Markers of community outbreak and facility type for mitigation of COVID-19 in long-term care homes in Ontario, Canada: Insights and implications from a time-series analysis

Purpose: The resident deaths among Long Term Care Home (LTCH) accounted for more than 65% of total deaths in the province of Ontario, Canada, during March 29 to June 3, 2020, yet not all LTCHs were severely affected. Methods: We carried out a retrospective cohort study, with case control for questions for which data allowed, with LTCH COVID-19 databases obtained from Ontario's Ministry of Long Term Care. We performed a combined temporal and spatial data analysis of COVID-19 cases and deaths among LTCH residents, identified trends, contributing factors, and early markers of LTCH outbreak severity. Results: Our analysis shows that for -profit LTCHs had higher death-to-bed ratio, also with an average rate of increase of death-to-bed ratio higher for for -profit homes than other types of management. We find from uni- and multi -variable analyses (linear and nonlinear) that staff infection has the strongest association with death-to-bed ratio from among the descriptor variables considered, reflecting the risk of the disease in the health region/ community. We also identify a delay of up to 8 days between the trends in fatalities among individuals outside LTCHs and that of LTCH residents. We did find an association between policy change to single LTCH/staff and reduction in weekly LTCH resident death, albeit with an expected time delay of about 7-10 days. Conclusions: The association between the risk of COVID-19 in the health region and the deaths among LTCH residents, and the delay between fatality among individuals residing outside and inside LTCHs suggests that fatality in a health region could be a predictor of outbreak in LTCHs within the same health region.