Determinants of place of death: a population-based retrospective cohort study

Background: As Canada's population ages, the location of end of life care (whether at home, extended care facility or hospital) may change depending on the location of death. We carried out a study to identify determinants of the place of death. Methods: Data on deaths in British Columbia between 2004 and 2008 were obtained from the Vital Statistics Agency. Place of death was categorized into home, extended care facility, hospital or other. Logistic regression analyses were used to estimate the effects of age, sex, marital status, residence, place of birth and cause of death on place of death using adjusted odds ratios and 95% confidence intervals (95% CI). Results: Of the 153,111 deaths in the study, 16.5% occurred at home, 29.0% in extended care, 51.0% in hospital and 3.5% occurred elsewhere. Male deaths were less likely to occur in extended care as compared with female deaths (odds ratio 0.73, 95% CI 0.71-0.75). Age (odds ratio 3.31, 95% CI 3.19-3.45 for those for >= 90 vs 70-79 years), marital status (odds ratio 1.42, 95% CI 1.38-1.47 widowed vs married), residence (odds ratio 0.80, 95% CI 0.76-0.83 rural vs Vancouver), place of birth (odds ratio 0.80, 95% CI 0.75-0.86 China vs Canada) and cause of death (odds ratio 3.91, 95% CI 3.69-4.13 dementia vs cancer) were also associated with death in extended care. Conclusions: Information on determinants of place of death can inform public health policy regarding care at the end of life and make resource allocation more efficient.