Predictive validity of subtypes of chronic affective disorders derived by cluster analysis

This study aimed to examine the predictive validity of the cluster-analytically derived 3 subtypes of chronic affective disorders that were proposed on the basis of cross-sectional symptoms and were shown to have some external validity with regard to psychosocial characteristics. Two-year naturalistic follow-up data were available for 38 (97%) of the original 39 patients with chronic affective disorders. The 3 subtypes, which were named psychotic subtype, late-onset female subtype and depressive personality subtype in the previous analyses, did not differ in their depressive severity or social functioning upon entry. In the course of the 2-year follow-up, however, there were significant differences in the Global Assessment of Functioning scores as well as the social adaptation in the economic sphere according to the Social Adjustment Scale-Self Report. All these indices revealed the psychotic subtype to have the worst outcome, followed by the depressive personality subtype, and the late-onset female subtype fared best.