Eating Disorder Examination Questionnaire (EDE-Q) in Norwegian Adults: Discrimination between Female Controls and Eating Disorder Patients

Abstract

Objective: To determine the optimal Eating Disorder Examination-Questionnaire (EDE-Q) global score to discriminate between female controls and patients by eating disorder (ED) diagnosis, body mass index (BMI) and age.

Method: A sample of 1845 control participants and 620 patients from specialty ED treatment centres.

Results: Mean global EDE-Q was 4.00 [standard deviation (SD) = 1.32] for patients and 1.25 (SD = 1.10) for controls. Receiver operating characteristic analyses demonstrated an area under the curve of 0.93 (95% CI: 0.91-0.94), with an optimal cut-off score of 2.50 (sensitivity = 0.86; specificity = 0.86), ranging from 2.09 for anorexia nervosa, 2.62 for bulimia nervosa and 2.63 for ED otherwise not specified. Optimal cut-off scores also varied according to BMI, ranging from 1.62 (BMI ≤ 18.0 kg/m(2) ) to 3.26 (BMI ≥ 30 kg/m(2) ), with less variability for age, ranging inversely from 2.16 (>40 years) to 2.70 (<20 years).

Discussion: The global EDE-Q score showed high discriminant validity, and findings illustrate the particular importance of considering BMI and diagnosis when applying cut-offs based upon the EDE-Q.

Keywords: Eating Disorder Examination-Questionnaire; ROC analysis; assessment; cut-off score; discriminant validity.