Psychometric properties of the Eating Disorder Examination-Questionnaire and psychopathology in Japanese patients with eating disorders.

Abstract

The Eating Disorder Examination-Questionnaire (EDE-Q) is one of the most widely used tools to assess the core psychopathology of eating disorders (ED). However, recent empirical findings did not support the original four-factor structure. The aims of the present study were to investigate the factor structure of the EDE-Q in Japanese ED patients, to test the reliability and convergent validity of the EDE-Q, to examine group differences between various ED groups and healthy participants, and to explore the main behavioral features of Japanese ED patients using the newly developed Japanese version of EDE-Q. A total of 148 ED patients and 469 healthy participants completed the EDE-Q, Eating Attitudes Test-26 (EAT-26), and Eating Disorder Inventory-2 (EDI-2). The factor structure, reliability, and validity of the EDE-Q were assessed in ED patients. Group differences were assessed using the new Japanese version of the EDE-Q (EDE-Q-J). The EDE-Q-J had three factors. Cronbach's alphas ranged from 0.83 to 0.93. Total score and subscale scores of "Dieting" and "Bulimia and Food Preoccupation" of EAT-26 and of "Drive for Thinness," "Body Dissatisfaction," and "Bulimia" of EDI-2 correlated with the global score and three subscale scores of the EDE-Q-J. For Japanese female ED patients, the EDE-Q-J had three subscales that were not consistent with the original subscales, but were interpretable. It demonstrated sufficient reliability and validity. Japanese female patients with restricting-type anorexia nervosa (AN-R) displayed less dissatisfaction with shape and weight than healthy participants. AN-R patients in Japan might present with a non-fat-phobic symptom profile.