledge dimin-ished mood or interest (American Psychiatric Association, 1994), andthe course of their somatic symptoms tends to mirror that of theirmood symptoms (Lipowski, 1988). On the other hand, the course of so-matic symptoms associated with somatization may not be tied asclosely to mood.

The overlap between somatization and depression is high. Whendepression and somatization are conceived and measured as dimen-sions, they are positively correlated (Katon et al., 1991; Simon et al.,1999). When defined by DSM categories, up to 50% of patients with so-matization disorder meet criteria for a lifetime diagnosis of majordepressive disorder (Simon & Von Korff, 1991). Relatively few patientsmeeting criteria for major depressive disorder meet criteria for soma-tization disorder, though up to 75% of patients with depression reportat least one medically unexplained symptom (Corruble & Guelfi, 2000;Simon et al., 1999).

Additional evidence for the overlap among somatization, anxiety,and depressive disorders comes from two other areas of research: lon-gitudinal research on somatization syndromes and research on thefamily members of patients with somatization. Longitudinal researchnot only suggests somatization syndromes are relatively stable, thoughwith some shifts in symptomatic focus (Lieb et al., 2002), but also maybe predictive of future depression and/or anxiety disorders (Hotopf,Carr, Mayou, Wadsworth, & Wesseley, 1998; Zwaigenbaum, Szatmari,Boyle, & Offord, 1999). A preexisting depression or anxiety disorder isassociated with a future somatization syndrome (Gureje & Simon,1999). Family members of patients with somatization appear to havehigher-than-expected rates of depression and anxiety disorders (Hud-son, Arnold, Keck, Auchenbach, & Pope, 2004).

CLINICAL CHARACTERISTICS

Much attention has focused on the illness behavior of patients with so-matization and the resulting impact of that behavior on the healthcaresystem. These patients disproportionately use and misuse healthcareservices. When standard diagnostic evaluations fail to uncover organicpathology, patients with somatization tend to seek additional medi-cal procedures, often from several different physicians. Patients mayeven subject themselves to unnecessary hospitalizations and surgeries,which introduce the risk of iatrogenic illness (Fink, 1992). One studyfound that patients diagnosed with DSM-III somatization disorderincurred nine times the U.S. per capita healthcare cost (Smith et al.,


1986a). Abridged somatization, multisomatoform disorder, CFS, fibro-myalgia, and IBS all have been associated with high healthcare utiliza-tion (Barsky, Orav, & Bates, 2005; Bombardier & Buchwald, 1996;Kroenke et al., 1997; Leong et al., 2003; Lloyd & Pender, 1992; Restak etal., 2003). Pilowsky (1969) used the phrase abnormal illness behaviorto describe the dysfunctional behaviors in which patients with hypo-chondriasis and somatization engage. According to Pilowsky, inappro-priate healthcare-seeking behavior is motivated by fear of diseaseand/or the potential rewards of the sick role. This is not to say thatall somatizers are overutilizers of mainstream medical services. Somefocus entirely on complementary and alternative healthcare, and somecease to seek any remedy after a lengthy history of unsuccessful treat-ment.

The abnormal illness behavior of patients with somatization ex-tends beyond medical offices and hospitals to patients workplacesand households. Somatizers withdraw from both productive and plea-surable activities because of discomfort, fatigue, and/or fears of exac-erbating their symptoms. Investigators have found 1860% of patientsdiagnosed with somatization disorder to be receiving disability pay-ments from either their employers or the government (Allen, Woolfolk,Escobar, Gara, & Hamer, 2006; Bass & Murphy, 1991). Estimates ofunemployment among patients with somatization disorder range from36% to 83% (Allen et al., 2006; Smith et al., 1986a; Yutzy et al., 1995).Whether working outside their homes or not, these patients reportsubstantial functional impairment. Some investigators have found thatpatients with somatization disorder are bedridden for 27 days permonth (Katon et al., 1991; Smith et al., 1986a). Likewise, patientsdiagnosed with subsyndrome somatization or a functional somaticsyndrome report significant reductions in their productive activities(Allen, Gara, Escobar, Waitzkin, & Cohen-Silver, 2001; Bombardier &Buchwald, 1996; Drossman et al., 1993; Escobar, Golding, et al., 1987;Gureje, Simon, et al., 1997; Kroenke et al., 1997; Leong et al., 2003;Wolfe et al., 1997). In fact, Katon and colleagues (1991) found a linearrelationship between the number of unexplained symptoms and theseverity of functional impairment.

In addition to their physical complaints, most patients with soma-tization complain of psychiatric distress. As many as 80% of patientsmeeting criteria for somatization disorder, CFS, fibromyalgia, or IBSmeet criteria for another lifetime DSM Axis I disorder, usually an anxi-ety or mood disorder (Bass & Murphy, 1991; Clark et al., 1995; Epsteinet al., 1999; Ford, Miller, Eastwood, & Eastwood, 1987; Swartz, Blazer,et al., 1986). When investigators consider only current psychiatric diag-noses, rates of psychiatric comorbidity associated with somatization


syndromes are closer to 50% (Allen et al., 2006; Camilleri & Choi, 1997;Epstein et al., 1999; Simon & Von Korff, 1991). Also, overall severity ofpsychological distress, defined as the number of psychological symp-toms reported, correlates positively with the number of functional so-matic symptoms reported (Katon et al., 1991; Simon & Von Korff,1991).

Although the association between physical and psychologicalsymptoms has been clearly established, the mechanism explaining thisassociation has yet to be identified. Mechanisms that may be involvedin this relationship include the following:

1. A common physiological, cognitive, or personality disturbancemay underlie both functional somatic and psychological symp-toms.

2. Mood disturbances may increase the likelihood of somaticcomplaints by lowering pain thresholds and triggering symp-tom amplification.

3. Chronic, unresolved, disabling physical discomfort may createemotional distress.

Whatever the mechanism might be, the f

Documents

Robert L. Woolfolk, Lesley a. Allen-Treating Somatization_ a Cognitive-Behavioral Approach (2006)