Impact of Extreme Heat, Wildfire Smoke, and Community-Level Social Vulnerability on Maternal Health - PROJECT SUMMARY/ABSTRACT Hypertensive disorders of pregnancy (HDP) and gestational diabetes mellitus (GDM) pose significant challenges to maternal health, affecting a substantial portion of pregnancies in the United States. These conditions disproportionately impact marginalized communities, highlighting the need for comprehensive research to address this critical public health issue. The primary objective of this research is to gather actionable information to enhance reproductive health and human development. This will be achieved by quantifying the individual and combined effects of extreme heat and wildfire smoke exposure on the incidence of HDP and GDM. The study aims to evaluate the impact of preconception and prenatal exposures to extreme heat and wildfire smoke, determine how combined exposures vary across the United States (U.S.), and explore potential modifications of these relationships by measures of social vulnerability. The study leverages a healthcare claims dataset, encompassing an estimated 6.5 million births during the study period and representing the largest study to date aimed at understanding the impacts of extreme heat and wildfire smoke exposure on maternal health outcomes, as well as the effects of combined exposures in the context of social vulnerability. A novel wildfire smoke dataset will be employed, including concentrations of total mass wildfire smoke-related PM2.5 (particulate matter ≤ 2.5 microns), individual PM2.5 constituents (sulfate, nitrate, elemental carbon, and organic carbon), and smoke-specific ozone. The analytical approach will involve time-to-event epidemiological models, spatial clustering, and machine learning kernel methods. Cox regression modeling within a distributed lag nonlinear framework will be used to estimate the risk of HDP diagnosis. K-means spatial clustering will be employed to quantify spatial heterogeneity in risk across all counties in the contiguous U.S. (CONUS), while Bayesian Kernel Machine Regression Distributed Lag Models (BKMRDLM) will be utilized to quantify overall effects and assess lagged responses. The study will evaluate exposure during the preconception period (0-4 weeks from last menstrual cycle), each week of gestation, and by trimester. The results of this study could provide crucial information to inform personal and clinical efforts to protect maternal health. By examining the complex interplay between environmental factors, social vulnerability, and HDP/GDM, this research has the potential to enhance our understanding of the risk factors associated with HDP and GDM, inform targeted interventions to reduce incidence, particularly in vulnerable populations, and guide policy decisions related to environmental regulations and public health measures. Furthermore, this study may serve as a foundation for additional research in maternal, fetal, and infant health. This methodological approach could be applied to investigate how mixtures of environmental, biological, and social stressors are associated with various maternal health outcomes beyond HDP and GDM. By addressing the complex interactions between these stressors, this research aims to contribute significantly to the field of reproductive health and ultimately improve outcomes for pregnant women and their offspring.
