Investigating the role of transport models in epidemiological studies of air pollution exposure and health effects

Urban air pollution has been linked with a range of health effects; in metropolitan areas on-road traffic is the largest contributor to local air pollution and therefore, exposure to traffic emissions could potentially explain the odds of air pollution related health effects among the general population. In this study, we investigate the possibility of using a transportation model developed for the City of Montreal for the purpose of deriving a measure of individual exposure to traffic emissions. This model has been extended with capability for modelling transport emissions at the individual and household levels. The resulting transport emissions are distributed across the modelling domain and validated against air pollution levels previously derived using a Land Use Regression (LUR) technique, a commonly used method for deriving individual exposure in epidemiological studies. We observe a medium and significant correlation between both datasets with strong correlations in the neighbourhood of all roads. Exposures were also derived for a sample of breast cancer survivors in the Montreal area using both methods (LUR and transportation model). We observe comparable odds ratios (OR) for breast cancer suggesting that a transport model can potentially act as a cost effective alternative to air pollution monitoring for epidemiological analysis of air pollution and its health effects.