Neighbourhood characteristics associated with the geographic variation in laboratory confirmed COVID-19 in Ontario, Canada: a multilevel analysis
Abstract
Purpose
There is limited information on the role of individual- and neighbourhood-level characteristics in explaining the geographic variation in the novel coronavirus 2019 (COVID-19) between regions. This study quantified the magnitude of the variation in COVID-19 rates between neighbourhoods in Ontario, Canada, and examined the extent to which neighbourhood-level differences are explained by census-based neighbourhood measures, after adjusting for individual-level covariates (i.e., age, sex, and chronic conditions).
Methods
We conducted a multilevel population-based study of individuals nested within neighbourhoods. COVID-19 laboratory testing data were obtained from a centralized laboratory database and linked to health-administrative data. The median rate ratio and the variance partition coefficient were used to quantify the magnitude of the neighbourhood-level characteristics on the variation of COVID-19 rates.
Results
The unadjusted median rate ratio for the between-neighbourhood variation in COVID-19 was 2.22. In the fully adjusted regression models, the individual- and neighbourhood-level covariates accounted for about 44% of the variation in COVID-19 between neighbourhoods, with 43% attributable to neighbourhood-level census-based characteristics.
Conclusion
Neighbourhood-level characteristics could explain almost half of the observed geographic variation in COVID-19. Understanding how neighbourhood-level characteristics influence COVID-19 rates can support jurisdictions in creating effective and equitable intervention strategies.
Related articles
Related articles are currently not available for this article.