Few studies of the risks of institutionalization of the elderly have had fully specified models using multivariate analysis, and several studies have examined highly selected populations, making their generalizability uncertain. We set out to examine the risks of institutionalization in elderly people in Nova Scotia. A case-control study, executed as part of the Canadian Study of Health and Aging examined 108 incident institutional cases and 533 community-dwelling elderly controls, using a standardized assessment interview conducted by trained interviewers. Multiple logistic regression analysis showed that advancing age, presence of dementia, functional impairment, poor self-rated health, recent hospital admission, and absence of a caregiver were important risks for entry into long term care. Institutions providing long-term care for the elderly need to be able to look after populations with a high prevalence of dementia and functional impairment.
