BACKGROUND Associations between different types of depression with clinical characteristics and putative vulnerability factors from several domains (health, disability, personality, familial psychopathology) were studied in a sample of elderly subjects, in order to find arguments that support or discard the notion of discontinuity in correlates of depression. METHODS Subjects were 183 depressed elderly persons from the general population, general practitioners and mental health care outpatient clinics, and a random elderly control group (n=74). Depression was subdivided into symptomatic (n=45), DSM-IV minor (n=73), and DSM-IV major (n=65) depression. The groups were compared with respect to a variety of variables from different domains. RESULTS Most variables showed a marked difference between the control group and the depressive groups, and smaller or no differences between depressive subtypes. CONCLUSIONS The data were collected cross-sectionally. CONCLUSIONS With a few notable exceptions, all subtypes of depression were significantly different from the control group, while differences between depressive subtypes were generally smaller. Depressive subtypes showed no discontinuum with respect to the associated variables. Similar associations in different subtypes of depression may indicate common vulnerability factors.