TRAFFIC-RELATED AIR POLLUTION AND CHILDREN’S RESPIRATORY HEALTH: DEVELOPING AND REFINING ESTIMATES OF RESIDENTIAL EXPOSURE

Abstract

Listen

Translate article

ISEE-521 Introduction: Several epidemiological studies have found associations between residential proximity to busy roads and adverse health effects. We recently conducted a school-based, cross-sectional study to examine respiratory health among children (8 to 11 years) living and attending schools at varying distances from high-traffic roads. Air pollutants, including particulate matter (PM10, PM2.5), black carbon (BC), and nitrogen oxides (NOX, NO,), were measured at each of ten schools. We found associations between traffic pollutants (especially BC, NOX, and NO {NOX- NO2}) and asthma and bronchitis episodes in the past 12 months, using a two-stage hierarchical model. In our analysis to date, we used pollutant levels measured at the school sites as estimates of the children’s exposure to traffic emissions. To refine exposure estimates, we obtained additional within-neighborhood measurements of NOX and NO2. We report here on the development of new measures of residential traffic exposure based on geographic information systems (GIS) technology. These metrics will then be compared with observed traffic pollutant concentrations. Methods: We utilize information on both school or residential address and traffic data using geographic information systems (GIS) technology to calculate various traffic metrics, including closest traffic volume within a given buffer radius, maximum traffic volume, and distance-weighted traffic density. Second, we develop a model incorporating prevailing wind direction and distance. Third, we present results of a spatial regression model similar to that developed by Brauer et al. (Epidemiology 2003; 14:228-39). This model uses information on several factors, such as local traffic density, distance to the highway and traffic counts, to explain the variation in school and neighborhood NOx/NO2 concentrations that were collected. Our last exposure measure is based on self-reported assessment of traffic from our intake questionnaires. Results: The GIS-based traffic metrics and spatial regression models will be presented and compared with previous results obtained using school-based or residential exposure measurements. Ultimately, these measures will be used to further explore associations between both bronchitis and asthma symptoms. Discussion: Different metrics of exposure to traffic pollutants can be used to explore exposure misclassification. By reducing measurement error, we will be able to elucidate more clearly the relationships of traffic to respiratory health outcomes. The results will help determine the relative importance of different approaches to refining exposure estimates and, in so doing, will provide methodological guidance for future traffic studies. Information generated from this study will be useful for both ambient air quality standard-setting and pollution control strategies.