Content area
Abstract
Outdoor air pollution, including fine particulate matter (PM2.5) air pollution, contributes to a range of adverse health outcomes and has a large population health impact. However, the standard method of measuring exposures to particulate air pollution as a mass concentration has limitations. Recently, emerging measures have been developed that account for the composition and toxicity of particles. The overall aim of this thesis was to describe within-city spatial variations in newly-developed measures of particle composition and toxicity (including multiple measures of particle oxidative potential as well as a measure of exposure to magnetite nanoparticles) across Canadian urban areas and to assess their effects on long-term health outcomes. To accomplish this aim, we completed three objectives that constitute the body of this manuscript-based thesis.
In Objective 1, we conducted monitoring campaigns at 124 sites in Montreal and 110 sites in Toronto, Canada to collect pollutant data, and developed land-use regression models to predict the spatial distributions of PM2.5 oxidative potential, production of reactive oxygen species, and magnetite nanoparticles. We used Bayesian lasso regression models with land-use characteristics from Geographic Information Systems databases to predict pollutant measures at unobserved points in order to create high-resolution exposure surfaces. We observed high spatial variability of oxidative potential measures (coefficients of variation 42.0-66.0%) and magnetite (coefficients of variation 69.7-75.4%) within each city relative to PM2.5 mass concentration (coefficients of variation 24.3-30.8%). Multivariable land-use regression models predicted elevated concentrations of oxidative potential, reactive oxygen species generation, and magnetite around highways, railways, and road intersections.
In Objective 2, we applied the estimates of exposure obtained in Objective 1 to determine how oxidative potential and the ability of particles to generate reactive oxygen species (ROS) may modify the relationship between long-term exposure to oxidant gases and cardiovascular mortality. We performed a retrospective cohort study of participants in the Canadian Census Health and Environment Cohort who lived in Toronto or Montreal. We used Cox proportional hazards models to estimate associations between outdoor concentrations of oxidant gases (Ox, a redox-weighted average of nitrogen dioxide and ozone concentrations) and cardiovascular deaths. Analyses were performed across strata of PM2.5 oxidative potential and ROS concentrations. We observed that spatial variations in outdoor Ox were associated with an increased risk of cardiovascular mortality (HR per 5 ppb = 1.028, 95% CI: 1.001, 1.055). The effect of Ox on cardiovascular mortality was stronger above the median of each measure of PM2.5 oxidative potential and ROS concentration (e.g., above the median of glutathione-based oxidative potential: HR = 1.045, 95% CI: 1.009, 1.081; below median: HR=1.000, 95% CI: 0.960, 1.043).
In Objective 3, we performed a retrospective cohort study in the Canadian Census Health and Environment Cohort to estimate associations between long-term exposure to magnetite nanoparticles in PM2.5 and the incidence of brain cancer. Cox proportional hazards models were used to estimate the association between exposure to magnetite nanoparticles in outdoor PM2.5 and incidence of brain cancer in Montreal and Toronto. We found no significant relationship between exposure to magnetite particles and incidence of malignant brain tumours (HR per IQR = 0.998, 95% CI: 0.988, 1.009). Moreover, we found no significant effect of PM2.5 or NO2 on brain cancer incidence





