Validation of the yale-brown obsessive compulsive scale modified for binge eating.

Objective: Establish the Yale-Brown obsessive compulsive scale modified for binge eating (YBOCS-BE) as a fit for purpose measure of treatment benefit in clinical trials of binge eating disorder (BED).

Methods: YBOCS-BE psychometric properties were evaluated with data from a Phase 2 randomized controlled trial of lisdexamfetamine dimesylate in 260 adults with BED. Assessments included: Cohen's effect size estimates of item-level sensitivity and scale-level external responsiveness; item-to-total correlations; Cronbach's alpha for internal consistency reliability; Spearman correlations against reference measures for construct validity; known-groups analyses for discriminating ability; t tests of within-group differences between baseline and post baseline visits for internal responsiveness; and multiple anchor-based approaches to estimate minimum clinically important change (MCIC).

Results: No significant distribution anomalies were seen. Items appear sensitive to treatment group differences. Item-to-total correlations were positive. Internal consistency is 0.81. Large correlations (>0.50) were seen between YBOCS-BE score change and the Clinical Global Impression-Improvement (CGI-I; 0.58) and score changes for the following; number of binge days (0.38), Clinical Global Impression-Severity (CGI-S; 0.57), the disinhibition (0.57) and hunger (0.52) subscales of the Three-Factor Eating Questionnaire (TFEQ), and the Barratt Impulsiveness Scale (BIS-11; 0.58). MCIC estimates range from -4 to -17.

Discussion: The YBOCS-BE was found to be a reliable and valid measure of an important and unique concept in BED-related clinical studies. Study limitations include using protocol-defined BED severity level and the exclusion of psychiatric comorbidities.