The main aim of the present study was to examine whether the well-established association between depression and social dysfunction still remains when effects of a coexistent anxiety disorder are eliminated from the data. As these effects strongly depend on the proportion of depressed subjects suffering simultaneously from an anxiety disorder, we first examined the frequency of mixed and pure depressive disorders and that of pure anxiety disorders (control subjects) in a community sample (n = 483). Using DIS/DSM-III criteria (reference period 6 months), pure anxiety disorders were most frequent (6%), followed by pure depressive disorders (3%) and the coexistence of anxiety and depression (2%). Cases suffering from both disorders were most severely afflicted in terms of psychopathology (persistence of symptoms, comorbidity regarding other mental disorders). At the diagnostic level, the association between depression and social dysfunction was only slightly influenced by effects resulting from comorbidity; at the level of actual symptoms, however, we found that cases suffering simultaneously from severe depression and severe anxiety were significantly more handicapped in their social lives than depressive subjects with only mild anxiety symptoms.