Article Text

See original article:

Download PDFPDF

Cognitive behaviour therapy was more effective than interpersonal psychotherapy for bulimia nervosa
  1. Cynthia M Bulik, PhD
  1. Virginia Commonwealth University, Richmond, Virginia, USA

Statistics from

OpenUrlCrossRefPubMedWeb of Science

QUESTION: What is the comparative effectiveness of cognitive behavioural therapy (CBT) and interpersonal psychotherapy (IPT) for the treatment of bulimia nervosa?


Randomised {allocation concealed*}, blinded (outcome assessors)*, controlled trial with 1 year of follow up.


2 outpatient treatment centres, California and New York, USA.


220 women (mean age 28 y, 77% white, mean body mass index 23 kg/m2) satisfying DSM-III-R criteria for bulimia nervosa. Exclusion criteria were severe physical or psychiatric conditions, current anorexia nervosa, current psychotherapy, use of psychotropic drugs, pregnancy, or previous CBT or IPT. 69% of patients completed the study, and all patients were included in the intention to treat analysis.


After stratification for history of anorexia nervosa, 110 patients were allocated to CBT, which focused on treating the eating disorder and associated cognitive disturbances, and 110 were allocated to IPT, which focused on achieving interpersonal change. Patients were given 19 individual sessions over a 20 week period. An independent quality control centre (Oxford, UK) trained and supervised the therapists and assessors to ensure similarity across treatments and assessments.

Main outcome measures

Eating disorder related symptoms and psychopathology.

Main results

Analysis was by intention to treat. At the end of treatment (20 wks), more patients who received CBT recovered (p<0.001), remitted (p=0.003), and met community norms for eating attitudes and behaviours (p=0.04) than did those who received IPT (table); no difference existed between groups for these outcomes at 1 year.

Cognitive behavioural therapy (CBT) v interpersonal psychotherapy (IPT) for bulimia nervosa


In the short term, cognitive behavioural therapy (CBT) was more effective than interpersonal psychotherapy (IPT) for achieving recovery and remission from bulimia nervosa. At 1 year, results from CBT and IPT were equivalent. Although IPT is effective, it takes longer to produce clinical change.


This 2 centre trial by Agras et al comparing CBT with IPT for the treatment of bulimia nervosa confirms the results of a previous study by Fairburn et al, which suggested that IPT may have therapeutic results equivalent to CBT.1, 2

The present study was a meticulously designed, controlled trial that allowed the authors to address important issues about the treatment of choice for bulimia nervosa.

The most critical message of this study is that although CBT and IPT lead to similar levels of clinical change at 1 year (and thus can be judged to be equally effective at 1 y), CBT must be considered to be a superior first line treatment by virtue of its more rapid effect on symptom reduction.

A rapid clinical improvement is an important dimension on which to base clinical decisions. Provided that a rapid treatment response is not transient, which in the case of CBT for bulimia nervosa it is clearly not, then CBT emerges as the clear winner. Many of the behaviours associated with bulimia nervosa can be physically damaging. In addition, with time, bulimic behaviors become more entrenched, and well worn behavioural patterns can be difficult to break. This study shows that by choosing CBT as a first line treatment, the duration of a patient's exposure to bulimic behaviours can potentially be minimised.

2 important questions remain. Firstly, how can we best approach those patients who did not achieve recovery or remission from CBT? Would IPT be a reasonable second choice for non-responders? Secondly, given that this study further establishes CBT as the psychotherapeutic treatment of choice for bulimia nervosa, how can we improve efforts to disseminate cognitive behavioural techniques to therapists in non-research settings to maximise the number of afflicted patients who receive this preferred treatment?



  • Sources of funding: National Institute of Mental Health; Wellcome Trust; Henry J Kaiser Family Foundation; Foundation Fund for Research in Psychiatry.

  • For correspondence: Dr W S Agras, Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, 401 Quarry Road, Stanford, CA 94305, USA. Fax +1 650 723 9807.

  • * See glossary.

  • Information provided by author.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

  • Correction
    BMJ Publishing Group Ltd, Royal College of Psychiatrists and British Psychological Society