Attributing School Air Pollution Levels to Features of the Built Environment in New York City

Abstract

Background: Exposure to traffic-related air pollution (TRAP) has been linked to asthma prevalence and exacerbations. Children spend a significant portion of their day in school where exposure to pollutants can be high based on geographic features of the built environment. The study objective was to determine the distributions of geographic indicators of traffic and built environment features and the association with school-based TRAP exposure in New York City (NYC).Methods: Spatial data on the proximity and density of roadways and built environment features were examined for a 250-meter buffer around NYC elementary and middle schools (N=1184). Annual average levels of particulate matter <2.5 microns (PM2.5), black carbon (BC), and nitrogen dioxide (NO2) were determined based on land use regression models of street-level pollution measured by the New York City Community Air Survey. Spatial data were categorized using k-means cluster analysis and clusters were associated with air pollution using analysis of variance.Results: Based on significant correlations with air pollutants, three geographic indicators were included for cluster analysis: number of bus stops; truck route density (total km route/km2 land); and percentage of area covered with trees (i.e. tree canopy). Four clusters were identified: 1. Higher truck route density (n=103 schools); 2. Higher number of bus stops (n=378); 3. Higher tree canopy (n=250); and 4. Lower in all 3 indicators (n=453). In general, the levels of PM2.5, BC, and NO2 were highest among schools with higher truck route density, followed by higher number of bus stops, and lowest with higher tree canopy (p=0.001).Conclusions: School-aged children may experience higher levels of air pollution based on their school’s proximity to geographic indicators of diesel vehicle emission. Geographic indicators may serve as surrogates of chronic school exposure to air pollution. Future studies will determine if clusters are associated with school asthma prevalence.