Background. There are few epidemiologic studies of the rate of antidepressant use in the elderly population, especially for community residents. We report findings on antidepressant use in the elderly population using data from a national survey in Canada which drew samples from both the community and institutional settings. Methods. Data for the present study came from the Canadian Study of Health and Aging (CSHA), a national prevalence study of dementia in which information was collected from 2914 elderly subjects on current drug use, place of residence (community, institution), depression, dementia, and self-reported health. Survey weights were constructed to reflect the national population and data were analyzed using the SUDAAN statistical software package. Results. The rate of antidepressant use was 4.1% (community 3.1%, institution 16.5%). Of those who were depressed, 9.4% were taking an antidepressant (community 4.2%, institution 36.0%). A logistic regression analysis showed that female gender, living in an institution, the presence of dementia, and the presence of a chronic physical disease, but not depression, were associated with increased antidepressant use. Conclusions. Our findings on the rate of antidepressant use in the elderly population are consistent with and extend previously published reports. We found evidence of underutilization of antidepressants in the treatment of geriatric depression, especially for community residents. However, this evidence needs to be interpreted with caution as the CSHA data on depressive symptoms were incomplete.