You are here:
Interaction between heterogeneous environmental quality domains (air, water, land, socio-demographic and built environment) on preterm birth.
Grabich, S., C. Gray, J. Jagai, Y. Jian, L. Messer, K. Rappazzo, AND D. Lobdell. Interaction between heterogeneous environmental quality domains (air, water, land, socio-demographic and built environment) on preterm birth. Society for Pediatric and Perinatal Epidemiology Annual Meeting, Miami, FL, June 20 - 21, 2016.
To explore interactions between each environmental domain pair of the EQI and PTB using National Center for Health Statistics data.
Environmental exposures are often measured individually, though many occur in tandem. To address aggregate exposures, a county-level Environmental Quality Index (EQI) representing five environmental domains (air, water, land, built and sociodemographic) was constructed. Recent studies showed associations between EQI and health effects, including preterm birth (PTB), using individual domains as well as overall EQI; however, the interaction between domains has yet to be investigated. We explored interactions between each domain pair and PTB using National Center for Health Statistics data (2000-2005; n=19,171,351). We stratified by rural urban continuum codes (RUCC) and used fixed slope, random intercept multilevel logistic models, adjusted for maternal age, education, marital status, and race, to assess how interactions between individual domains relate to PTB (<37 weeks gestation). We calculated prevalence odds ratios (POR (95% confidence intervals (95%CI))) comparing residence in a county with poor environmental domain interactions (5th quintiles) to the referent of those who resided in the best quintile for each domain (1st quintiles). Trend across quintiles and interaction p-value <0.05 determined significance. Significant interactions were found between the air and land domain across RUCC models. For example, in the most rural counties the POR comparing the highest (worse quality) to lowest EQI quintile for air was 1.54 (1.28-1.86); land 1.13 (0.91-1.40) and interaction (0.74 (0.57-0.95). In the most urban environments, interactions were consistent across many of the environmental domains including socioeconomic with built, water and air. Comparing highest to lowest EQI quintile for built was 0.74 (0.63-0.88); land 0.68 (0.56-0.84); interaction 1.38 (1.06-1.80). Our results suggest interactions exist between different environmental domains with PTB, which should be considered in future statistical analyses. This abstract does not necessarily reflect EPA policy.