This review concluded there was some limited evidence that pure and guided self-help interventions based on cognitive behavioural therapy (CBT) may provide some benefit to individuals with bulimia nervosa or binge-eating disorder. The conclusions take into account the limitations of the data, but may not be reliable given concerns about the review methods.

To determine the efficacy of self-help interventions for the treatment of bulimia nervosa and binge-eating disorder.

MEDLINE and PsycINFO were searched for published studies. Search terms, but not dates, were reported. Reference lists of published relevant reviews and studies were searched for additional studies.

Any study that evaluated the effects of an acute pure self-help or guided self-help intervention on the symptoms of eating disorder in individuals with bulimia nervosa or binge-eating disorder was eligible for inclusion in the review. Studies that included only individuals with bulimia nervosa and/or subthreshold bulimia nervosa (eating disorders not otherwise specified; EDNOS) or mixed populations with mainly bulimia nervosa patients were all classified as bulimia nervosa studies.

Most included studies assessed guided self-help in bulimia nervosa populations; the rest assessed pure self-help or (one study) both intervention types. A small number of studies also assessed self-help interventions combined with medication or cognitive-behavioural therapy (CBT). Most self-help interventions were comprised of training manuals, training sessions with a healthcare provider or social worker and/or homework assignments. Both controlled and uncontrolled studies were included. Comparison groups, where relevant, included other treatments such as CBT, other forms of self-help and other treatments such as medication and waiting list controls. Outcomes included abstinence from binge eating, laxative use and/or vomiting, improvement rates and eating disorder symptoms.

The authors stated neither how papers were selected for review nor how many reviewers performed the selection.

The authors did not state that they assessed validity.

The authors stated neither how data were extracted for the review nor how many reviewers performed the data extraction. Percentage values were extracted for abstinence, improvement and symptom outcomes.

Studies were grouped by intervention and population in a narrative synthesis.

Twenty-six studies were included in the review: two randomised controlled trials (RCTs; 195 patients); 16 non-randomised controlled studies (1,306 patients); and eight non-controlled studies (279 patients). Sample sizes ranged from eight to 120 patients. Completion rates in the studies varied from 30.8% to 100%.

All eight uncontrolled trials of self-help for bulimia nervosa or EDNOS reported reductions in binge eating (range 33% to 85%) and self-induced vomiting (17% to 80%). Studies reported abstinence rates of between 12.8% and 47% for binge eating, 27% and 65.2% for vomiting and 26.8% and 50% for binge-eating and vomiting.

Seven out of the eight studies that compared self-help with waiting list control for bulimia nervosa or binge-eating disorder reported a significant advantage for self-help. Improvement rates ranged from 25% to 87% for self-help and 6% to 19% for waiting list control; abstinence rates ranged from 6% to 53% for self-help and 0% to 13% for waiting list control.

Two of eight studies that compared self-help to another active comparator found significant differences in favour of guided self-help in comparison with pure self-help (one study) and standard management (one study). Six studies found no significant differences between study groups.

Three studies that compared a combination of self-help with medication for bulimia nervosa or binge-eating disorder reported mixed findings.

Two studies that compared a combination of self-help and CBT with CBT alone reported no significant differences between the two intervention types.

There was some limited evidence to suggest that self-help and guided self-help interventions based on the principles of CBT provided some benefit to individuals with bulimia nervosa and binge-eating disorder.

This review answered a clear research question but used wide inclusion criteria for both study design and outcome. The authors provided few details of their literature search and so it was difficult to assess whether relevant data may have been missed. Publication bias was possible as only published studies were eligible for inclusion. It was difficult to assess the risk of bias in the review process as the authors did not report their methods.

With no assessment of study validity, the reliability of the data was unclear. As the review included mostly small studies that used a wide range of study designs (which included non-controlled studies and studies that did not follow outcomes over time), the data may not have always been reliable. The studies also varied in intervention, control, population and outcome definition, which suggested that the use of narrative synthesis was appropriate.

The authors' conclusions took into account the limitations of the data, but the conclusions may not be reliable given concerns about the review methods.

Practice: The authors stated that it may be appropriate to offer self-help and guided self-help interventions based on CBT for individuals with bulimia nervosa and binge-eating disorder when no other established treatment is available.

Research: The authors stated that further research was required to compare self-help interventions for bulimia nervosa and binge-eating disorder with established interventions such as CBT and other credible comparison treatments.

The review was supported by grant DK53635 from NIDDK.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.