You are here:
County-level environmental quality is differentially associated with individual- and county-level infant mortality by race
Citation:
Jagai, J., L. Messer, K. Rappazzo, AND D. Lobdell. County-level environmental quality is differentially associated with individual- and county-level infant mortality by race. Presented at Society for Pediatric and Perinatal Epidemiologic Research, Boston, MA, June 17 - 18, 2013.
Impact/Purpose:
To assess associations between county-level environmental quality and individual- and county-level infant mortality and how the associtions vary by race.
Description:
Human health is affected by simultaneous exposure to stressors and amenities, but research typically considers single exposures. In order to account for multiple ambient environmental conditions, we constructed an Environmental Quality Index (EQI) using principle components analysis with data representing five environmental domains (air, water, land, built and sociodemographic). Using U.S. linked births/infant deaths data for 2002 (4,027,479 birth records; 27,527 infant deaths), we report on associations between domain-specific and the overall indices for infant mortality (IM). Race-stratified linear regression estimated prevalence differences (PD(95% confidence interval(CI))) in county-level rates of IM (n=3141) for quartiles of index exposure. Fixed slope, random intercept multilevel logistic models estimated odds ratios (OR(95%CI)) for index quartiles and individual IM. Results differed by racial group and environmental domain. Race stratified analyses demonstrated residence in a county with poor environmental quality (4th quartile) compared to the best quality (1st quartile) was associated with IM in county-level analyses (PD=0.003(0.004,0.0016)) but not in individual-level analysis (OR=0.96(0.86,1.05)) for blacks however, this association was not seen in whites. The air index demonstrated a strong association in blacks in county-level analyses; increasing quartiles of the air index were associated with PDs for IM (e.g., 4th quartile PD=0.014(0.011,0.017) compared with 1st quartile). The land, water, built and sociodemographic indices had mixed results with IM for both county- and individual-level analyses. This work considered associations both with cases and with rates of IM and these environments appear linked to both individual and county-level outcomes. (This abstract does not necessarily reflect EPA policy.)