Abstract
During the first wave of COVID-19, three-quarters of Canadian deaths were among those age 80 and older. We examined whether age, chronic disease load, sex, or place was the strongest predictor of such deaths. A cross-sectional analysis of administrative data from 1 January 2020 to 30 October 2020 for the population of Ontario (n = 15,023,174) was performed. Using logistic regression analysis, we determined whether place of residence (community dwelling, community dwelling with formal home care, or long-term care facility), age group, sex, or chronic disease burden was most strongly associated with the outcome of death within 60 days of a positive SARS-CoV-2 PCR test. Overall, there were 2766 deaths attributed to COVID-19. The age-related odds of dying increased from 6.1 (age 65-74) to 13.4 (age 85 or older) relative to those aged <65 years. This age effect was dwarfed by an odds ratio of 117.1 for those living in long-term care versus independently in the community, adjusted for age, sex, and chronic disease burden. The risk of death from COVID-19 aligned much more with social realities than individual risks. The disproportionate mortality arising specifically from institutional residence demands action to identify sources and ameliorate the harms of living in such facilities.