Some of the variables associated with major depression in community samples may be nonspecific correlates of mental ill-health. The study objective was to evaluate this hypothesis by comparing two groups of psychiatric patients with the general population. Subjects were recruited from psychiatric inpatients at a general hospital in Calgary, Alberta, Canada, by dissemination of a consent form. Consenting subjects completed the mood disorders section of the Composite International Diagnostic Interview (CIDI). All subjects were interviewed to evaluate a set of variables that may be risk factors for major depression. The measurement instruments were identical to those used in a national survey so that comparisons to population-based data were possible. The psychiatric inpatients differed dramatically from community subjects in terms of stress, traumatic life events, recent life events, and social support. However, differences were not observed between inpatients with major depression (according to the CIDI) and inpatients without major depression. Associations between certain variables and major depressive disorders in community populations may reflect nonspecific associations with mental ill-health. This may occur because of nonspecific impacts of these variables on the etiology or prognosis of mental disorders or some nonspecific impact of mental illness itself.