You are here:
EVALUATING HETEROGENEITY IN INDOOR AND OUTDOOR AIR POLLUTION USING LAND-USE REGRESSION AND CONSTRAINED FACTOR ANALYSIS
Impact/Purpose:
Epidemiologic studies of exposure to air pollution have typically relied on data from centrally located ambient air quality monitors. However, such data are not sufficient for capturing the spatial variability of pollutant concentrations at the local scale, in particular at the within-city, or intra-urban, scale at which traffic-related air pollution is both highest and most variable. The ideal approach would be to measure each individual’s personal exposure to traffic-specific pollutants over time, but this is difficult, intrusive, expensive, and generally not feasible for very large populations. Investigators have consequently sought ways to predict, or to model, individual-level exposures from more readily available data.
Dr. Jonathan I. Levy of the Harvard School of Public Health has proposed an approach to extend and improve upon existing GIS-based methods for predicting intra-urban exposures. An underlying goal of his study was therefore to explore ways to reduce exposure-measurement error and to improve the accuracy and precision of the associations reported in epidemiologic studies of air pollution.
Description:
Investigators will explore how land-use regression and source-apportionment techniques can be used to characterize individual-level exposure to both indoor and outdoor air pollution sources. Investigators will utilize health and air monitoring data from an ongoing prospective cohort study on childhood asthma in Boston, Massachusetts to model variability in outdoor and indoor residential air pollution, identify potential sources, and evaluate the effectiveness of various indoor exposure surrogates for predicting childhood asthma development. This study will also evaluate how these approaches might reduce error in individual exposure assessment and thereby improve epidemiologic estimates of the effects of traffic-related air pollution on health.