Secondary Logo

Journal Logo

Associations Between Spatially Resolved Estimates of Traffic-related Pollution and Acute Morbidity: Assessing Agreement of Results Among Multiple Exposure Assignment Approaches

Sarnat, Jeremy A.1; Sarnat, Stefanie Ebelt1; Crooks, James2; Isakov, Vlad2; Touma, Joe2; Ozkaynak, Haluk2; Mulholland, James3; Russell, Armistead3; Kewada, Priya1

doi: 10.1097/01.ede.0000391755.25463.0c
Abstracts: ISEE 22nd Annual Conference, Seoul, Korea, 28 August–1 September 2010: Enhancing Exposure Assessment for Air Pollution Health Studies: EPA/NERL Cooperative Agreement Program

1Emory University, Atlanta, GA; 2US Environmental Protection Agency, Research Triangle Park, NC; and 3Georgia Institute of Technology, Atlanta, GA.

Abstracts published in Epidemiology have been reviewed by the societies at whose meetings the abstracts have been accepted for presentation. These abstracts have not undergone review by the Editorial Board of Epidemiology.


Back to Top | Article Outline


A common method for assigning exposure in population-based epidemiologic studies of air pollution is to use surrogate measurements from central monitoring sites. This practice may introduce exposure misclassification in epidemiologic analyses for pollutants that are spatially heterogeneous (ie, nitrogen dioxide, elemental carbon, carbon monoxide). To investigate the potential impact of misclassification within observed health risk estimates, we created several tiers of exposure refinement to estimate daily ambient concentrations at 225 Atlanta-area ZIP code centroids. These tiers include estimates that vary in their approaches for modeling pollutant spatial heterogeneity, as well as surrogates of pollutant infiltration.

Back to Top | Article Outline


The current analysis presents results using the following 6 exposure tiers: (i) central site monitoring data, (ii) spatially interpolated monitoring data, (iii) Community Multiscale Air Quality model outputs, (iv) AERMOD atmospheric dispersion modeling outputs, (v) Community Multiscale Air Quality-AERMOD hybrid and blended model outputs, and (vi) spatial and temporal surrogates of air exchange rates. Each metric was applied in (1) a daily time-series study of emergency department visits and (2) a case-crossover study of cardiac events in patients with implantable cardioverter defibrillators.

Back to Top | Article Outline


Varying degrees of spatial and temporal variability were observed for different exposure metrics. Variability was more pronounced for traffic-related (CO, NOX) versus regional (O3, PM2.5) pollutants.

Back to Top | Article Outline


Preliminary analyses obtained for each exposure metric show that finer spatial resolution in ambient concentrations and incorporation of exposure factors in analyses yield different pollutant exposure estimates compared to ambient monitoring data alone. We will also discuss the interpretations and limitations of each exposure assignment method for air pollution epidemiology.

© 2011 Lippincott Williams & Wilkins, Inc.