Long‐term outcome of inpatients with bulimia nervosa—Results from the Christina Barz Study

OBJECTIVE: To assess the long-term outcome and identify outcome predictors in a very large sample of inpatients treated for bulimia nervosa (BN). METHOD: Out of a total of 2,033 patients admitted consecutively to specialized treatment, 1,351 patients (mean age at treatment 25.94) were assessed for follow-up on average 11 (SD 6) years after admission. Also a very long-term (21 years) subsample (N = 147; mean age 25.92) was defined. Bivariate and logistic regression analyses identified predictors of poor outcome. RESULTS: For more than 70% of the patients follow-up information could be gathered. Severity of eating disorder (ED) and other symptoms decreased over time but remained higher than in healthy controls, using published normative data. Remission rate was 38% after 11 years and 42% in the subsample after 21 years. Out of the total sample of N = 2,033 patients, 49 had died (2.4%). Persistent BN was found in 14.2% and the most frequent crossover was to ED not otherwise specified. Predictors of poor outcome were fewer follow-up years, higher drive for thinness, higher age at treatment, and less global functioning. DISCUSSION: Based on clinical indicators, patients presented with a high level of ED and psychiatric symptomatology. With less than half of the patients remitted after 22 years, efforts are needed to improve treatment outcome.