Replication and evaluation of a proposed two-factor Binge Eating Scale (BES) structure in a sample of bariatric surgery candidates

Abstract

BackgroundThe Binge Eating Scale (BES) is a widely-used self-report measure of binge eating severity. Hood et al. [1] reported a 2-factor structure for the BES in a sample of bariatric surgery candidates, with factors labeled feelings/cognitions and behavioral manifestations. The present study aims to replicate and extend the factor structure obtained by Hood et al. [1] by testing the utility of a bifactor model that removes binge eating severity variance, which the total BES score purports to assess, in another sample of bariatric surgery candidates. MethodsBariatric surgery candidates (n = 517; 71.2% women; mean body mass index = 49.50 kg/m2; SD = 10.17) were sampled. Twenty-four percent met DSM-IV-TR diagnostic criteria for binge eating disorder. ResultsConsistent with previous research, a 2-factor structure for the BES was supported; however, the 2 factors were substantially correlated (r = .89). A bifactor model significantly improved model fit, supporting the presence of a higher-order severity factor accounting for a significant amount of variance. This factor was primarily marked by binge eating severity as demonstrated by associations with number of objective binge eating episodes and DSM-IV-TR diagnosis. In the bifactor model, the feelings/cognitions factor was modestly associated with self-reported mood psychopathology. However, contrary to expectations, the behavioral manifestation factor was not associated with empirically derived behavioral problems. ConclusionsThe current findings partially converge with those of Hood et al. [1] However, the use of the BES as a 2-factor measure is not recommended at this point because of lack of incremental validity demonstrated by the behavioral manifestation factor. Continued use of the BES as a unidimensional measure of binge eating severity, in conjunction with a comprehensive clinical interview, can provide useful guidance for presurgical treatment recommendations.