Presentation description

Inversions are extreme poor air quality events in which warm air traps cold air and pollutants near the ground. In Utah, multiple inversion episodes each winter impact nearly 3 million people along the Wasatch Front. Few studies have investigated the effect of inversion-driven exposures on preterm birth. Our objective is to evaluate the relationship between preterm birth and exposure to inversion events. Using Utah birth certificates from 2013-2016, we linked the maternal residential address at delivery to known dates of inversions and to a granular air quality model of daily PM2.5, NOx, O3, and temperature. We modeled the association between inversion events and preterm birth across gestational weeks using a distributed lag model (DLM), which accounts for weekly time intervals while distinguishing between correlated exposure over time. DLMs jointly fit the correlation, exposure, and structure. This allows us to identify critical windows in which exposure to inversions has a significant association with preterm birth. Our preliminary model for inversion-specific PM2.5 suggests early pregnancy may be a critical window for exposure. In upcoming work, we will interpret model curves to precisely identify critical windows and the strength of their association with preterm birth. Finally, we will map these results to visualize inversion exposure and preterm birth incidence. These findings may inform public health policies aimed at protecting pregnant individuals, particularly during inversion events. Notably, while inversions are not themselves preventable, their relationship with adverse outcomes may be modifiable through policies that reduce emissions during forecasted inversions.