Connectivity between long-term care homes and subsequent SARS-CoV-2 outbreaks.

BACKGROUND

To examine the relationship between individual workers employed at more than one LTCH (inter-LTCH connectivity) across LTCH and SARS-CoV-2 outbreaks.

METHODS

We conducted a retrospective cohort study using secondary, aggregate data (surveillance and mobile geolocation data) from 179 LTCH in the Greater Toronto Area of Ontario, the province where close to one-third of the Canada's SARS-CoV-2 cases among long-term care homes residents were reported, between 2020-02-26 and 2020-08-31. The main exposure of interest was the inter-LTCH connectivity, generated from geographic location data procured across mobile apps. Three outcomes were examined: 1) at least one SARS-CoV-2 diagnosis among residents, 2) cumulative cases among residents in each facility, and 3) time to first outbreak.

RESULTS

The median degree of connectivity for LTCH that experienced an outbreak (59%; 106/179) was 1.2 times the degree of those without an outbreak (6 compared to 5). LTCH with higher inter-LTCH connectivity also had larger numbers of residents and beds, and were more likely to have for-profit ownership. After adjusting for facility-level and neighbourhood-level factors, every additional connection to another LTCH increased the odds of an outbreak in the respective LTCH by 8% (adjusted odds ratio=1.08, 90% credible interval [CrI]: 1.02-1.09). Inter-LTCH connectivity was also associated with higher risk of earlier occurrence of a first SARS-CoV-2 case (adjusted hazard ratio=1.05, 90%CrI: 1.02-1.09), but not with outbreak size.

CONCLUSIONS

Staff cohorting was associated with reduced importation risk of SARS-CoV-2 cases into LTCH. However, once importation has occurred, other facility-level factors including facility infrastructure and staff benefits are more important in shaping outbreak size. Implementing these structural strategies to meet the LTCH workers and residents' needs are pivotal to prevent and manage future respiratory virus outbreaks.