Abstract
BACKGROUND: Orthorexia nervosa (ON) is defined as a pathological fixation on eating healthy and pure food. In this study, it was aimed to evaluate the psychometric properties of the Turkish version of the Düsseldorf orthorexia scale (DOS) in a clinical sample. METHODS: A total of 385 individuals, 117 with generalized anxiety disorder (GAD), 108 with major depressive disorder (MDD), 56 with obsessive-compulsive disorder (OCD) and 104 healthy controls, participated in the study. Hamilton Depression Rating Scale (HAM-D), Hamilton Anxiety Rating Scale (HAM-A), Yale-Brown Obsessive-Compulsive Scale (Y-BOCS), Eating Attitude Test (EAT-40), Orthorexia Nervosa Inventory (ONI) and Düsseldorf Orthorexia Scale (DOS) were applied to all participants. RESULTS: The DOS demonstrated high internal consistency (Cronbach's alpha = 0.87) and good construct validity. Confirmatory factor analysis supported a one-factor structure, explaining a significant portion of variance among responses. In the analyses performed to test the convergent validity of DOS, a positive correlation was found with ONI and EAT scores (p < 0.001 for each). There was no significant correlation between DOS and HAM-D and HAM-A scores (p > 0.05 for each). The severity of orthorexic symptoms measured by DOS was similar between each patient group and healthy control group. DOS was applied to 70 more participants two weeks apart and the test-retest reliability was determined as 0.99. CONCLUSIONS: This study shows that the Turkish version of DOS is valid and reliable in clinical samples and healthy individuals, is largely consistent with ONI, and that this scale can be used in studies investigating ON.