Hazardous Air Pollutants and Maternal Health

There is little known about the health effects from ambient levels of hazardous air pollutants (HAPs), particularly in regard to maternal health. We help fill this gap by investigating whether pregnant women from diverse sociodemographic backgrounds face systematic differences in HAPs exposure, assess the impact of ambient HAP levels on gestational hypertension (GH) and gestational diabetes mellitus (GDM), and determine whether historically underserved groups experience disproportionately more severe health effects at comparable exposure levels. We compile data on the population of pregnant women and singleton births in North Carolina from 2002-2016, and combine these data with a comprehensive, spatially explicit measure of ambient HAP concentrations. We estimate multivariate regression models that employ high-resolution fixed effects and an instrumental variables approach based on reported fugitive emissions to identify the plausibly causal effects of ambient HAP levels on maternal health. The results demonstrate that pregnant women from historically underserved subsets of the population face higher baseline risks of GH and GDM, and are exposed to greater levels of HAPs during pregnancy. We find evidence that a one percent increase in HAPs leads to a roughly 0.03% increase in the risks of GH and GDM, and examine heterogeneity in these health effects based on race, ethnicity, and proxies for income. Considering the average 31% reduction in ambient HAP concentrations during our study period, these results suggest a reduction of 61 to 68 statistical cases of GH in North Carolina each year, and a comparable decrease of 57 to 59 statistical cases of GDM. Our quantified estimates of the maternal health effects can be used to inform future policy decisions and demonstrate how HAPs exacerbate maternal health disparities. Key Words: air pollution; environmental justice; hazardous air pollution; HAP; maternal health; gestational hypertension; gestational diabetes mellitus