Exploring COVID-19 Morbidity and Mortality During the First Three Epidemic Waves in Ontario, Canada: A One Health Perspective to Assessing Risk

Abstract

Livestock farming serves to support human sustenance and livelihood, but these systems also emit atmospheric particulate matter ≤ 2.5 µm (PM2.5) and ammonia (NH3), which are known respiratory stressors. Over three epidemic waves in Ontario, Canada, prolonged exposure to PM2.5 and NH3 were explored as risk factors for COVID-19 incidence and mortality. Through multilevel negative binomial principal component (PC) regression modeling, regional variations in PM2.5 were positively associated with COVID-19; the strength of this association declined as the pandemic continued. Compared to livestock farming, fuel combustion appeared to have had a more prominent role in the observed association of PM2.5 with COVID-19. There was a minor inverse association between NH3 and COVID-19, suggesting that livestock farming communities, as opposed to more urbanized communities, had a tendency toward a decreased risk of COVID-19 health outcomes; this result may reflect confounding. In this thesis, PC regression served as an effective tool for enabling a robust One Health risk factor analysis. PC regression can be recommended for studying intricate relationships in the One Health context.