eating disorders - Science Direct

6 downloads 0 Views 2MB Size Report
Correspondence should be addressed to William G. Johnson, PhD, .... psychopathology of the disorder rather than efforts to increase eating (e.g., Walsh,. 1995 ...
Clinical Psychology Review, Vol. 16, No. 6, pp. 457-478. 1996 Copyright 0 1996 Elswier Science Ltd Printed in the USA. Ail rights reserved 0272-7358/96 $15.00 + .OO

Pergamon

PI1 SO272-7358(96)00030-X

EATING DISORDERS: EFFICACY OF PHARMACOLOGICAL AND PSYCHOLOGICAL INTERVENTIONS William

G. Johnson, janice University

Y Tsoh, and Paula /. Vamado

of Mississippi Medical

Center

ABSTRACT. Eating disorders represent a spectrum of overlapping conditions that combine disturbances in eating with various fm of traditional psychopathology. Medications am oflen helpful in the management of the psychopathology associated with anorexia nervosa but no compound has been shown to assist weight gain reliably or alter other con?features of the condition. Contingency management and other behavior therae @nxedums are effective in pwwting weight gain in anmexia neroosa. The limited effectiveness of cognitive interventions for anmexia neruosa may be due to th.e complex of physical symptoms associated with knu body weight. Antid+ressants reduce binging and purging in bulimia neroosa and binge eating dish although this action appears to be ino@xnoknt of any antidep-essant effect. Cognitive/behavioral interventions also reduce binging and pqing, and compations of this therapy with medication indicate that psychotherapy alone is more effective than medication alone. Mmeovo; changes jn~ duced by cognitive/behavioral interventions endure longer than medication where higher relapse rates are common. Most studies also reveal no advantage of medication overcognitive/behavioral therapy alone in th reduction of bulimic symptoms. Copyright 0 1996 Elsevier Science Ltd

EATING DISORDERS represent a class of deviant behavior that has presented formidable challenges to both the diagnostic nomenclature and the broad range of therapies. Estimates of the prevalence of eating disorders are less than 1% for anorexia nexvosa (AN), l-3% for bulimia nervosa (BN), and between 2 and 5% for binge eating disorder (BED). While these estimates reflect the proportion of community populations satisfying formal diagnostic criteria, the prevalence of caloric restriction, binge eating, purging, and other pathogenic weight control behaviors (PWCBs) is much higher, ranging from 15 to 40% (Schlundt &Johnson, 1990, Spitzer et al., 1992). Eating disorders have profound health, economic, and personalsocial consequences. AN is associated with a significant mortality (5-15%), and BN is closely related to depression and various Correspondence should be addressed to William G. Johnson, PhD, Department of Psychiatry and Human Behavior, University of Mississippi Medical Center, 2500 North State Street, Jackson, MS 39216-4505. 457

458

W G. Johnson, J. I! Tsoh, and II J. Varnado

personality disorders. Patients who meet diagnostic criteria for AN and BN have major disruptions in their lives. Also, those with frequently occurring PWCBs experience decreased efficiency at work and school and disturbed interpersonal relationships as well as an overall diminished quality of life. Eating disorders are very complex conditions that fuse disturbances in eating with various forms of other, traditional psychopathology. They can be a humbling experience for clinician and researcher alike. Despite this complexity, progress in understanding the nature of eating disorders and their treatment is apparent. In this first section, the diagnostic criteria and core cognitive/behavioral characteristics of the three major eating disorders (AN, BN, BED) will be reviewed. The core features of these disorders are specific cognitions and behaviors that can be targets of thempeutic interventions. In later sections, the fundamentals of pharmacological and cognitive/behavioral interventions as they are applicable to eating disorders will be presented. Finally, studies evaluating the efficacy of these interventions will be considered. DIAGNOSTIC

CRITERIA

Anorexia Nervosa The formal Diagnostic and Statistical Manual of Mental Dim&m, 4th Edition (DSM-nlT; American Psychiatric Association [APA], 1994) criteria for the diagnosis of AN are the simultaneous presentation of a refusal to maintain an acceptable weight (85% of ideal weight), an intense fear of gaining weight, body image disturbance, and amenorrhea. This diagnosis is further classified into restricting and binge/purging types. The diierentiation of bulimic from restricting types of AN is the presence of binge eating and/or the use of purging such as self-induced vomiting and/or laxative abuse in the former. Bulimia Nervosa The DSM-IV criteria (APA, 1994) for BN include the presence of recurrent (at least 2 x per week for the previous 3 months) binge eating, behaviors to compensate for caloric intake, and selfevaluation overly influenced by body image and weight. The diagnostic criteria include two subtypes of BN which are differentiated on the basis of the type of compensatory behavior: purging and nonpurging. The purging-type displays self-induced vomiting, laxatives, diuretics or enemas, and the nonpurging type uses fasting and/or exercise as the primary method to compensate for caloric intake. Currently, little empirical data are available supporting these two types, and purgative behaviors occur in the overwhelming majority of the clinical presentations of BN (e.g., Mitchell, 1992; Mitchell, Hatsukami, Eckert, & Pyle, 1985). Binge Eating Disorder Due to its prevalence and distinction from other eating disorders, BED is the most recent addition to the eating disorders classification. However, BED was not included as a formal diagnositc category in DSM-IV but rather as one warranting further study (Spitier et al., 1992; Spitzer et al., 1993). Surveys estimate that from 22 to 55% of sub jects presenting for weight loss treatment have significant problems with binge eating (de Zwaan, Nutzinger, & Schoenbeck, 1992; Cormally, Black, Daston, & Rardin, 1982). The core feature of BED is binge eating in the absence of compensatory behaviors such as self-induced vomiting or laxative abuse. Briefly, binge eating episodes are those in which a large amount of food is eaten in a discrete period of time (e.g., within a 2-hour period) with a feeling of a lack of control over the eating episodes. The

Eating Disorders

459

binge eating must occur on average at least twice per week for a 6-month period. Finally, the binge eating does not occur during the course of BN or AN. Additional criteria for BED include rapid eating, eating until uncomfortably full, eating when not hungry, eating alone due to embarrassment, and feeling disgusted, depressed, or guilty after overeating. Eating disorders, as with other syndromes such as schizophrenia, are best considered as a spectrum of interrelated conditions with considerable overlap, and where patients move from one diagnostic category to another. So, although it is convenient to refer to AN, BN, and BED as separate entities, the empirical evidence strongly suggests that these three eating disorders are a spectrum of overlapping conditions (Andersen, 1983; Guiora, 1967). Core Characteristics of Eating Disorders The three eating disorders and their distribution over core cognitive/behavioral characteristics are displayed in Table 1. As indicated, the fear of obesity is uniformly strong over each disorder. Dietary/eating habit control is most prominent in AN-restrictor type, less obvious in the AN-bulimic type, and negligible in BED where most patients are obese. The compensatory behaviors of exercise, self-induced vomiting, and laxative abuse vary over the disorders. Obligatory exercise is most common in AN-restrictor type and BN, and it is rare in BED. Both self-induced vomiting and laxative abuse are most salient in BN, present in AN-bulimic type, and seen occasionally in BED. Body image disturbance is a unique and perplexing characteristic of eating disorders that combines both image distortion and dissatisfaction (Schlundt & Johnson, 1990). Distortion refers to perceptual errors wherein the body is experienced as much larger than its actual size. These distortions are typically limited to perceptions of one’s own body, and are most common in AN. In contrast, body dissatisfaction refers to discontent with the total body or specific parts, and is more common in BN and BED. The variable nature of body dissatisfaction in AN is dependent on body weight with satisfaction more likely when an emaciated body shape/ low weight is achieved.

TABLE 1. Cognitive and Behavioral Features of Eating Disorders

Eating Disorders Anorexia Nervosa Cognitive and

Behavioral Features Fear of obesity Dietary/eating habit control Compensatory behavior Exercise Self-induced vomiting Laxative abuse Body image Distortion Dissatisfaction

Restricting

Bulimic

Binge Bulimia Nervosa Eating Disorder

strong

strong

strong

strong

strong

variable

variable

slight

strong slight slight

strong strong strong

variable strong strong

slight slight slight

strong variable

variable variable

slight strong

slight strong

460

W! G. Johnson, J. I: Tsoh, and tl J. Varnado

These core characteristics of eating disorders represent the behavioral/cognitive/affective complex of variables that are the focus of most therapeutic interventions. As shall be obvious, pharmacological and cognitive-behavioral therapy (CBT) interventions offer somewhat different approaches to alter these core characteristics. CBT interventions directly attack these core characteristics, while pharmacological interventions attempt to change their neurophysiological substrates or to modify the psychopathology associated with them. PHARMACOTHERAPY

FOR EATING DISORDERS

A variety of external/environmental and internal/neurophysiological influences act in concert to affect eating (Hobel & Hernandez, 1993; Schlundt &Johnson, 1990). These influences may include food and nutrient deprivation, stress, energy depletion via work/exercise, aspects of the social environment such as the presence of others, and various psychological states (e.g., feelings of rejection, loneliness, and boredom). It is not unusual for the same set of internal or external conditions to produce either an increase or a decrease in caloric intake depending on prior history. On a phenomenological level, for example, feeling low in energy, lightheadedness, and an empty sensation are linked with hunger and the urge to eat, yet these same internal states can also be associated with a desire to restrict intake. In contrast, fullness is generally related to satiety, but it is not unusual for eating to continue unabated after a full meal (Johnson, Schlundt, Barkley, Carr-Nangle, 8c Engler, 1995). In terms of the clinical problems associated with eating disorders, medications are needed that increase and decrease caloric intake. In the case of AN, for example, a chronic state of hunger is apparent but typically denied in the early phases of the condition, yet with increasing chronicity and weight loss, a true anorexia, or lack of appetite, develops. Conversely, with BED and obesity, where a presumably abundant supply of energy would seemingly decrease the desire to eat, yet binge eating is common. Paradoxically, it appears that greater deviations at both extremes from a normal height/weight ratio are associated with biobehavioral mechanisms that make eating problematic - either too much in the case of BED/obesity or too little with AN. Pharmacological influences on Eating Drug influences on eating behavior can be quite specific, ranging from affecting the onset of both hunger and satiety, to altering nutrient selection and the hedonic value of food, meal interval, frequency of meals, rate of eating, and energy intake (Blundell & Hill, 1993). Drugs act on various neurophysiological substrates to produce these behavioral effects, and studies have focused on the monoamines such as norepinephrine, serotonin, and dopamine, as well as opioid and gut-brain peptides and various hormones. As summarized by Leibowitz (1988)) eating is stimulated or inhibited by the action of these neurotransmitters, peptides, and hormones depending on their site of action in the hypothalamus. Drugs that alter their function can be expected to affect eating. One result of this research has been the identification of a number of overlapping neurophysiological systems involved in eating, which is undoubtedly related to its importance for survival. Also, as the data on the sideeffects of psychotropic compounds attest, many of these neurophysiological systems also serve as substrates for other behaviors. So, the effects of a compound that alters eating may extend to other neurophysiological substrates and behaviors. In addition to the neurophysiological influences on eating, social, and psychological events such as the social situation, location, time of day, and mood also exert

Eating Lkorders

461

strong control over eating behavior. In fact, a binge eating episode, rather than resulting in a decrease in subsequent binging as a result of satiety, actually increases the probability that the next intake will be a binge (Johnson et al., 1995). Pharmacological Influences on Psychopathology Depression, obsessive-compulsive disorder, anxiety disorders, and various personality disorders are the most common conditions that covary with eating disorders. A wide variety of drugs have been used under the assumption that the eating disorder is functionally related to the associated psychopathology. For example, Hudson, Pope, Yurgelun-Todd, Jonas, and Frankenburg (1987) suggested that BN is a variant of depression, while AN has been linked with obsessivecompulsive disorder (Halmi et al., 1991). The medication targets the symptoms of psychopathology, and improvements are then presumed to decrease PWCBs. Interestingly, some drugs intended to affect the associated psychopathology of eating disorders have had a significant impact on eating behavior, which is independent of its influence on psychopathology. This finding will be apparent in the discussion of antidepressants and BN. EVALUATING

PHARMACOLOGICAL

INTERVENTIONS

FOR ANOREXIA NERVOSA

With few exceptions, pharmacotherapy for AN is primarily directed at the associated psychopathology of the disorder rather than efforts to increase eating (e.g., Walsh, 1995; Walsh & Delvin, 1992). Unfortunately, all studies evaluating therapy for AN including pharmacotherapy have been hampered by small sample sizes and poor methodological designs. Neurolep tics Chlorpromazine was used initially for AN to disrupt the delusional-like cognitions toward body and food, but it has not been found effective in inducing cognitive changes or weight gain (Hoffman 8c Halmi, 1993). Other neuroleptics such as pimozide and sulpiride have been used, and controlled studies examining these drugs in the context of behavior therapy have failed to show a significant effect as compared to placebo conditions (Vandereycken, 1984; Vandereycken 8c Pierloot, 1982). In reviewing drug therapy for AN, researchers do not recommend neuroleptics due to their lack of efficacy, side-effect profiles, and the availability of other compounds (Hoffman & Halmi, 1993; Judd, Norman, 8c Burrows, 1987; Vandereycken, 1987). Antidepressants A variety of antidepressants, including tricyclics (TCA) and monoamine oxidase inhibitors (MAOI) have been used to treat AN based on their ability to decrease depressive symptoms and, in the case of the TCAs, to increase caloric intake. For the TCAs, amitriptyline and clomipramine have been evaluated in controlled studies with AN. When compared to placebo in a 5-week double-blind trial within the context of a behavior modification program, amitriptyline showed no signficant effect, and a number of sideeffects were reported (Biederman et al., 1985). In a largescale study, amitriptyline, cyproheptadine (CYP) and placebo were compared over a 4week period. Amitriptyline was found to have a marginal effect over placebo, yet no different than CYP in weight gain or on psychopathology measures. Similar to other studies, significant side-effects of amitriptyline were noted (Halmi, Eckert, LaDu, & Cohen, 1986).

W G. Johnson, J. I! Tsoh, and !?J, Varnado

462

The effect of clomipramine in conjunction with a behavioral intervention was examined in 16 AN inpatients who were followed at 1 and 4 years. No differences were found between the drug and placebo conditions, but clomipramine did produce greater increases in appetite, hunger, and calorie consumption during the first 2 months of treatment (Crisp, Lacey, & Crutchfield, 1987; Lacey 8c Crisp, 1980). MAOIs (e.g., phenelzine) and trazodone have been studied in open clinical trials without sucess (Hudson, Pope, Jonas, & Yurgelun-Todd, 1985). More recently, the use of fluoxetine on refractory AN inpatients with significant depression has been explored (Gwirtsman, Cuze, Yager, 8c Gainsley, 1990). Preliminary results were encouraging with most patients showing improvement in weight gain and depressive symptoms. Because of its relatively benign side-effect profile, further studies of fluoxetine are warranted. Appetite Stimulants Given its antagonistic effect on serotonin, CYP has been used to increase caloric intake and to promote weight gain. In an initial study, CYP induced minimal weight gain in one third of outpatients (Vigersky & Loriaux, 1977). In a larger study, CYP was compared to placebo on weight gain among 81 AN inpatients using a 2 (CYP vs. placebo) x 2 (behavioral modification vs. control) design. There was a small nonsignificant difference favoring the drug condition after 35 days. As mentioned above, CYP was compared to amitriptyline and placebo with minimal effects on weight gain and depression over placebo and no difference when compared to amitriptyline (Halmi et al., 1986). A purported appetite stimulant, tetrahydrocannabionol (THC), which is the active metabolite of marijuana, was compared to diazepam in a double-blind crossover study over a 4-week period, but was not found effective in producing weight gain (Gross et al., 1983). Miscellaneous

Lithium combined with behavior therapy was tested on 14 patients in a double-blind controlled study over a 28day protocol (Gross, Ebert, & Faden, 1981). Although a marginal effect on weight gain was reported during the third and fourth week of treatment, the use of lithium in medically compromised AN is not advisable due to its sideeffects. Other drugs such as clonidine, an adrenergic agonist (Casper, Schlemmer, & Javaid, 1986), and metoclopramide, an antiemetic agent (Moldofsky, Jeuniewic, & Garfinkel, 1977), were tested using small sample, placebo-controlled crossover studies but were not found useful. Anti-anxiety drugs, such as lorazepam (Andersen, 1987) have also been suggested for AN patients but no empirical data are available. SUMMARY

OF PHARMACOTHERAPY

FOR ANOREXIA

NERVOSA

Our current knowledge of pharmacotherapy of anorexia nervosa is limited by a handful of controlled studies, most of which were conducted within the context of an inpatient behavior modification or psychotherapy program (Hoffman & Halmi, 1993; Walsh, 1995). Most studies examining treatment efficacy of pharmacotherapy focused primarily on weight gain with several reports including measures of psychopathology (e.g., depression). To date, no medication has been shown to change eating behavior reliably, assist weight gain or modify its fear, or alter body image disturbance. EVALUATING

PHARMACOTHERAPY

FOR BULIMIA

NERVOSA

A number of drugs have been evaluated for BN, and the consensus is that antidepressants are the most efficacious and safe (e.g., APA, 1993; Fait-burn & Hay, 1992; Hoffman

Eating Lkomhs

463

& Halmi, 1993; Walsh, 1991, 1995). Other drugs including carbamazepine (Kaplan, Garfinkel, Darby, 8c Garner, 1983), lithium (Hsu, 1984; Hsu, Clement, Santhouse, & Ju, 1991), and appetite suppressants such as fenflummine and naltrexone (Hoffman & Halmi, 1993) appear to be less effective and are generally not recommended due to their side-effects and potential toxicity (APA, 1993; Fairbum & Hay, 1992). Tricykc Antidepressants In an initial report, imipramine was compared to a placebo condition for 6 weeks. Over 85% of subjects on imipramine (vs. 10% in placebo) exhibited at least a 50% decrease in binge eating (Pope, Hudson, Jonas, & Yurgelun-Todd, 1983). When imipmmine was given to all subjects after the double-blind phase, 90% reported at least a 50% reduction in binge eating and 35% were in remission during follow-up periods ranging from 1 to 8 months. Imipramine was also associated with decreased depression and preoccupation with food. Increased global improvement and a minimal dropout-rate (9.5%) due to sideeffects were also noted. A subsequent 16-week double-blind controlled trial of imipramine found over 70% of patients had reduced binging and purging at 16 weeks with a remission rate of 30% (Agras, Dorian, Rirkley, Arnow, & Bachman, 1987). Close to half of the subjects reported less depression at 6 and 16 weeks (46% and 51%, respectively), with a significant difference from the placebo condition evident at 6 weeks only. While both studies indicate the efficacy of imipmmine on binge eating, it is unclear whether the reduction in binging and purging is associated with the decline in depression. The effectiveness of imiptamine was also examined in a comparison with naltrexone among nonpurging binge eaters. When compared to naltrexone and placebo in 22 bulimics (with the majority using laxatives) and 33 obese binge eaters, no significant reduction in binge frequency was found in either medication condition (Alger, Schwaberg, Bigaouette, Michalek, & Howard, 1991). Desipramine, a metabolite of imipmmine, was compared to a placebo in 22 nondepressed subjects with BN (Hughes, Wells, Cunningham, & Ilstrup, 1986). After 6 weeks, over 90% of subjects receiving desipramine (vs. 19% in placebo) showed a decline in binge eating, which was noted in the first week of treatment and maintained at a l-month follow-up. Subjects in the control group were then given desipramine with 68% ceasing binging at 10 weeks. Another double-blind crossover study examined the effect of desipramine in a 6-week outpatient program (Barlow, Blouin, Blouin, & Perez, 1988). Close to half of the subjects (46.8%) reported a reduction in binging (vs. 2.4% in placebo) with one subject achieving abstinence. The antibulimic effect of the drug was again noted during the first week of treatment. Desipramine also has been reported successful in treating nonpurging bulimics. McCann and Agras (1990) reported a remission rate of 60% with desipramine (vs. 15% in placebo) among 23 nonpurging bulimics over a 12-week trial. However, in both of these studies (Barlow et al., 1988; McCann & Agras, 1990), one-third of the patients dropped out due to sideeffects of desipramine. Furthermore, treatment benefits disap peared when medication was discontinued (cf. Hughes et al., 1986). Desipmmine was not associated with decreases in depression or other psychopathology in these studies. Monoamine Oxidase inhibitors The first controlled trial of MAOIs (Walsh, Stewart, Roose, Gladis, & Glassman, 1984) evaluated phenelzine in 20 BN patients, with 12 patients having concomitant dysthymia or major depression, over 8 weeks. All patients receiving phenelzine reduced

464

M! G. Johnson, J. I: Tsoh, and F!J. Varnado

binging by at least 50% (vs. 2 of 11 in placebo) with many reporting significant sideeffects. Walsh et al. (1988) continued the study with 50 additional BN patients using the same protocol and found that phenelzine significantly reduced binging as compared to placebo (65% vs. 5%) and yielded a higher abstinent rate (35% vs. 4%). Unfortunately, 24 out of 27 responders discontinued phenelzine due to sideeffects. Because of the dietary restrictions required for MAOIs, they are recommended only for patients who can reliably maintain a tyramine-free diet (Pope & Hudson, 1986; Wells & Logan, 1987), which may be impractical when treating BN. More recently, brofaromine was tested in an &week double-blind controlled study but showed nonsignificant effects (Kennedy et al., 1993). Selertive Serotonin Reuptake Inhibitors (SSRl) After the introduction of fluoxetine in 1988 for the treatment of depression (Herzog & Sacks, 1993), SSRIs were quickly applied to eating disorders and now appear to be the most commonly prescribed antidepressant for BN (Abbott & Mitchell, 1993). The popularity of SSRIs such as fluoxetine, paroxetine, and sertraline are no doubt related to their modest side-effect profile, and their effectiveness in the treatment of depression. In addition, the theoretical prospects of appetite suppression on high doses of SSRIs was suggested as a desirable property that might make the medication more appealing to BN patients (e.g., Fluoxetine Bulimia Nervosa Collaborative Study Group [FBNCSG], 1992; Walsh, 1991). Following preliminary open trials of fluoxetine, a large multicentered study by the FBNCSG (1992) compared the efficacy of 60 mg vs. 20 mg in a placebwontrolled 8 week trial on 387 BN patients. The higher dose (60 mg) of fluoxetine was clearly superior to the 20 mg dose and placebo with median reduction rates of 67%, 45%, and 33% for binge eating and 57%, 45%, and 26% for vomiting, respectively. Dropout rates related to side-effects were minima1 across all groups. Compliance rates, defined as taking at least 80% of the medication, were similar across all conditions ranging from 80 to 87%. In another large study, Goldbloom and Olmsted (1993) tested fluoxetine (20 vs. 60 mg) in a placebo-controlled 8-week study on 382 BN outpatients. The two doses yielded a similar response with reductions on standardized self-report measures of eating disorders symptoms. Atypical Antidepressants Other atypical antidepressants such as mainserain, trazodone, and buproprion have been evaluated in controlled studies for their efficacy in treating BN. However, research findings do not support the use of these medications given the options of other more efficacious compounds (Pope, Keck, McElroy, & Hudson, 1989; Sabine, Yonace, Farrington, Barratt, & Wakeling, 1983) or, as with buproprion, their severe side-effects (Horne et al., 1988). SUMMARY

OF PHARMACOLOGY

FOR BULIMIA

NERVOSA

In general, most antidepressants appear to reduce bulimic symptoms. A large majority of patients decrease binge eating by at least 50% with a corresponding decrement in vomiting, and remission rates ranged from one-third on imipramine to about twothirds of patients on desipramine or fluoxetine over a minimal period of 6-10 weeks. The relative impact of these medications on core characteristics of BN compared to changes in depression is uncertain. Interestingly, while earlier studies with desipramine

Eating Disurders

465

reported no influence on body image disturbance (McCann & Agras, 1990; Barlow et al., 1988) more recent research with SSRIs have reported improved body satisfaction (FBNCSG, 1992; Goldbloom & Olmsted, 1993). Among the antidepressants, SSRIs are preferred over TCAs due to their lower sideeffect profile. Given these positive results, a number of unanswered questions remain regarding the use of medication for BN (e.g., Abbot & Mitchell, 1993; Hoffman & Halmi, 1993; Walsh, 1995; Walsh 8c Delvin, 1992). Perhaps the most important question regards the mechanism of action of antidepressants given the lack of covariation between the reduction in bulimic symptoms and the antidepressive effect. In fact, the action of TC& and SSRIs should not be apparent until 3-4 weeks of medication; however, as noted, several studies report reductions in bulimic symptoms during the first 3 weeks of therapy. It does not appear then, that the antibulimic effect is related to the relief of subjective depression, as a number of studies show no relationship between levels of depression and BN symptoms. Also, compliance with the prescribed medication and the relationship of serum levels of drugs and treatment efficacy or antibulimic effects are generally not evaluated. Thus, while antidepressants are effective for BN symptoms as compared to other drugs, their mechanisms of action are unclear. Furthermore, there is no clear consensus regarding dosage. Most studies test one standard dose (except for the multicenter fluoxetine study), while other studies adjusted the dosage. Another important unresolved issue is the time course of active medication, as relapse rates seem especially high after medication is discontinued. For example, Pope, Hudson, Jonas, and Yurgelun-Todd (1985) reported a remission rate of 73% in those followed (8 of 11). Most subjects had to remain on the medication to maintain the treatment effect with only three able to discontinue successfully. In addition, there has been limited examination of the relationship between efficacy and length of drug treatment (e.g., Agras et al., 1992). Iastl y, most investigations of pharmacothempy on BN focus on patients who compensate calorie intake through self-induced vomiting, and thus the genet-&ability of these findings remains unknown for nonpurging bulimics or BED. COGNITIVE/BEHAVIORAL

THERAPY

The integration of cognitive psychology into social learning theory provided a framework merging cognitive and behavioral therapeutic techniques (Schlundt &Johnson, 1990). A variety of CBT interventions have been developed for the treatment of eating disorders. In general, CBT targets both the cognitive distortions and the behaviors associated with disordered eating (Fairburn, 1981). The CBT model proposes that eating disorder patients believe that their weight and shape are of fundamental importance and that both must be kept under strict control (Fait-bum, 1985). The goals of CBT are the modification of abnormal cognitions regarding the personal significance of body shape and weight and the replacement of dysfunctional dietary restraint and purging with more normal eating and activity patterns (Schlundt &Johnson, 1990). CBT includes a variety of procedures: the presentation of the cognitive view of the dis order, self-monitoring of relevant thoughts and behaviors, education, cognitive restructuring, the use of self-control measures to establish regular meals and activity, and the introduction of “feared” and avoided foods into the diet (Fairburn, 1981). COMPONENTS

OF CBT

Eating disorders are multifaceted, and specific interventions are often necessary to target different core features of the disorders (see Table 1). The following interventions are commonly used in the CBT in an integrated manner. Detailed descriptions

W G. Johnson, J. I! Tsoh, and E!J. Varnado

466

of the procedures techniques Wilson

and empirical

can be found

(1993))

studies related

in Agras

and Schlundt

(1991),

to each of the following

Fairburn

and Johnson

(1985),

intervention

Fairburn,

Marcus,

and

(1990).

Self-Monitoring Self-monitoring

refers to the recording

ated events on a specific important

relationships

affective

between

antecedents

by patients

time or event schedule. target behaviors

(Schlundt,

of a target behavior

Self-monitoring

and its associ-

records

and their situational,

often reveal

cognitive,

and

1989).

Exposure With Response Prevention Exposure

with Response

ing (Rosen

Prevention

8c Leitenberg,

the fear of weight

1982)

gain subsequent

feared foods and experiences and urge to vomit. Purging discomfort

(ERP)

after eating

focuses

to eating.

the feelings is prevented

vomit-

that reduces

the patient

eats specific

bloatedness,

and the patient

fear of weight gain,

is taught that the feelings

over time.

and Exposure With Response Prevention

Temptation

and Exposure Similar

with Response

Prevention

to ERP, prevention

of binge

(TERP) eating

is a method

is achieved

to eliminate

through

repeat-

ed exposure to the situational, cognitive, and affective cues that elicit craving binge eating without allowing the binging response to occur. Thus, the functional between

the triggering

association

between

(Johnson,

Cot-r&an,

of

and that the fear of weight gain and

Temptation

binge eating.

self-induced

as a behavior

In practice,

of fullness,

a large meal are normal,

urge to purge subside

on decreasing

which is conceptualized

events and binging

are weakened

those cues and more appropriate

through coping

extinction

strategies

and link

while the

is established

& Mayo, 1987).

Energy Balance Training Energy

Balance

eating

disorders

Training

energy regulation patients

with eating

ing, and self-induced Cognitive Cognitive ceptions Substitute

that promote

model

that considers

to manage

The goal of EBT is to develop

long-term

are ineffective

such as skipping vomiting,

skills required

weight management.

in weight

meals,

regulation

restrictive

which perpetuate

dieting,

weight and healthful

eat-

The majority of as they engage fasting,

binge

in eat-

their problems.

Interventions interventions that contribute thoughts

al adjustment, depression

et ai., 1987).

disorders

practices

is based on the energy balance

in self-management

(Johnson

ing and activity patterns selfdefeating

(EBT)

a deficit

target cognitive

are developed

and interpersonal

(Beck,

distortions,

to the maintenance

that promote relations

overgeneralizations,

of pathogenic a healthful

similar

and misper-

weight control lifestyle,

to the interventions

behaviors.

positive personproposed

for

1976).

Personal Social Problem Solving Personal problem

Social Problem Solving (PSPS) teaches patients situations that typically lead to negative moods

to deal more effectively with and pathogenic weight con-

Eating Lhbra!m

467

trol behaviors (Johnson, Schlundt, & Jarrell, 1986). Problem solving has five general steps including problem definition, generation of alternative coping behaviors, evaluation of outcomes, decision making, and verification. Relaxation Training Relaxation training includes teaching progressive muscle relaxation and visual imagery to cope with anxiety that might provoke pathogenic weight control behaviors. Body Image Therapy Body image therapy focuses on the following body image problems: misperception of and dissatisfaction with one’s body, irrational cognitive beliefs about the body, and ineffective ways of coping with problem situations that elicit these problems. Awareness is promoted by identifying problematic thoughts and feelings, then pro viding feedback from self-report measures of body image, body fat measures, and viewing videotapes. The gamut of CBT interventions described earlier are used to identify and challenge beliefs and values related to body image as well as to substitute distorted, unrealistic views with more constructive perceptions and interpretations. Equally, if not more important, are efforts to decrease the importance of personal appearance and physical attractiveness. Relapse Prevention Relapse prevention aims at increasing awareness of the possibility of relapse and preparing for situations that may interfere with the maintenance of progress. Relapse prevention involves identifying high-risk situations which may lead to problem behaviors and generating plans and strategies to cope with these situations in order to minimize the chance of a relapse by using a variety of CBT interventions. EVALUATING

CBT FOR ANOREXIA

NERVOSA

The initial applications of CBT to AN began with explicit behavioral techniques such as contingency management and other operant procedures on inpatient units. Currently, the majority of programs for AN either explicitly or implicitly use such treatment techniques to normalize eating and promote weight gain (Halmi, 1985). The development of cognitive therapy (Beck, 1976), and the recognition that the pathology of eating disorders extended beyond a disturbance in eating behavior per se, hastened the integration of cognitive components with behavior therapy (BT). The emotional and cognitive disturbances observed in eating disorders often exacerbate and maintain the disturbed eating patterns (Schlundt &Johnson, 1990). While the goal of BT with AN is to increase eating and weight gain, cognitive interventions seek to replace thoughts that exacerbate and maintain PWCBs and replace them with ones that promote healthful eating and activity patterns. Given the scarcity of cases and their medical complications, research studies on AN with several exceptions are limited to singlecase designs that focus on increasing eating and weight (Bemis, 1987). In a series of singlecase experiments, positive and negative reinforcement, informational feedback, and the size of meals, among other variables have contributed to increased caloric intake and weight gain (Agras, Barlow, Chapin, Abel, & Leitenberg, 1974; Bianco, 1972; Blinder, Freeman, & Stunkard, 1976; Garfinkel, Kline, & Stancer, 1973). Also, reinforcement during hospitalization and contracting for readmission to the hospital if target weight was not maintained have been

W G. Johnxm,J

468

Y Tsoh, and PJ

Varnado

effective in maintaining target weight (Poole 8c Sanson-Young, 1978). One controlled study of operant procedures found that a behavioral group gained slightly but not significantly more weight than a ward milieu therapy program (Eckert, Goldberg, Halmi, Casper, 8c Davis, 1979). The addition of cognitive techniques similar to treatments proposed for BN have been examined with AN with mixed results (Cooper & Fairburn, 1984). CBT, BT and a control group were compared in a randomized trial with eight outpatients. While the interventions produced similar clinical improvements, the small number of patients and pretreatment differences limit definitive conclusions (Channon, De Silva, Hemsley, & Perkins, 1989). SUMMARY

OF CBT FOR ANOREXIA NERVOSA

As noted previously, the severity of medical complications in AN often limits the use of wait-list or no-treatment control groups. Thus, with several exceptions, the majority of studies for AN have used single-case designs that have evaluated specific BT interventions. While these data have demonstrated the efficacy of various behavioral interventions, the results of investigations of CBT have been less convincing with methodological problems confounding their interpretation. Other studies have examined the influence of family therapy (le Grange, Eisler, Dare, & Russell, 1992; Russell, Szmukler, Dare, & Eisler, 1987), contrasted behavioral family systems therapy with ego-oriented family therapy (Robin, Siegel, Koepke, Moye, & Tice, 1994), and compared cognitive-analytic therapy with an educational approach (Treasure, Todd, Brolly, Tiller, Nehmed, & Denman, 1995). While the results of these studies have been promising, they usually compare two interventions without a no-treatment control group over a small number of subjects for a short period of time. Thus, their findings must be viewed with caution. EVALUATING

CBT FOR BULIMIA

NERVOSA

CBT is directed specifically toward the abnormal eating patterns and cog&ions dis played by bulimics. The interventions include education about the dangers of vomiting and purgative abuse, food monitoring and dietary management, the identification of antecedents to loss of control, problem-solving, cognitive restructuring, ERP, relapse prevention, and body image therapy (Fair-bum, 1981; Schlundt & Johnson, 1990). Initially, CBT was examined in uncontrolled singlecase studies to determine its promise in the treatment of BN. For example, a combination of interventions (ERP, EBT, and PSPS) was used in a 16-week treatment program (Johnson et al., 1986). Eight of 12 subjects completed treatment and decreased purging by 90% with significant decreases also noted on measures of social anxiety and depression. A second generation of controlled studies has compared CBT, BT alone, and with wait-list and no-treatment conditions. CBT Compared

to BT

Several studies have compared CBT to BT interventions, and, in general, these studies find no differences in the reduction of binging and purging between the CBT and BT interventions. However, decreased concern and preoccupation with dieting, greater confidence in identifying sensations of hunger and satiety have been noted for CBT (Freeman, Sinclair, Turnbull, & Annandale, 1985; Wolf & Crowther, 1992; Yates 8c Sambrillo, 1984). CBT has been shown to be superior to BT in other comparisons (e.g., Fairburn, Jones, Peveler, Hope, & O’Connor, 1993; Thackwray, Smith, Bodfish, & Meyers, 1993).

Eating Disorders

469

Psychoeduca tional lnterven tions for Bulima Nervosa Several studies describe psychoeducational programs that combine the presentation of information about eating disorders, nutrition, and eating habits with group support. These programs, while not explicitly CBT interventions, do utilize many CBT techniques such as goal setting, cognitive restructuring, and EBT. They report substantial reductions (e.g., 70%) in binge/purge episodes with maintenance in the follow-up phase as well as improvements in depression and self-esteem. While recognizing these overall clinical improvements, most subjects continued to purge, and gains over a control group were minimal (Connors, Johnson, & Stuckey, 1984, Wolchik, Weiss, 8c Katzman, 1986). CBT and ERP The additive effects of ERP to CBT have been investigated in several studies. Cognitive restructuring plus ERP was more effective than cognitive restructuring alone in reducing binging and vomiting with treatment gains maintained at l-year follow-up (Fairbum, Kirk, O’Connor, 8c Cooper, 1986, Wilson, Rossiter, Rleifield, 8c Lindholm, 1986). In another comparison of ERP over multiple settings with and without CBT, all active treatments showed significant improvement on a variety of dimensions with no significant differences across conditions. However, ERP was more effective on decreasing BN symptoms in a test meal (Leitenberg, Rosen, Gross, Nudelman, 8c Vara, 1988). One study found that the inclusion of ERP did not enhance treatment gains of CBT (Agras, Schneider, Arnow, Raebum, & Telch, 1989). In this study ERP was administered concordant with or following CBT. The failure to find an effect for ERP may be related to the timing of the interventions as a previous report indicates that ERP is not successful unless it is delivered prior to efforts at normalizing eating (Johnson et al., 1986). CBT Versus Wait-list and Psychotherapy Controls In studies comparing CBT with a wait-list control, the CBT group displayed signiflcandy decreased bulimic episodes and maintained treatment gains at several months (Lacey, 1983; Lee 8c Rush, 1986). CBT has also been compared to nondirective/sup portive therapy, and results at a 3month follow-up indicated that over one-third of the CBT group were abstinent compared to 11% of the nondirective group with decreases noted in associated psychopathology for both groups (Kirkley, Schneider, Agras, & Bachman, 1985). In a comparison of CBT and a short-term focal therapy, both conditions produced significant improvements in BN symptoms, however subjects treated with CBT showed greater improvements in general psychopathology and social adjustment (Fairburn et al., 1986). INTERPERSONAL THERAPY (IPT) FOR BULIMIA NERVOSA: EFFICACY AND COMPARISON TO CBT Much like CBT, IPT was first introduced as a treatment for depression (Kleiman, Weissman, Rounsaville, SCChevron, 1984). In its application to eating disorders, it targets personal stress and current interpersonal relationships rather than dietary issues or body shape and weight. The rationale given to patients for the use of IPT is that binge eating often begins in the context of an unsatisfactory interpersonal situation and that anxiety and depression emanating from the interpersonal interchanges often trigger binge eating, leading to a loss of control over food intake (Agras, 1991). Various forms of CBT have been compared to BT and IPT in the treatment of BN, and

470

W G. Johnson, J. K Tsoh, and PJ

Varnao!u

both CBT and IPT achieved equivalent effects in the reduction of binge eating and vomiting with over 90% reductions in both symptoms maintained at a l-year follow-up (Fairburn, Jones, Peveler, Hope, & O’Connor, 1993). However, CBT and BT achieved treatment gains faster than IPT, which also had a higher dropout rate. The authors acknowledged that further research is needed to determine the efficacy of IPT as a treatment for BN (Fairburn et al., 1993; Jones, Peveler, Hope, 8c Fairbum, 1993). The long-term outcome for CBT, BT, and IPT has also been examined, and over half of the subjects in the active interventions were relatively symptom-free and received no diagnosis at follow-up (Fairburn et al., 1995). Of these subjects, 82% still had no bulimic episodes at a l-year follow-up and 71% remained abstinent at longerterm follow-up. Subjects in the CBT and IPT conditions fared better than the subjects in the BT conditions with IPT subjects twice as likely and CBT subjects three times as likely to be in remission than those treated with BT (Fairbum et al., 1995). With its focus on interpersonal events as triggers of bulimic symptoms, the uniqueness of IPT from CBT is procedural rather than conceptual. For example, the assessment conducted prior to CBT typically includes an analysis of the antecedents to dysfunctional cognitions and behaviors (Johnson et al., 1987; Schlundt, 1989). The implementation of personal social problem solving, for example, concentrates on the gamut of antecedents to bulimic symptoms, whether interpersonal or otherwise. As such, IPT should be considered as a form of CBT with a focus on interpersonal antecedents and not a distinct form of therapy. SUMMARY

OF RESEARCH ON CBT FOR BULIMIA

NERVOSA

Overall, CBT has been shown to be superior to no treatment and waiting list control groups (Agras et al., 1989; Freeman, Barry, Tumbull, & Henderson, 1988; Lacey, 1983; Lee & Rush, 1986; Leitenberg et al., 1988; Wolf & Crowther, 1992). Also, the studies reviewed indicate that CBT is superior to behavior change procedures alone including ERP and IPT (Fairburn, Jones et al., 1993; Leitenberg et al., 1988; Thackwray et al., 1993). In a review, Agras (1991) concluded that CBT led to full recovery of BN in over 50% of cases, and follow-up data suggests that CBT is durable with 40% of patients being abstinent at 1 year (Fairburn, 1985). It is important to note that CBT programs have varied in content, and that only a few studies have included rigorous manipulation checks or obtained credibility ratings. Also, the prime outcome variables have been binging and purging, and few studies have examined use of laxative or other purgatives, excessive exercise, diet, or body image. While several studies have included follow-up asessment, this has not been standard. EVALUATING

CBT FOR BINGE EATING DISORDER

Treatment of BED typically consists of two complimentary objectives: weight management and the reduction of binge eating. Treatments directed at weight reduction are indicated given that the majority of BED patients are obese. Weight reduction programs typically include many of the CBT interventions previously discussed. Other treatment strategies have been directed toward the cessation of binge eating. Although these objectives are similar, the literature has largely addressed them seperately (Johnson & Torgrud, 1996). Weight Management

Early reports indicated that obese binge eaters lost significantly less weight than nonbinge eaters in behavioral treatment programs (Keefe, Wyshogrod, Weinberger, &

EatingDisorders

471

Agras, 1984). Furthermore, dropout and relapse rates appear to be higher for obese individuals with BED (Keefe et al., 1984; Marcus, Wing, & Hopkins, 1988; Yanovski, 1993); however, a recent study has not confirmed these results (Laporte, 1992). CBT for Binge Eating in Binge Eating Disorder Binge eating can be effectively treated with CBT with abstinence rates from 50 to 80% (Smith, Marcus, & Kaye, 1992; Telch, Agras, Rossiter, Wilfley, & Kenardy, 1990), and relapse rates lower than pharmacological interventions (cf. McCann & Agras, 1990). Treatment of binge eating alone does not result in weight loss, and Agras (1993) has suggested that weight loss should become a focus of treatment only after binge eating has been addressed. IPT for Binge Eating Disorder CBT, IPT, and a wait-list control were compared for the treatment of nonpurging BN (Wilfley et al., 1993). CBT focused on the elimination of binge eating rather than weight reduction, and IPT targeted current interpersonal relationships with a rationale provided to subjects that binge eating was a response to interpersonal disturbances and consequent negative moods. Both CBT and IPT were effective in the reduction of binge eating, and treatment gains persisted at G-month and l-year followups, while no effects were found for weight loss. One potential treatment strategy is the use of IPT after an unsuccessful trial of CBT. This issue was examined in BED with subjects who did not stop binge eating after 12 weeks of group CBT (Ag r-as et al., 1995). Fifty-five percent of the subjects improved with CBT, and IPT led to no further improvement in the remainder. PHARMACOTHERAPY

VERSUS PSYCHOTHERAPY

Several studies have compared pharmacological and psychotherapeutic interventions and their combination. These investigations have used various antidepressants and CBT with BN. Imipramine has been compared to psychotherapy in a randomized trial contrasting group psychotherapy with placebo or imipramine, imipramine alone, and placebo only for 12 weeks (Mitchell et al., 1990). The active interventions yielded significant improvement in bulimic symptoms compared to the placebo only condition, and subjects receiving imipramine alone showed a comparable reduction in binging (64%) and remission (16%) to previous trials with imipramine. Psychotherapy was superior to the drug conditions in the reduction of binging and purging frequencies, and the combined psychotherapy and drug condition offered no advantage over psychotherapy alone. As expected, dropout rates in both drug conditions (with or without psychotherapy; 25% and 43%, respectively) due to side-effects were higher than in psychotherapy-alone condition. Relapse was also much higher (78% for imipramine alone) as compared to 21% of subjects who received group psychotherapy with or without imipramine. In another comparison of pharmacotherapy and psychotherapy, Agras et al. (1992) evaluated the treatment efficacy of various lengths of desipramine treatment and CBT separately as well as combined. Seventy-one BN outpatients were randomly assigned to one of the five conditions: a 16-week or 24week treatment of desipramine (50-350 mg/d), individual CBT for 15 sessions (-3 follow-up sessions), or CBT combined with either the 16- or the 24week treatment of desipramine. At 16 weeks, both combined treatment and CBT were superior to desipramine alone in reducing binging and

472

W G. Johnson, J. I! Tsoh, and f? J. Vimnado

purging with abstinence rates at approximately 65%, 50%, and 35%, respectively. At 32 weeks, further reductions in binging were obvious in the combined treatment as compared to both 16 and 24 weeks of medication alone. Abstinence rates for the 24 week combined, CBT only and 24week medication were 70%, 55%, and 42%, respectively. Considerable relapse rate in binging was observed in the medication group at 32 weeks. Sixty-one of these subjects with BN were followed for a year, and both the 24week combined and CBT alone groups were superior in reducing binging and purging as compared to the medication-alone group. Abstinence at the l-year follow-up was significantly better for the 24week combined group (78%, 7 out of 9 subjects) than the 16-week medication only group (IS%, 2 of 11; Agras, Prossiter et al., 1994). Desipramine alone was also compared to CBT alone and a combined CBTdesipramine condition over 20 weeks (Leitenberg et al., 1994). At the end of treatment, abstinence rates for CBT and the combined condition were similar (83% vs. 80%) with none in the medication only group being abstinent. Improvements in depression and body image were also noted for the CBT and combined groups, and dropout rates were minimal in the CBT (17%) and combined groups (29%) compared to the desipramine condition (57%). These results suggest no additional benefit from desipramine over the 20-week CBT intervention. Fichter et al. (1991) evaluated 40 patients receiving 60 mg fluoxetine vs. placebo in the context of intensive inpatient behavioral psychotherapy over 35 days. Both groups showed significant reductions in hinging and purging, as well as improvements in other outcome measures such as eating attitudes and psychopathology. There were no differences between the groups suggesting minimal effects of fluoxetine over the psychotherapy intervention. In another placebocontrolled trial, Fahy, Eisler, and Russell (1993) compared CBT combined with d-fenfluramine, an appetite supressant, to CBT with a placebo over an 8-week treatment. They found no additive effects of d-fenfluramine, with both groups showing significant improvements in bulimic symptoms. At the end of treatment, 31% and 26%, respectively, stopped binging and purging for 2 weeks. Remission rates at 16 weeks were 37% for binging and 47% for purging. No outcome difference between the combined treatment and CBT alone was reported at 24 weeks and 1 year. One study with BED subjects evaluated CBT, a weight loss intervention, and desipramine in an additive design. CBT targeted binge eating and was implemented for 3 months followed by the weight loss intervention with and without desipramine for another 6 months. The weight loss intervention ignored binge eating and focused on dietary, exercise, and life-style changes, and it was implemented solely for 9 months or with the combined intervention. After the first 3 months, CBT lead to greater decreases in binge eating than the weight loss intervention with the latter intervention generating more weight loss. Desipramine did not enhance reductions in binge eating, yet the medication did add to the effectiveness of the weight loss intervention at a Smonth follow-up (Agras, Telch et al., 1994). CONCLUSIONS While medication may be helpful in the management of the psychopathology associated with AN, the literature clearly indicates that current medications do not promote weight gain nor influence the core characteristics of the disorder. The initial positive results on a group of difficult AN patients with fluoxetine are promising but preliminary. In contrast, medication with BN is effective in reducing bulimic symptoms when

Eating Llimmh

473

compared to placebo conditions. These positive results occur early in the course of treatment and are observed in a large portion of patients. Several reports also note positive changes in the fear of obesity, dietary/eating habits, and body image disturbance. SsRIs are preferred over other antidepressants primarily for their modest sideeffect profiles. A similar outcome to that shown for BN in reduced binge eating should be apparent for BED. BT has proven effective in promoting weight gain in AN during hospitalization, and cognitive interventions do not appear to augment the basic therapy. Attentional deficits, lethargy, and physical symptoms associated with low body weight my limit the impact of CBT during the weight restoration phase. Whether CBT has a greater effect on other core characteristics of AN, such as body image disturbance after weight restoration, is an open question. Studies evaluating the efficacy of CBT for BN indicate that it leads to a reduction in binge/purge cycles in over half of patients, and the initial data on its application with BED are promising. Comparisons of pharmacotherapy and CBT for BN suggest that psychotherapy alone is more effective than medication alone in reducing bulimic symptoms, and that symptomatic changes pro duced by psychotherapy endure at follow-up in contrast to medication where relapse rates are high. With one exception, research reports indicate that the addition of medication to psychotherapy yields no advantage over psychotherapy alone in the reduction of bulimic symptoms. Dropout rates on medication are from 2-3 times higher than that of CBT, and relapse rates following medication withdrawal are correspondingly high. These results are remarkably similar to the findings comparing pharmacotherapy and psychotherapy for depression (Munoz, Hollon, McCrath, Rehm, 8c VandenBos, 1994; Wexler & Cicchetti, 1992). Clinical recommendations regarding the choice of therapy are straightforward. For AN, BT and hospitalization depending on the presence of physical symptoms and the failure of outpatient treatment are indicated with medication oriented to the management of associated psychopathology. CBT is currently the treatment of choice for BN due to its positive response rate, durability, and low dropout rate. Medication is indicated for BN with cormorbid depression or when a rapid reduction in binge eating is desired, as may be the case with high rates of binge eating. However, medication must be combined with CBT to be effective with the majority of patients. The cost of the respective interventions may also be a factor in treatment decisions. Given a standard 100 mg/day dose, the cost of medication for 1 month is as follows: desipramine - $8, sertraline - $120, and fluoxetine - $280. In contrast, CBT for one session may cost $90 with monthly fees ranging from $360 to $720 depending upon the number of sessions per week. Several trends in the current literature on the nature of eating disorders and their treatment are evident. The notion of eating disorders as a spectrum of closely related conditions will foster more specific assessments of core cognitive and behavioral features and their interrelationships. So, rather than treating a diagnostic category such as AN or BN, interventions will focus on more distinct cognitive and behavioral targets. This specificity in evaluation will also be matched in pharmaco and psychotherapy for eating disorders. A closer alliance between the development of medications and basic research on the neurophysiological substrates of eating behavior will be evident as will the continued development of CBT interventions.

Acknozuledgements - We appreciate the assistance of Cheryl A. Osten, Frederick G. Grieve, and James Novalany for their help in the preparation of this manuscript.

W G. Johnson,J

474

I! Tsoh, and F(J Varnada

REFERENCES Abbott, D. W., & Mitchell, J. E. (1993). Antidepressants vs. psychotherapy in the treatment of bulimia nervosa. Pqchophawnacolqp

Bulletin, 29, 115-l 19.

treatments of bulimia netvosa. Journal of ClinicalAychiahy,52, 29-33. Agras,W. S. (1993). Short-term psychological treatments for binge eating. In C. G. Fairbum & G. T W&on (Eds.), Binge eating: Nature, treutment and assessment New York: Guilford Press. Agras, W. S. (1991). Nonpharmacologic

Agtas, W. S., Barlow, D. H., Chapin, H. N., Abel, G. G., & Leitenberg, H. (1974). Behavior modification of anorexia nervosa. Archives ofGeneral Psychiohy, 34279-286. Agras, W. S., Dorian, B., Kirkley, B. G., Amow, B., & Bachman, J. (1987). Imipramine in the treatment of bulimia: A double-blind controlled study. Intemational Journal of Eating L&n&s, 6, 29-38. Agras, W. S., Rossiter, E., Amow, B., Schneider, J. A., Telch, C. F., Raebum, S., Bruce, B., Perl, M., & Koran, L. M. (1992). Pharmacologic

and cognitive-behavioral treatment for bulimia nervosa: A controlled com-

parison. American Joumai of Psychiatry, 149,82-87.

Agms, W. S., Schneider, J. A., Amow, B., Raebum, S. D., & Telch, C. F. (1989). Cognitive-behavioral and response-prevention treatments for bulimia nervosa. Journal of Gmsulting and Clinical P@ology, 57, 215-221. Agras, W. S., Telch, C. F., Amow, B., Eldredge, K, Detzer, M. J., Henderson, J., & Marnell, M. (1995). Does interpersonal therapy help with patients with binge eating disorder who fail to respond to cognitivebehavioral therapy? Journal of Gmsulting and Clinical Ps@ologv, 63,356-366. Agras, W. S., Telch, C. F., Arnow, B., Eldredge, K, Wihley, D. E., Raebum, S. D., Henderson, J., & Mamell, M. (1994). Weight loss, cognitive-behavioral, additive design. &ha&r

and desipramine

treatments in binge eating disorder. An

Therafi, 25.225-238.

Alger, S. A., Schwaberg, M. D., Bigaouette, J. M., Michalek, A. V., & Howard, L. (1991). Effect of a tricyclic antidepressant and opiate antagonist on bingeeating

behavior in normoweight

bulimia and obese,

binge-eating subjects. American Journal of Clinical Nutrition, 99,865-871. American Psychiatric Association. ( 1993). Practice guideline for eating disorders. Americun J&n&

of Psychiutry, 150,212-228. American Psychiatric Association. ( 1994). Diugmstic and surtirtiad manual of tnentnl dimrim (4th ed.) . Washington, DC Author.

Andersen, A. E. (1983). Anorexia nervosa and bulimia: A spectrum of eating disorders. Journal of Adolescent

He&h Care, 4,15-21. Andetsen, A. E. (1987). Uses and potential misuses of anxiety agents in the treatment of anorexia netvosa and bulimia nervosa. In P. E. Garfinkel & D. M. Garner (Eds.), The sole of dmg tmztrnents foreatingdbrders (pp. 59-73). New York: Bmnner/Mazel. Barlow, J., Blouin, J., Blouin, A, & Perez, E. (1988). Treatment of bulimia with despmmine: A double blind crossover study. Canadian Joumal of Psych&y,

33, 129-133.

Beck, A. T. (1976). Cognitive therapy and the emotiunol disorders. New York: International Universities Press. Bemis, K M. (1987). The present status of operant conditioning for the treatment of anorexia nervosa. &hat.&

Mod$catbn,

Bianco, F. J. (1972).

11.432-463. Rapid teatment of two cases of anorexia netvosa. Jo-1

of Behavior Them@ und

Experimental Psychiahy, 3,223-224. Biederman, J., Herzog, D. 8.. Rivinus, T. N., Harper, G. P., Ferber, R. A., Rosenbaum, J. F., Harmants, J. S., Tondorf, R., Orsulak, P., & Schildkraut, J. J. (1985). Amitriptyline in the treatment of anorexia nervosa.

Journal

ofClinicalPsycht@annacology, 5,10-16.

Blinder, B. J., Freeman, D., & Stunkard, A. J. (1976). Behavior therapy of anorexia nervosa: Effectiveness of activity as a reinforcer of weight gain. A&an

Journal of Psychiatry, 126, 1093-1098.

Blundell, J. E., & Hill, A. J. (1993). Binge eating: Psychobiological mechanisms. In C. G. Fairbum & G. T. Wilson (Eds.), Binge eating: Nature, acscsnncnt, ond treabnent (pp. 266-224). New York: The Guilford Press. Casper, R. C., Schlemmer, R. G., & Javaid, J. I. (1986). A placebo-controlled dine in acute anorexia nervosa. Psych&y

crossover study of oral cloni-

Reseanh, 24249-266.

Channon, S., De Silva, P., Hemsley, D., & Perkins, R. (1989). A controlled trial of cognitive-behavioural

and

behavioural treatment of anorexia netvosa. Behaviour Researh Therafi, 27,529-535. Connors, M. E., Johnson, W. G., & Stuckey, M. K (1984). Treatment of bulimia with brief psychoeducational therapy. AmericanJournal of Psychiatry, 141,1512-1516. Cooper, P.J., & Fairbum, C. G. (1984). Cognitive behavior therapy for anorexia nervosa: Some preliminary findings. Journal of Psychannrxztic Reseanzh, 28.493-499. Crisp, A. H., Lacey, J. H., & Crutch&Id, vosa: An in-patient study. BrilMJoumal

M. (1987). Cl omipramine and drive in people wih anorexia ner-

of Psych&y,

150,355-358.

Eating Dirorak

475

de Zwaan, M., Nutzinger, D. O., & Schoenbeck, G. (1992). Binge eating in overweight women. comprehensive Psych&y, 33,256-261. Eckert, E. D., Goldberg, S. G., Hahni, K A., Casper, R. C., 8c Davis,J. M. (1979). Behavior therapy in anorexia nervosa. British Joumal of Rych&y, 134,55-59. Faby, T. A., Eisler, I., & Russell, G. F. M. (1993). A placebocontrolled trial of d-fenflutamine in bulima nervosa. Bsitish Joum& ofPyhiaq, 162,597-603. Fairbum, C. (1981). A cognitive behatioural approach to the treatment of bulimia. psYch&gi& A4cdicinc, II, 707-711. Fairbum, C. G. (1985). Cognitive-behavioral treatment for bulimia. In D. M. Gamer & P. E. Garfinkel (Eds.) , Handbook of psychotherapy for anorexia neruosa and bulimia (pp. 169-192). New York Guilford Press. Fairburn, C., & Hay, P. J. (1992). The treatment of bulimia nervosa. Annah of Medkine, 24,297-X)2. Fairbum, C. G., Jones, R., Peveler, R. C., Hope, R. A, & O’Connor, M. (1993). Psychotherapy and bulimia netvosa: Longer-term effects of interpersonal psychotherapy, behavior therapy, and cognitive behavior therapy. Archives of Ceneml Psych&y, 50,419-428. Fairburn, C. G., Kirk, J., O’Connor, M., & Cooper, P. J. (1986). A comparison of two psychological treatments for bulimia nervosa. Behnviour Reseavzh and Therapy, 24,629-643. Fairbum, C. G., Marcus, M. D., & Wilson, G. T. (1993). Cognitive-behavioral therapy for binge eating and bulimia nervosa: A comprehensive treatment manual In C. G. Fairbum & G. T Wilson (Eds.), Binge eating: Natusc, aswsmen t, and hzatment (pp. 361-404). New York: Guilford Press. Fairburn, C. G., Norman, P. A., Welch, S. L., O’Connor, M. E., Doll, H. A, & Peveler, R. C. (1995). A prospective study of outcome in buhma nervosa and the long-term effects of three psychological treatments. Anhives of Cenenzl Psych&y, 52,X)4-312. Flchter, M. M., Leibl, II, Rief, W., Brunner, E., Schmidt-Aubetger, S., & Engel, R R (1991). Fluoxedne versus placebo: A double-blind study with bulimic inpatients undergoing intensive psychothempy. P-hi&y, 24, l-7. Fhroxetine Bulimia Netvosa Collaborative Study Group. (1992). Fluoxetine in the treatment of bulimia nervosa: A multicenter, placebocontrolled, double blind trial. Anhim of Cmeral Psych&y, 49,139-147. Freeman, C. P L., Barry, F., Tumbull, J., & Henderson, A. (1988). Controlled trial of psychotherapy for bulimia nervosa. British Medical Jamal, 2%. 521-525. Freeman, C., Sinclair, F., Tumbull, J., & Annandale, A. (1985). Psychotherapy for bulimia: A controlled study. Journal of Psychiatric Research, 19,473-478. Garfinkel, P. E., Kline, S. A., & Stancer, H. C. (1973). Treatment of anorexia nervosa using operant conditioning techniques. Journal of Newmu and Men& Diseaws, 157,428-433. Goldbloom, D. S., & Ohnsted, M. P. (1993). Pharmacotherapy of bulimia nemosa with fluoxetine: Assessment of clinically significant attitudinal change. A me&an Journal of Psychiatty, 154770-774. Gormally, J., Black, S., Daston, S., & Rardin, D. (1982). The assessment of binge eating severity among obese persons. Addict& Behaviors, 7,47-55. Gross, H. A, Ebert, M. H., & Faden, V. B. (1981). A double-blind controlled study of lithium carbonate in primary anorexia nervosa. Journal of Clinical P@oph ammto&ly, 1,376-381. Gross, H., Ebert, M. H., Faden, V. B., Goldberg, S. C., Raye, W. H., Caine, E. D., Hawks, R., & Zinberg, N. (1983). A double-blind trial of delta 9-tetrahydrocannabinol in primary anorexia netvosa. Jmrnal of CiinicalPsychophamurcsl+Iy,3,165-171. Guiora, A. Z. (1967). Dysorexia: A psychopathological study of anorexia and bulimia. Americanfonrrnaf of Aychiaby, 124,391-393. Gwirtsman, H. E., Cure, B. H., Yager, J., & Gainsley, B. (1990). Fluoxetine treatment of anorexia nervosa: An open clinical trial. Journal of Clinical Pychiutty, 51, 378-382. Halmi, R A. (1985). Behavioral management for anorexia nervosa. In D. M. Garner 8c P. E. Garfinkel (Eds.) , Handbook of p+otherapy fw anon&a nemosa and bulimia (pp. 147-159). New York GuiIford Press. Halmi, K A, Eckert, E. D., LaDu, T. J., & Cohen, J. (1986). Anorexia nervosa: Treatment efficacy of cypro heptadine and amitriptyline. Archives of CeneralpSgchiatq, 43,177-181. Halmi, K A., Eckert, E., Marchi, I?, Sampugnaro, V., Apple, R., 8c Cohen, J. (1991). Comorbidity of psychiatric diagnoses in anorexia nervosa. Arcfiiva of CmeraL Psych&&y, 48,712-718. Herzog, D. B., & Sacks, N. R. (1993). Bulimia nervosa: Comparison of treatment responders vs. nonresponders. Psych@mmuco& B&in, 29,121-125. Hobel, B. G., & Hemandez, L. (1993). Basic neural mechanisms of feeding and weight regulation. In A. J. Stunkard & T. Wadden (Ed.%), C&&y: Theory and therupy (2nd ed., pp. 43-62). New York: Raven Press. Hoffman, L., & Hahni, E. (1993). Psychopharmacology in the treatment of anorexia nervosa and bulimia nervosa. Psychiatlic Clinics of North Amen&, 6,767-778. Home, R. L, Gerguston, J. M., Pope, H. G., Hudson, J. I., Lineberry, C. G., &her, J., 8~ Cato, A. (1988). Treatment of bulimia with bupropion: A muhicenter controlled trial.JoumalofUiniAychia&y, 49.262-266.

476

W G. Johnson,J. K Tsoh, and PJ

VumMto

Hsu, L. K. (1984). Treatment of bulimia with lithium. Anwicu n Journal of Psych&y, 141,1260-1262. Hsu, L. K, Clement, L., Santhouse, R., & Ju, E. S. (1991). Treatment of bulimia nervosa with lithium carbonate: A controlled study. Journal of Neroous and Mental LXseases, 179, 351-355. Hudson, J. I., Pope, H. G., Jonas, J. M., & Yurgelun-Todd, D. (1985). Treatment of anorexia nervosa with antidepressants. Journal of C.%cal P+ophannacology, 5,17-23. Hudson, J. I., Pope, H. G., Yurgelun-Todd D., Jonas J. M., & Frankenburg F. R (1987). A controlled study of lifetime prevalence of affective and other psychiatric disorders in bulimic outpatients. American Joumal of Psychiatry, 144, 128‘3-1287. Hughes, P. L., Wells, L. A., Cunningham, C. J., & Ilstrup, D. M. (1986). Treating bulimia with desipramine. Anhives of&n& P~chtiq, 43,182-186. Johnson, W. G., Corrigan, S. A., & Mayo, L. L. (1987). Innovative treatment approaches to bulimia nervosa. &ha&r Mod+cation, 11,373-388. Johnson, W. G., Schlundt, D. G., Barclay, D. R., CamNangle, R. E., & Engler, L. B. (1995). A naturalistic functional analysis of binge eating. Behuvim T&rue, 26,101-118. Johnson, W. G., Schlundt, D. G., 8c Jarrell, M. P. (1986). Exposure with response prevention, training in energy balance, and problem solving for bulimia nervosa. Intffnatirmal Journal of Eating Disalns, 5, 35-45. Johnson, W. G., & Torgrud, L. (1996). Assessment and treatment of binge eating disorder. In J. K. Thompson (Ed.), Eating disorders, obesity, and body image. Washington, DC: American Psychological Association. Jones, R., Peveler, R. C., Hope, R. A., & Fairburn, C. G. (1993). Changes during treatment for bulimia nervosa: A comparison of three psychological treatments. Behaviarr Reseanlt and Ttiapy, 31,479-485. Judd, F. K, Norman, T. R., & Burrows, G. D. (1987). Pharmacotherapy in the treatment of anorexia and bulimia nervosa. In P. J. V. Beumont. G. D. Burrows, & R. C. Casper (Eds.), Handbook of eating dirmdm. Part I: Anaxia and bulimia nervosa (pp. 361-370). New York: Elsevier. Kaplan, A. S., Garfinkel, P. E., Darby, P. L., & Gamer, D. M. (1983). Carbamazepine in the treatment of bulimia. American Joumal of Psych&y, 140.1225-1226. Keefe, P. H., Wyshogrod, D., Weinberger, E., 8c @as, W. S. (1984). Binge eating and outcome of behavioral treatment of obesity: A preliminary report. BehavioTResearch and Therapy, 22,319-321. Kennedy, S. H., Goldbloom, D. S., Ralevski, E., Davis, C., D’Souza, J. D., & Lofchy, J. (1993). Is there a role for selective monoamine oxidase inhibitor therapy in bulimia nervosa? A placebontrolled trial of brw faromine. Journal ofClinicalPsychopharmacology, 13, 412-422. Kirkley, B. G., Schneider, J. A., Agras, W. S., & Bachman, J. A. (1985). Comparison of two group treatments for bulimia. Journal of Gm.w%ng and Clinical Psychology, 53,43-48. Kleiman, G. L., Weissman, M. M., Rounsaville, B. J., & Chevron, E. S. (1984). Znt@emmal p~ch&en@y of depnkon. New York: Basic. Lacey, H. (1983). Bulimia nervosa, binge eating, and psychogenic vomiting: A controlled treatment study and long term outcome. Britirh MedicalJournal, 286,1609-1613. Lacey, J. H., & Crisp, A. H. (1980). Hunger, food intake and weight: The impact of clomipramine on a refeeding anorexia nervosa population. Postgradua& Medical Joumar 56.79-85. Iaporte, D. J. (1992). Treatment response in obese binge eaters: Preliminary results using a very low calorie diet (VLCD) and behavior therapy. Addictive Behauims, 17.247-257. Lee, N. F., & Rush, A. J. (1986). Cognitive-behavioral group therapy for bulimia. International Journal of EatingLknden, 5,599-615. Ie Grange, D., Eisler, I., Dare, C., Russell, G. F. M. (1992). Evaluation of family treatments in adolescent anorexia nervosa: A pilot study. International Journal of Eating D&n&x, 12, 347-357. Leibowitz, S. F. (1988). Brain neurotransmitters and drug effects on food intake and appetite: Implications for eating disorders. In B. T. Walsh (Ed.), Eating behavior in eating disonfzrs (pp. 21-35). Washington, DC: American Psychiatric Press. Leitenberg, H., Rosen, J. C., Gross, J., Nudelman, S., Xc Vara, L. S. (1988). Exposure plus response prevention treatment of bulimia nervosa. Journal of Consulting and Clinical Psychology, 56,535-541. Leitenberg, H., Rosen, J. C., Wolf, J., Vara, L. S., Detzer, M. J., & Srebnik, D. (1994). Compaxison of cognitive-behavior therapy and desipramine in the treatment of bulimia nervosa. BehuuiuurRes~h Therapy, 32, 37-45. Marcus, M. D., Wing, R. R., & Hopkins, J. (1988). Obese binge eaters: Affect, cognitions, and response to behavioral weight control. Journal of Gmsulling and Clinical Psychology, 56,433-439. McCann, U. D., & Agras, W. S. (1990). Successful treatment of nonpurging bulimia with desipramine: A double-blind, place~ontrolled study. American Journal ofPsych&y, 147, 1509-1513. Mitchell, J. E. (1992). Subtyping bulimia nervosa. In&national Joumal of EatingDicordm, Jl, 327-332. Mitchell, J. E., Hatsukami, D., Eckert, E. D., & Pyle, R. L. (1985). Characteristics of 275 patients with bulimia. A&an Journal of Psych&&y, 142,482-485.

Euting Dish

477

Mitchell, J. E., Pyle, R. L., Eckert, E. D., Hatsukami, D., Pomeroy, C., & Zimmerman, R. (1990). A comparison study of antidepressants and structured intensive group psychotherapy in the treatment of bulimia nervosa. Archives o/General Psychiafry, 47,149-157. Moldofsky, H., Jeuniewic, N., & Garfinkel, P. E. (1977). Prelimiaty report on metoclopramide in anorexia nervosa. In R. A. Vtgersky (Ed.), An& turwsa (pp. 373-376). New York: Raven Press. Munoz, R E, Hollon, S. D., McGrath, E., Rehm, L. P. & VandenBos, G. R. (1994). On the AHCPR Depression in Primary Care guidelines: Further considerations for practitioners. Amekun Psycholagist, 49, 42-61. Poole, A_ D., & Sanson-Young, R. W. (1978). A behavioural programme for the management of anorexia nervosa. Australian and New Zculand Journal of Rychiahy, 12,49-53. Pope, H. G., Jr., & Hudson, J. I. (1986). Antidepressant drug therapy for bulimia: Current statusJournal of Clinical Psych&y, 47, 339-345. Pope, H. G., Jr., Hudson, J. I., Jonas, J. M., & Yurgelun-Todd, D. (1983). Bulimia treated with imipramine: A placebo-controlled, double-blind study. America1 Journal of Psychiatry, 140,554558. Pope, H. G., Hudson, J. I., Jonas, J. M., & Yurgehm-Todd, D. (1985). Antidepressant treatment of bulimia: A twcyear follow-up study. Journal of Clinical Psycho@a-ology, 5,320-327. Pope, H. G., Jr., Reck, P. E., Jr., McElroy, S. L., & Hudson, J. I. (1989). A placebo-controlled study of trazodone in bulimia nervosa. Journal of Clinical psychophmmocdogy, 9,254-259. Robin, A. L., Siegel, P. T., Roepke, T., Moye, A. W., & Tice, S. (1994). Family therapy versus individual therapy for adolescent females with anorexia netvosa. Deuelopnu?ntal and Behauiura! Pcdiahics, 15,111-I 16. Rosen, J. C., SCLeitenberg, H. (1982). Bulimia netvosa: Treatment with exposure and response prevention. Behavior Therapy, 13,117-124. Russell, G. F. M., Szmukler, G. I., Dare, C., & Eisler, I. (1987). An evaluation of family therapy in anorexia nervosa and bulimia nervosa. Anhives of Gmcrol Psych&y, 44,1047-1056. Sabine, E. J., Yonace, A., Farrington, A. J., Barratt, K H., & Wakeling, A. (1983). Bulimia nervosa: A placebocontrolled double-blind trial of minaserin. Btitish Journal of Clinical Psychopha-o&y, 15, 195s-202s. Schlundt, D. G. (1989). Assessment of eating behavior in bulimia netvosa: The self-monitoring analysis system. In W. G. Johnson (Ed.), Advances in eating disorders (Vol I, pp. 1-41). New York: JAI Press. Schlundt, D. G., &Johnson, W. G. (1990). Eating disonlm: Asressmen t and heatmmt. Needham Heights, MA: Allyn & Bacon. Smith, D. E., Marcus. M. D., & Raye, W. (1992). Cognitive-behavioral treatment of obese binge eaters. Intematiaal Joumal of EatingDiso&rs, 15257-262. Spitzer, R. L., Devlin, M., Walsh, B. T., Hasin, D., Wing, R. R., Marcus, M. D., Stunkard, A., Wadden, T., Yanovski, S., Agras, S., Mitchell, J., & Nonas, C. (1992). Binge eating disorder: A multisite field trial of the diagnostic criteria. ZntnnafionalJournal of Eating Disorder, II, 191-203. Spitzer, R. L., Yanovski, S. Z., Wadden, T., Wing, R. Marcus, M., Stunkard, A, Devlin, M., Mitchell, J., Hasin, D., & Home, R. L. (1993). Binge eating disorder: Its further validation in a mulitisite study. Inkmational Journal of EatingDisorders, 13.137-153. Telch, C. F., Agras, W. S., Rossiter, E. M., Wilfley, D., & Renardy, J. (1990). Group cognitive-behavioral treatment for the nonpurging bulimic: An initial evaluation. Joumal of Consulring and Clinical Psychlogy, 58, 629-635. Thackwray, D. E., Smith, M. C., Bodtish, J. W., & Meyers, A W., (1993). A comparison of behavioral and cognitive-behavioral interventions for bulimia nervosa.&u& of Consultingand ClinicalPsycholo~, 61,639~645. Treasure, J., Todd, G., Brolly, M., Tiller, J., Nehmed, A, & Denman, F. (1995). A pilot study of a randomised trial of cognitive analytical therapy vs. educational behavioral therapy for adult anorexia netvosa. Bchaviour Rexarrh and Therapy, 33.363-367. Vandereycken, W. (1984). Neuroleptics in the short-term treatment of anorexia nervosa: A double-blind placebo-controlled study with sulpiride. British Journal of Psych&y, 144, 288-292. Vandereycken, W. (1987). The use of neuroleptices in the treament of anorexia nervosa patients. In P E. Garfinkel & D. M. Garner (Eds.), Thelok~ofdmgtmztmentsforeatingdis~ (pp. 74-89). NewYork Brunner/ Mazel. Vandereycken, W., & Pierloot, R. (1982). Pimozide combined with behavior therapy in the short-term treatment of anorexia nervosa: A double-blind placebo-controlled crossover study Acta Pyhiattica Scandinavica, 66,445-450. Vigersky, R. A., & Loriaux D. L. (1977). The effect of cyproheptakine in anorexia nervosa: A double-blind trial. In R. Vigersky (Ed.), An& nervosa New York: Raven Press. Walsh, B. T. (1991). Psychopharmacologic treatment of bulimia nervosa. J~mal of Clinical Psychiahy, 52, 34-38. Walsh, B. T. (1995). Pharmacotherapy of eating disorders. In K D. Brownell & C. G. Fairbum (Eds.), Eating disonfers and obesity: A comp&m.G~ handbook (pp. 313-317). New York: Guildford Press.

478

N! G. Johnson, J. I! Tsoh, and l? J. Varnado

Walsh, B. T., & Devlin, M. J. (1992). The pharmacologic treatment of eating disorders. P@&ric Clinics of North America, 15,149-160. Walsh, B. T., Gladis, M., Roose, S. P, Stewart, J. W., Stetner, F., & Glassman, A. H. (1988). Phenelzine vs placebo in 50 patients with bulimia. Archivesof&neral Psychiatry, 45,471-475. Walsh, B. T., Stewart, J. W., Roose, S. P., Gladis, M., & Classman, A. H. (1984). Treatment of bulimia with phenelaine: A double-blind, placebo-controlled study. Archiws of Gmnal Pqchiatry, 41,1105-l 109. Wells, L. A, & Logan, R M. (1987). Pharmacologic treatment of eating disorders: A review of selected literature and recommendations. Ps+somdiq 28,470-479. Wexler, B. E., & Cicchetti, D. V. (1992). The outpatient treatment of depression: Implications of outcome research for clinical practice. TheJournal of Menial and Nmvmu L&am, 180,277-286. Wilfley, D. E., Agras, W. S., Telch, C. F., Rossiter, E. M., Schneider, J. A., Cole, A C., Siiord, L., & Raebum, S. D. (1993). Group cognitivcbehavioral therapy and group interpersonal psychotherapy for the nonpurging bulimic individual: A controlled comparison. Journal of Gmsulting and Clinical psYchalogy,61, 296-305. Wilson, G. T., Rossiter, E., Rleifield, E. I., & Lindholm, L. (1986). Cognitive-behavioral treatment of bulimia nervosa: A controlled evalution. Behauiour Rtsumh and Therapy, 24.277-288. Wolchik, S. A., Weiss, L., & Katzman, M. A. (1986). An empirically validated, short-term psychoeducational group treatment for bulimia. InternationalJoumal of EatingL&ndcrs, 5,21X34. Wolf, E. M., & Crowther, J. H. (1992). An evaluation of behavioral and cognitive-hhavioral group interventions for the treatment of bulimia nervosa in women. Intcmattil J-al of EatingLkonfm, 1 I, 3-15. Yanovski, S. 2. (1993). Binge eating disorder: Current knowledge and future directions. ObesifyR+srearrh,I, 306-324. Yates, A. J., & Sambrillo, F. (1984). Bulimia nervosa: A descriptive and therapeutic study. &hautiur Rwarch and Tkrapy, 22,503-517.