Bulimia Nervosa: What 25 Years of Research Tells Us

Thirty years ago, Gerald Russell first outlined the features of bulimia nervosa (BN)-dread of overeating, compensatory measures, and morbid fear of gaining weight and getting fat (Psycho Med 1979; 9:429). Results of a recent survey of the literature show that 25 years later, for many patients the outcome is still unsatisfactory.

Dr. Hans-Christoph Steinhausen and Sandy Weber reviewed 79 study series covering 5,653 patients with BN, to evaluate recovery, improvement of symptoms, chronicity, crossover to another eating disorder, mortality rates, and psychiatric comorbidities (Am J Psychiatry 2009; 166:1331. Getting a clear picture of each variable was challenging because studies used different measures, and after the 1990s, there was an increasing reliance on the DSM-IV and ICD-10 criteria. For example, 27 studies used a three-level classification of global outcome (recovery, improvement, chronicity), and six of these studies included information about crossover to another eating disorder. According to this procedure, nearly 45% of patients showed full recovery, whereas 27% improved considerably and nearly 23% had a protracted course. In another 27 studies, only two outcome parameters were used; in 19 studies, recovery and chronicity were given, and recovery increased to almost 60% on average; 30% of patients had a chronic course; and 10% crossed over to another eating disorder.

Dropout rate, follow-up, and intervention

According to the authors, several conclusions can be drawn from analyses of three central variables: dropout rate, duration of follow-up evaluation, and type of intervention. First, there are some signs that patients who dropped out of follow-up studies may have had a more favorable course of BN. Second, duration of follow-up evaluation was the variable with the strongest effect on outcome.

The dropout rate had a strong effect on rates of recovery and crossover diagnoses. The effects of duration of follow-up were robust among the three variables. The recovery rate was strongest after 4 years of BN. As for the type of intervention, psychotherapy scored highest, followed by medical therapy, then behavioral therapy. Crossover to another eating disorder was observed most often with behavior therapy, less often with medical therapy, and not at all with psychotherapy.

When prognosis was evaluated, most studies showed that the duration of illness had no effect on prognosis. In five studies, a short duration was beneficial, whereas it was negative in another study; the authors found no significant association between age at onset and outcome. Mostly, however, results were mixed—positive in some studies, negative in others. Positive social factors including social adjustment, close social relationships and high socioeconomic status were all identified as positive prognostic factors, whereas a high amount of psychological distress and low job status were considered negative factors.

A search for variables in family history and environment that might affect the course of the illness led to few significant findings. Nonsignificant factors were found for parental socioeconomic status, sexual abuse, and quality of family bonding.

With a mean crude mortality rate of 0.32%, including a number of deaths not caused by BN, the illness was definitely less fatal than anorexia nervosa, as was reported in a parallel review of anorexia nervosa by Steinhausen et al (Am J Psychiatry 2002; 159:1884), where the morality rate was 5%. However, the authors point out that the frequencies of comorbid psychiatric disorders were high for both disorders. Affective and neurotic/anxiety disorders ranked highest and a sizeable proportion of patients had personality disorders at follow-up.

Thus, after a quarter century of research, BN remains a serious disorder with unsatisfactory recovery and improvement rates. Many patients with BN continue to have chronic eating disorder problems and other comorbid psychiatric disorders over a long period of their lives. According to the authors, future research could benefit from more collaborative and prospective studies based on large unselected samples and standardized assessment approaches.