In our Western society, we have several different types of eating dysfunction, all of which are unique and tragic in their own right. Despite their individuality, however, they all have several overlapping symptoms that are key to their classification and severity. For Bulimia Nervosa (BN) and Binge Eating Disorder (BED), one of the core features is binge eating, which can be defined objectively by number of calories eaten in a given time or subjectively by the feelings of the binger. Binge eating has many different aspects that are of interest to scientists and clinicians alike. One of those interests has to do with the reduction of this symptom among populations being treated for their respective disorder. Because both disorders are relatively new to the scientific and clinical world, there is much debate over how to treat, define, and possibly distinguish this phenomenon between the two disorders. For the purpose of this review, I intended to examine the evidence for the best outcomes, both immediate and long term, for reduction of binge eating in these disorders.
Clinicians treat BN and BED in several different ways, some of which are more effective than others, especially at reducing the core symptom of binge eating. Although many treatments exist, those most researched in terms of treatment outcome are interpersonal therapy (IPT), cognitive behavioral therapy (CBT), behavior therapy (BT), and pharmacological therapy. These treatments are not designed to simply treat binge eating; instead, they are used to treat the whole of the disorder. However, by understanding the workings of the therapy and differences in treatment outcomes among core features, science can expand upon its understanding of the causes of BN and BED. Thus, it is important to know which therapies affect which symptoms effectively.
Interpersonal therapy (IPT) is a derivative of therapy designed for depressed individuals, which originated from the New Haven-Boston Collaborative Depression Project (Fairburn et al 1993). It borrows from the psychodynamic school of psychology, but concentrates on interpersonal functioning and involves three steps: 1) the first is an intensive analysis of the interpersonal context which the disorder develops and is maintained 2) the second involves addressing the interpersonal problems and contracting with the patients to work on these problems 3) the third deconstructs feelings about termination of therapy, reviews progress and outlines future work (Fairburn et al. 1993).
Cognitive behavioral therapy (CBT) is currently the most commonly used type of psychotherapy (Peterson & Mitchell 1999). It is based on the cognitive view of bulimia nervosa, which weighs attitudes towards shape and weight as most significant for maintenance of the disorder. Therapy actually includes a combination of behavioral and cognitive directives aimed at changing patterns of behavior, attitudes of shape and weight, and cognitive distortions such as low self esteem (Fairburn et al. 1993).
Behavioral therapy (BT) is similar to CBT but concentrates on behavioral procedures used to normalize eating habits. The major features are regaining control over eating, establishing a regular pattern of eating, and cessation of dieting (Fairburn et al. 1993). The latter is a controversial issue with BED because of the high incidence of obesity in this disorder (Wilfey et al. 1997).
Pharmacotherapy is the use of drugs, mainly selective seratonin reuptake inhibitors (SSRI) or MAO inhibitors. The list of these antidepressant drugs includes imipramine, desipramine, phenelzine, bupropion and fluoxetine (Agras et al 1992). The use of strictly pharmacological treatment is touted by one study to be the most cost-effective treatment for eating disorders (Agras 1997). However, biases are present in science and this comment may reflect the views of the author, W. Stewart Agras, who is a MD and not a clinical psychologist.
Several review papers have looked at the question of which type of therapy is most effective in treating bulimia nervosa and binge eating disorder (Peterson & Mitchell 1999; Wilfey et al 1997), but little detail is discussed about the reduction of binge eating specifically. CBT is the psychotherapy most successful in immediate reduction of symptoms, including binge eating, and long-term abstinence from the characteristics of each disorder (Wilfey et al. 1997). However, Wilfey et al. (1997) claims that long-term effects of IPT are comparable to CBT in cases of BN, but not enough is known about treatment for BED to make any firm conclusions. What is interesting about this conclusion is that IPT does not deal directly with behaviors associated with eating disorders, yet it is successful in reducing the core symptoms of BN.
The debate between psychotherapy and pharmacotherapy is at least two-dimensional. The first dimension is less scientific and more important to clinicians who face more pressure to efficiently and effectively serve their patients. Although, Agras (1997) claims that 24 weeks of medical treatment was 1,982$ and 3,230$ for CBT, the reality of the problem of cost comes from health insurance coverage plans and the influence of drug companies on this system. It is an issue too broad for this paper, but one worth mentioning because it is vital to the reality in treatment of eating disorders. The second dimension is that of true effectiveness. Peterson & Mitchell (1999) conclude that CBT is the most effective psychological therapy and more effective than medication alone. However, they concede that a combination of medicine and CBT is moderately more effective than CBT alone for reducing symptoms of BN.
With this background discussed, the rest of the paper looks at the experimental evidence that measures the effectiveness of treatments on a number of core symptoms. Only the results on reduction of binge eating will be discussed in hopes of determining if the core symptom of binge eating should be thought of as identical in both BN and BED as well as determining which therapy is most effective at reducing binge eating.
Fairburn et al (1993) studied the effectiveness of three psychotherapies on the frequency of binge eating and purging in seventy-five consecutively referred cases of BN. The only exclusion criterion was concurrent anorexia nervosa, which may have allowed incidences of depression or personality disorders to complicate the data since they do occur with BN (Wilfey et al. 1997). An additional problem that occurs in many of the clinical trials (Wilfey et al. 1997) is attrition rate. In this particular study, the BT group lost 48% of its sample by the final assessment and a total of 33% of the original seventy-five withdrew by the final assessment. These assessments were done after the nineteenth and final session (18 weeks), 4, 8, and 12 months after treatment. The mean frequency of binge eating (objective) in the CBT group fell from 18.5 to 0.8 and from 17.2 to 1.1 in IPT. There was no significant difference between the two reported decreases. However, rates of total abstinence of binge eating were not reported as well as the data on BT because of the high drop out rate. Additionally, only objective binge eating was measured which may be why this study finds an equal reduction rate in binge eating between the different therapies. It is important to note that these findings were maintained up to one year following the end of treatment, which suggests that objective overeating can be reduced, by both IPT and CBT.
Fairburn et al. (1991) compared CBT, BT and IPT on seventy-five randomly assigned bulimics aged 17 and older who were first referred by MD’s and then screened to make sure they met DSM-III-R criteria for BN before entering experimental groups. This study did not control subjects for any other eating disorders or psychological symptoms that could have the possibility of contamination effects on binge eating reduction. The authors measured the core features of BN with the eating disorder examination (EDE) and found that all three therapies reduced the mean occurrence of both subjective and objective binges. There was no significant difference between therapies, which further complicates the picture of binge eating. If inclusion of cognitive elements in therapy does not predict the increased reduction of binge eating, then it is important to ask the question of whether or not the cognitive characteristics of binge eating are important to BN? Additionally, if behavioral and cognitive treatments are equally as effective at reducing binge eating, then perhaps this phenomenon is the result of a much more complicated picture of internal functioning that has many root stems, any of which will kill the core of the event. I find it important to address this concern because instances of binge eating in BED are thought of as similar, although they are less likely to include purging behaviors. To understand the most important features and direct further research, these thoughts about binge eating should be addressed. However, hasty conclusions are not rewarded as this paper failed to report the variance of subjects’ scores and did not report the abstinence levels of any of the therapies. This information could explain the doubts about the cognitive, behavioral, and interpersonal features of binge eating, which are seemingly equally powerful as evidenced by this study.
The comparison of pharmacology and psychotherapy brings to question another dimension of binge eating—biologic factors. If biologic factors are the core importance in binge eating, then drug therapy should be superior to psychotherapy. Agras et al. (1992) addressed this question in a controlled comparison of CBT and desipramine. There were seventy-one randomly assigned bulimics recruited by advertisements and referrals. The exclusion criteria included unfit for use of anti-depressant drugs, concurrent anorexia, drug or alcohol abuse, psychosis, or depression with suicidal risk. This is an improvement over the studies comparing just psychotherapy. Additionally, a psychologist blind to treatment assessed the subjects at zero, sixteen, twenty-four, and thirty-two weeks. There were five groups: desipramine withdrawn at sixteen or twenty-four weeks, combined treatment with medication withdrawn at sixteen or twenty-four weeks, and fifteen sessions of CBT. Medication at sixteen weeks reduced binges per week from a mean of 5.5 to 3.5(16 weeks), 3.7(24 weeks) but an increase in mean binges occurred at 32 weeks. This means that medication effects lasted approximately eight weeks after cessation, but rebounded by twenty-four weeks. This observation calls to question the sustained effects reported in the twenty-four week medicine condition, which only includes and eight week follow up. The results showed a sustained decrease from a mean of 5.9(pretreatment) to 3.3(32 weeks). However, this finding cannot be seen as superior to the sixteen-week group because of the timing of assessments. Both CBT and the combined treatment were significantly more effective at reducing binge eating (81.7% and 82.6 % respectively). There was a marked difference in the reports of abstinence in each condition as well. The combined treatment groups were 65% abstinent, CBT 50%, and the desipramine groups 38.8% abstinent at the 32-week assessment.
Agras et al (1991) sheds some light onto the question of importance of biologic factors to binge eating. Although pharmacotherapy reduced binge eating, these effects were not sustained and there was a marked difference in abstinence rates among groups. Additionally, there was no placebo control, so the effects of the medication cannot be accurately determined. However, this study suggests that the cognitive and behavioral components are more important to the core of binge eating than biologic imbalances in brain chemistry.
Another study done by Walsh and colleagues (1997) addressed the difference between CBT, psychodynamically oriented supportive therapy and desipramine in a placebo-controlled study. The 120 women went through similar exclusion criteria as Agras et al. (1991) and had DSM-III-R bulimia for at least one year. This study improved upon Agras et al. (1991) by allowing for fluoxetine replacement if desipramine was ineffective or not tolerated. A second improvement is that patients were required to keep daily records of binges, which may be more accurate than the recall methods of earlier studies. There was not a statistical difference between CBT with medication vs. CBT without medication in reduction of binge eating. Supportive psychotherapy was not as effective as CBT n reducing binges and CBT with and without medication was better than medication alone. Despite the differences in CBT vs. medication, the results of the drug vs. placebo test reveal that pharmacotherapy did have a significant effect on reducing binge eating. This evidence however is not long term and still remains to answer the question of whether or not drug therapy is effective in the long-term maintenance of reduced binge eating.
Agras et al (1994) did a one-year follow-up to psychological and pharmacological treatments of BN to determine which were the most successful for sustaining the reduction of bulimic symptoms. Using the same design as in Agras et al. (1991), with the exception of using an eighteen-session version of CBT as opposed to fifteen and a smaller sample size of sixty-one, the authors found similar results. The question of a rebound effect occurring after eight weeks in Agras et al. (1991) was answered as the twenty-four week drug group sustained a low mean of binge eating at the one-year follow-up. In comparison of the CBT with and without drug treatment, the combined treatment showed and increased in percentage drop in mean number of binges. However, the standard deviation was 7.7 compared to 3.8 of the CBT group, which suggests that there was a greater variance in outcome with the combined treatment, possibly due to side effects of the medication. This study suffered from several of the recurring problems. For instance, abstinence rates were unreported. Additionally, the sample included only females and had subjects report binges on a seven-day recall, which only captures subjective binges.
With the information on different types of therapy discussed, the next question becomes are the results consistent with BED. Only one suitable study was found and it reported that there were similar effects between BED and BN when treated with pharmacological and CBT oriented therapies. Obviously each is modified slightly to address the differences in the disorders, but reduction in binges seems most effected by CBT. However, the author suggests that mediation be thought of as a first treatment option because of its apparent cost effectiveness. Further research needs to be done on both BN and especially BED in order to flush out any differences that may exist in the nature of binge eating.
The evidence on reduction of binge eating suffers from unreported data, sample bias, and lack of long-term follow up studies. Additionally, mixed results exist on the efficacy of psychotherapies when compared to each other, which questions the understanding of the phenomenon of binge eating. Could binge eating be a construct that has several strong roots founded in behavioral, cognitive, interpersonal, and biologic characteristics, each of which when treated could eliminate the presence of overeating episodes? More research looking at the specific features of binge eating, for instance subjective vs. objective, would help answer this question. Additionally, does this phenomenon occur differenently in men or even in BED? Research is needed that examines these variables. For now, researchers agree that treatment of BN should include CBT for the best outcome and may or may not need pharmacological assistance (Wiley et al. 1997; Peterson & Mitchell 1999). Although CBT seems a logical application for BED, more research needs to be done in order to provide the best possible treatment for those suffering from this disorder. Finally, research needs to be done on the predictive variables for treatment outcome because their presence may help explain the nature of binge eating as well as better enable clinicians to choose the most effective therapy for their individual patients.
Agras, W. S. (1997). Pharmacotherapy of Bulimia Nervosa and Binge Eating Disorder: Longer-Term Outcomes. Psychopharmacology Bulletin, 33(3): 433-36.
Agras, W. S., Elise, M. R., Arnow, B., Telch, C. F., Raeburn, S. D., Bruce, B. & Koran, L. M. (1994). One-Year Follow-Up of Psychological and Pharmacologic Treatments for Bulimia Nervosa. Journal of Clinical Psychiatry, 55: 179-183.
Agras, W. S., Rossiter, E. R., Arnow, B., Schneider, J. A., Telch, C. F., Raeburn, S. D., Bruce, B. Perl, M., & Koran, L. M. (1992). Pharmacologic and Cognitive-Behavioral Treatment for Bulimia Nervosa: A Controlled Comparison. American Journal of Psychiatry, 149: 82-87.
Fairburn, C. G., Jones, R., Pevleer, R. C., Hope, R. A., O’Connor, M. (1993). Psychotherapy and Bulimia Nervosa. Archives of General Psychiatry, 50: 419-428.
Fairburn, C. G., Norman, P. A., Welch, S. L., O’Connor, M. E., Doll, H. A., & Robert, C. P. (1995). A Prospective Study of Outcome in Bulimia Nervosa and the Long-term Effects of Three Psychological Treatments. Archives of General Psychiatry, 52: 304-312.
Fairburn, C. G., Jones, R., Peveler, R. C., Carr, S. J., Solomon, R. A., O’Connor, M. E., Burton, J., Hope, R. A. (1991). Three Psychological Treatments for Bulimia Nervosa. Archives of General Psychiatry, 48: 463-469.
Peterson, C. B. & Mitchell, J E. (1999). Psychosocial and Pharmacological Treatment of Eating Disorders: A Review of Research Findings. Journal of Clinical Psychology, June: 686-697.
Walsh, B. T., Wilson, G T., Loeb, K L., Devlin, M. J., Pike, K. M., Roose, S. P., Fleiss, J. & Waternaux, C. (1997). Medication and Psychotherapy in the treatment of Bulimia Nervosa. American Journal of Psychiatry, 154: 523-531.
Wilfey, D. E., Cohen, L. R. (1997). Psychological Treatment of Bulimia Nervosa and Binge Eating Disorder. Psychopharmacology Bulletin,33(3): 437-454.
Psychology DepartmentThe Health Psychology Home Page is produced and maintained by David Schlundt, PhD.
Vanderbilt Homepage | Introduction to Vanderbilt | Admissions | Colleges & Schools | Research Centers | News & Media Information | People at Vanderbilt | Libraries | Administrative Departments | Medical
|Return to the Health Psychology Home Page|
|Send E-mail comments or questions to Dr. Schlundt|