A factor analytic comparison of three commonly used depression scales (HAMD, MADRS, BDI) in a large sample of depressed inpatients

Abstract

Abstract Background Quantifying depression mainly relies on the use of depression scales, and understanding their factor structure is crucial for evaluating their validity. Methods This post-hoc analysis utilized prospectively collected data from a naturalistic study of 1014 inpatients with major depression. Confirmatory and exploratory factor analyses were performed to test the psychometric abilities of the Hamilton Depression Rating Scale, the Montgomery Asberg Depression Rating Scale, and the self-rated Beck Depression Inventory. A combined factor analysis was also conducted including all items of all scales. Results All three scales showed good to very good internal consistency. The HAMD-17 had four factors: an "anxiety" factor, a "depression" factor, an "insomnia" factor, and a "somatic" factor. The MADRS also had four factors: a “sadness” factor, a neurovegetative factor, a “detachment” factor and a “negative thoughts” factor, while the BDI had three factors: a "negative attitude towards self" factor, a "performance impairment" factor, and a "somatic" factor. The combined factor analysis suggested that self-ratings might reflect a distinct illness dimension within major depression. Conclusions The factors obtained in this study are comparable to those found in previous research. Self and clinician ratings are complementary and not redundant, highlighting the importance of using multiple measures to quantify depression.