Near-Road Exposures to Traffic-Generated PM2.5, NOx and Black Carbon and the Risk of Daily Mortality: A Case-Crossover Study

Abstract

Background: A large proportion of the US population lives and works near highways where levels of traffic-generated air pollutants are the highest. While these pollutants have been linked to adverse health, typical exposure estimates assume similar time trends across space and do not account for localized differences in traffic.Aim: We aimed to quantify associations between short-term variations in traffic-generated PM2.5, NOx and black carbon (BC) and non-accidental, cardiovascular (CVD) and respiratory mortality in near-road populations using finely resolved spatiotemporal traffic data.Methods: We conducted a time-stratified case-crossover study of non-accidental, CVD, and respiratory mortality among individuals living within 1 km of all highways in the Washington State Puget Sound. Leveraging highly resolved spatiotemporal traffic data (monitored each half-mile at 5-minutes intervals) and the Research Line source dispersion model (R-LINE), we predicted hourly concentrations of traffic-generated PM2.5, NOx and BC for each subject. We then ran conditional logistic regression models, adjusted for weather, holidays, influenza, and background PM2.5.Results: For each inter-quartile range increase in 24-hour average pollution levels, we found higher odds of respiratory mortality (PM2.5: OR: 1.01, 95%CI: 0.95-1.08, NOx: 1.02, 95%CI: 0.94-1.09, BC: 1.02, 95%CI: 0.96-1.08) and lower odds of cardiovascular mortality (PM2.5: OR: 0.97, 95%CI: 0.94-0.99; NOx: 0.96, 95%CI: 0.9-0.99, BC: 0.97, 95%CI: 0.94-1.00). For both outcomes and pollutants, we found stronger associations among residents within 300 m of a highway (PM2.5: 1.08, 95%CI: 0.98-1.19 for respiratory mortality and 0.95, 95%CI: 0.91-0.99 for cardiovascular mortality). No associations were found for non-accidental or stroke mortality.Conclusion: Short-term exposures to traffic-generated air pollution may influence the odds of mortality in near road populations, though not always in the expected direction.