Satellite-Derived Estimates of Ambient Fine Particulate Matter and Pediatric Emergency Department Visits in Georgia, 2002-2010: a Case-Crossover Study

Abstract

Introduction: There is substantial evidence that ambient particulate matter less than 2.5 microns in diameter (PM2.5) plays a role in the exacerbation of asthma and respiratory symptoms. Methods: ZIP code level emergency department (ED) visits among children in the U.S. state of Georgia were obtained from 1/1/2002 to 6/30/2010. Outcomes of interest were asthma or wheeze (N=524,760), bronchitis (N=67,632), otitis media (N=548,197), pneumonia (N=120,498), and upper respiratory infection (N=1,021,685). Daily 24-hour average satellite-derived estimates of PM2.5 were estimated using land use variables, meteorology, and remotely sensed aerosol optical depth measurements from the Multiangle Implementation of Atmospheric Correction (MAIAC) algorithm at 1-km resolution. Daily grid-level estimates were averaged to obtain daily ZIP code-level estimates (N=691 ZIP codes); the ZIP code-level estimates were used in the epidemiologic analysis if at least 70% of the grid-level estimates were non-missing. Overall and urban/suburban/rural-specific relative risks (RR) (based on the NCHS urban-rural classification scheme) for a 10 µg/m3 increase in same day PM2.5 concentration were estimated using conditional logistic regression with matching variables ZIP code, year, and month in the framework of a case-crossover analysis. Results: RRs for a 10 µg/m3 increase in same-day estimated PM2.5 were 1.013 (95% CI: 1.003, 1.023) for asthma or wheeze, 1.004 (0.978, 1.031) for bronchitis, 1.005 (0.996, 1.014) for otitis media, 0.999 (0.979, 1.019) for pneumonia, and 1.014 (1.007, 1.021) for upper respiratory infection. No consistent effect modification by level of urbanicity was observed. Conclusion: We observed that increases in PM2.5 concentrations were associated with same day increases in pediatric ED visits for asthma or wheeze and for upper respiratory infection.