Psychometric evaluation of the youth eating disorder examination questionnaire in children with overweight or obesity

Abstract

This study evaluated the psychometric properties of the Youth Eating Disorder Examination Questionnaire (YEDE-Q) and its utility for detecting loss of control (LOC) eating (i.e., eating episodes, regardless of size, involving a perceived inability to control what or how much one is eating) among school-age children with overweight or obesity. Identifying eating pathology, particularly LOC eating, in this population may facilitate treatment that improves weight outcomes and reduces eating disorder risk. Children with overweight or obesity (N = 241; 7-11 years) completed the YEDE-Q and abbreviated Child EDE (ChEDE) to assess LOC eating, prior to entering a weight management treatment trial. Confirmatory factor analyses (CFA) were conducted on children's YEDE-Q responses and compared to the standard adult EDE-Q factor structure and newer, alternate factor structures. CFA supported a three-factor structure, which distinguished youth with versus without LOC. The YEDE-Q showed low accuracy for detecting LOC eating as measured by the ChEDE, which served as the gold-standard benchmark (AUC = 0.69). Among children who endorsed LOC eating, more episodes per month were reported on the YEDE-Q than ChEDE (p < .001). The YEDE-Q may not have utility as a screener for identifying true cases of LOC eating among school-age children with overweight or obesity. Further evaluation of the YEDE-Q and the alternate three-factor structure is warranted.

Keywords: assessment; children; eating disorder pathology; loss of control eating; overweight.

FIGURE 1

Receiver operating characteristic analysis for cases screening positive for loss of control eating on the Youth Eating Disorder Examination Questionnaire compared to the Child Eating Disorder Examination. Area under the curve = 0.69 (SE = 0.04; Asymptotic 95% CI = 0.62 – 0.77); Asymptotic p < .001