Norms, normality, and clinical significant change: implications for the evaluation of treatment outcomes for eating disorders.

OBJECTIVE: The implications of the use of national norms of the Eating Disorders Inventory (EDI) are investigated. METHOD: A German version of the EDI was administered to a representative national sample (n = 650), a repeatedly measured community sample (n = 207), and a national inpatient sample with primary diagnoses of anorexia or bulimia nervosa (n = 959). Resulting scale distributions were compared with those of North American samples of the handbook. RESULTS: Normal ranges on several scales markedly differed when German community norms were used. When estimates of rates of clinical significant changes were based on German norms instead of on North American norms, rates were markedly lower for scales measuring core diagnostic features (e.g., binging/purging, Drive for Thinness) and higher for those measuring correlates (e.g., Maturity Fears, Interpersonal Distrust). DISCUSSION: Neglecting differences in national norms can lead to substantial misinterpretation of empirical findings in the diagnosis of eating disorders and the evaluation of treatment outcome.