Use of the Hamilton and Montgomery-Asberg depression scales in institutionalized elderly patients

In screening for a study of drug treatment of major depression, the authors obtained data on depressive symptoms in elderly residential care patients (N = 116; average age 84 years; 81% women). Principal-components analysis (with varimax rotation) of the Hamilton Rating Scale for Depression yielded a four-factor solution (accounting for 47.2% of variance): core depression, anxiety, insomnia-hypochondriasis, and cognitive-ideational symptoms. The Montgomery-Asberg Depression Scale yielded two factors (54% of variance)-core depression and anxiety. Core depression factor scores from both scales (but not other factor scores) predicted mortality. The association of core depression with mortality (in subsets of patients for which data on these covariates were available) remained significant after measures of illness burden and disability were controlled.