Evaluating Influential Factors for Spatial Variability of the Effect of Air Pollution on Term Birth Weight Using Bayesian Hierarchical Models

摘要

Studies suggest that air pollution has adverse effects on pregnancy outcomes. Such effects can be modified by socio-demographic and environmental factors. Few studies have investigated the impact of these factors on spatial variability of the effects of air pollution. We developed a two-stage Bayesian hierarchical model to evaluate spatial variability of the effects of air pollution on the weight of term birth infants (≥37 gestational weeks) across census tracts and the influential factors for such spatial variability. Based on the birth certificate records from 2001 to 2008 in Los Angeles County, California, USA, we developed a two-stage hierarchical non-linear model to evaluate the spatial variability of the effects of air pollution on term birth weight across Census tract and the influence of socio-demographic (including ethnicity and socioeconomic variables) and environmental (including land-use and greenness variables) factors on the variability of the effects. Air pollution exposure was modeled at individual level for nitrogen dioxide (NO2) and nitrogen oxides (Nox) using spatiotemporal models. We found adverse effects of air pollutants on term birth weights (-1.47 g per ppb increment in NO2 and -0.69 g per ppb increment in Nox). The effects of NO2 and Nox were spatially clustered. Spatial variability of such effects was affected by socio-demographic (ethnicity, median family income, maternal education, and commuting means and time to work), and environmental (distance to freeways/highways, proportion of land-use for agriculture, heavy industry and park/recreation, and greenness) factors. This study contributes new findings on influence of socio-demographic and environmental factors on the spatial variability of the effects of air pollution on term birth weight.