Reliability and validity of the Global Assessment of Functioning Scale in clinical outpatients with depressive disorders

Abstract

The Global Assessment of Functioning Scale (GAF) is widely used to assess psychological, social and occupational functioning. The validity and reliability of the GAF in clinical practice have only scarcely been studied in naturalistic samples. A total of 432 outpatients with a current major depressive disorder (MDD) were evaluated with routine outcome monitoring (ROM). At baseline the GAF score was assessed by the treating clinician and at ROM baseline and follow-up sessions also by a trained test nurse. Sociodemographic data, the Mini International Neuropsychiatric Interview Plus and scores on the Montgomery-Äsberg Depression Rating Scale, Beck Depression Inventory-revised, Brief Symptom Inventory and Short Form-36 were assessed. At baseline, the mean GAF score by the clinician was 54.8 (range 35-85), and this was systematically lower than the mean GAF score by the test nurse of 57.5 (range 31-88). GAF scores by the clinician and test nurse correlated weakly (r = 0.26). The GAF scores of the clinicians correlated strongly with disease severity, and social and physical functioning. The GAF showed rather poor inter-rater reliability as well as poor discriminant validity with disease severity and physical limitations in a large naturalistic sample of outpatients with MDD.