Environmental factors associated with Baylisascaris procyonis infection from a population of raccoons in Toronto, Ontario, Canada

Although researchers often investigate differences in pathogen prevalence and transmission between urban and non-urban environments, little work has been done examining factors associated with variation observed within an urban population. Many wildlife species that are successful in urban environments carry zoonotic pathogens, and their proximity to human populations can increase the risk of human exposure. One such pathogen is Baylisascaris procyonis, the raccoon roundworm, the larval stage of which can cause severe neurological disease in humans and many other bird and mammal species. We investigated the relationships between environmental factors, human land use and land cover types, and raccoon and human demographic factors, and the prevalence of B. procyonis infection in 391 raccoons submitted to the Canadian Wildlife Health Cooperative from Toronto, Ontario, Canada, between 2013 and 2016. Using mixed multivariable logistic regression models, we identified five variables that were significantly associated with B. procyonis infection in raccoons, after controlling for host age, host sex, and season of collection. Total green space was negatively associated with the odds of infection. Residential zoning demonstrated a quadratic relationship, with the greatest odds of infection associated with moderate levels of residential zoning. We identified significant two-way interactions between sex and the presence of a ravine, age and the building land cover classification, and age and the amount of grass/shrub coverage. These relationships emphasize the complexity of host-pathogen relationships in human-altered environments. Thus, development of risk mitigation strategies may be challenging; landscape modification could have unexpected outcomes on infection prevalence in some wildlife populations.