Exploring airshed and watershed exposure pathways as mediators of the association between proximity to surface mining and adverse birth outcomes in Central Appalachia

Background Previous work has determined an association between proximity to active surface mining in coal producing counties within Central Appalachia and an increased risk of preterm birth (PTB) and low birthweight (LBW); however, the relative importance of specific exposure pathways explaining this association remains poorly understood. Multiple potential exposure pathways to surface mining activities exist during gestation, including inhalation of particulate matter (airshed exposure), or via exposure to impacted surface waters (watershed exposure). Here, we perform a mediation analysis to explore these pathways and the odds of adverse birth outcomes. Methods We obtained birth records acquired through health departments in WV, KY, VA and TN between 1990 and 2015. Surface mine extents for each year were identified through remotely sensed Landsat imagery. Corresponding airsheds were estimated using the HYSPLIT4 atmospheric trajectory model, while watersheds were assigned using United States Geological Survey’s Watershed Boundary Dataset boundaries. We performed logistic regression to determine associations between exposure and the odds of preterm birth, low birthweight, and term low birthweight, and iteratively included within our models: a) the percent of active surface mining landcover within a 5 km buffer of maternal residence; b) the cumulative potential exposure to air pollutants via the airshed experienced at the maternal residence, and; c) the percent of land experiencing surface mining within the watershed of residency. Results Our baseline models found that active surface mining was associated with an increased odds of PTB (1.09, 1.05 – 1.13) and LBW (1.06, 1.02 – 1.11), while controlling for significant predictors. When mediators were added to the base model, the association between active mining and birth outcomes was reduced (PTB: 1.04, 0.99 – 1.09; LBW 1.04, 0.99 – 1.10), while the odds of PTB and LBW increased with airshed exposure (PTB: 1.14, 1.11 – 1.18; LBW: 1.06, 1.03 – 1.10). Conclusions Results were consistent with a hypothesis of mediation via airshed, but suggested mediation via watershed was less likely. These results suggest that air pollution resulting from surface mining activities may be the primary exposure pathway explaining the association between maternal residence proximity to active surface mining and PTB and LBW.