The gender gap in depression reconsidered: the influence of marital and employment status on the female/male ratio of treated incidence rates.

BACKGROUND: The consistently observed predominance of female over male rates in depression - in treated as well as in untreated populations - has never been satisfactorily explained. Among the many possible biological and psychosocial explanations, marital and employment status have not been extensively studied and virtually nothing is known about the combined effect of these variables on sex differences in depression. A main reason for this lack of knowledge is the limited number of cases available in epidemiological studies. METHODS: The present paper examines the combined effects of marital and employment status on sex differences in depression rates by analysing in-patient admission rates of all depressed patients aged 18-67 admitted for the first ever time to the Department of Psychiatry of the University of Vienna from a strictly defined catchment area over a period of 42 months. Stepwise Poisson regression analyses were carried out in order to identify the relative contribution of these variables to the variance of first ever in-patient admission rates for depression. RESULTS: A total of 2599 depressed patients fulfilled the inclusion criteria. When analysed separately, sex, marital status and employment status were shown to have distinct influences, with the "not married" carrying a two-fold higher risk than the married (2:1), and female sex (1.7:1) as well as not being employed (1.7:1) showing similar but smaller effects. In the detailed combined analysis, marriage was significantly less advantageous for women than for men, while sex differences disappeared completely in the widowed group. Also, there was no sex difference in the employed divorced; in the employed widowed there was even a slight preponderance in men. The highest rates were found in not employed divorced women, the lowest in employed married men. CONCLUSION: While in-patient admissions are certainly selective in relation to epidemiological data, the large sample made it possible to perform combined analyses of sex, marital status and employment status. It was shown that the statement of a female preponderance in depression, which was found for the total sample, is a gross oversimplification. If marital and employment status are considered simultaneously, the sex differences disappear in some subgroups and in some are even reversed. We suggest that the combined influence of marital and employment status should be studied in epidemiological studies as well before conclusions about the influence of sex on depression rates are drawn.