Comparing eating disorder examination questionnaire factor structures in veteran men and women

Abstract

The present study aims to investigate the factor structure and measurement invariance by gender of the Eating Disorder Examination Questionnaire (EDE-Q)—a commonly used measure evaluating eating disorder symptom severity— in veterans. The present study used data from a 2022 survey study conducted with a nationally representative sample of 405 veterans. Competing factor structures based on prior literature were compared using confirmatory factor analyses (CFA). Multigroup CFA was used to evaluate measurement invariance among men and women (n = 401). None of the full-item factor structures, including the original four-factor model, were supported. A brief seven-item, first-order three-factor structure demonstrated best model fit and highest scale reliability. A bifactor model that included a general factor and three brief factors demonstrated adequate fit; however, factor loadings for two of the specific factors were low, and internal consistency of all three specific factors in this model was poor. Both the best-fitting first-order and the bifactor models demonstrated measurement invariance by gender. The present study strongly supports the use of a brief, seven-item three-factor EDE-Q with veterans. Moreover, there was some evidence for the appropriateness of a global score from the seven-item EDE-Q, although it may not fully capture eating disorder symptom severity in veterans. Further, the brief seven-item EDE-Q is appropriate for gender comparisons. The results of this study have important clinical and research implications for the use of the EDE-Q to evaluate eating disorder symptom severity in veteran men and women.