The outcome of bulimia nervosa: findings from one-quarter century of research

OBJECTIVE: The present review addresses the outcome of bulimia nervosa, effect variables, and prognostic factors. METHOD: A total of 79 study series covering 5,653 patients suffering from bulimia nervosa were analyzed with regard to recovery, improvement, chronicity, crossover to another eating disorder, mortality, and comorbid psychiatric disorders at outcome. Forty-nine studies dealt with prognosis only. Final analyses on prognostic factors were based on 4,639 patients. RESULTS: Joint analyses of data were hampered by a lack of standardized outcome criteria. There were large variations in the outcome parameters across studies. Based on 27 studies with three outcome criteria (recovery, improvement, chronicity), close to 45% of the patients on average showed full recovery of bulimia nervosa, whereas 27% on average improved considerably and nearly 23% on average had a chronic protracted course. Crossover to another eating disorder at the follow-up evaluation in 23 studies amounted to a mean of 22.5%. The crude mortality rate was 0.32%, and other psychiatric disorders at outcome were very common. Among various variables of effect, duration of follow-up had the largest effect size. The data suggest a curvilinear course, with highest recovery rates between 4 and 9 years of follow-up evaluation and reverse peaks for both improvement and chronicity, including rates of crossover to another eating disorder, before 4 years and after 10 years of follow-up evaluation. For most prognostic factors, there was only conflicting evidence. CONCLUSIONS: One-quarter of a century of specific research in bulimia nervosa shows that the disorder still has an unsatisfactory outcome in many patients. More refined interventions may contribute to more favorable outcomes in the future