Multi-pollutant case-crossover models of all-cause and cause-specific mortality and hospital admissions by age group in 47 Canadian cities.

Most of the existing epidemiological studies have investigated adverse health effects of multiple air pollutants for a limited number of cities, thus the evidence of the health impacts is limited and it is challenging to compare these results because of different modeling approaches and potential publication bias. In this paper, we expand the number of Canadian cities, with the use of the most recent available health data. A multi-pollutant model in a case-crossover design is used to investigate the short-term impacts of air pollution on various health outcomes in 47 Canadian main cities, comparing three age groups (all-age, senior (age 66+), non-senior). The main findings are that a 14 ppb increase of O3 was associated with a 0.17%-2.78% (0.62%-1.46%) increase in the odds of all-age respiratory mortality (hospitalization). A 12.8 ppb increase of NO2 was associated with a 0.57%-1.47% (0.68%-1.86%) increase in the odds of all-age (non-senior) respiratory hospitalization. A 7.6 μgm-3 increase of PM2.5 was associated with a 0.19%-0.69% (0.33%-1.1%) increase in the odds of all-age (non-senior) respiratory hospitalization.