Ultrafine particles, PM2.5 chemical constituents, and cardiovascular disease - PROJECT SUMMARY The goal of this F31 fellowship application is to support and promote my (Erin Burman’s) training as a pre- doctoral student in the Department of Environmental Health of the Harvard T.H. Chan School of Public Health and the Graduate School of Arts and Sciences at Harvard University. My long-term research interests focus on the association of air pollution on cardiovascular (CV) conditions and mortality. I have assembled a group of leading experts as my Sponsors and Co-Sponsors including Drs. Francine Laden, Andrea Bellavia, Eric Rimm, and Joel Schwartz. Air pollution is the second leading risk factor for death worldwide, with nearly half of these deaths due to cardiovascular disease (CVD). Airborne particulate matter (PM) is well-understood to be particularly harmful, with much research and regulation focusing on PM with diameter < 2.5 microns (PM2.5). However, PM2.5 is a complex mixture composed of various size fractions and a variety of chemicals, and it is unclear which of these constituents drive PM2.5’s toxicity. Identifying particularly harmful PM constituents has been challenging because constituents have complex correlation structures that are difficult to model with conventional statistical methods. Confounding bias has also made identifying causal effects challenging. In this study, we will investigate the effects of long-term exposure to ultrafine particles (UFPs) with diameter < 100 nm as well as fifteen chemical components of PM2.5 – including elemental carbon (EC), ammonium (NH4), nitrate (NO3), organic carbon (OC), sulfate (SO4), and elements Br, Ca, Cu, Fe, K, Ni, Pb, Si, V, and Zn -- on CV conditions and mortality. We leverage new spatial and spatiotemporal models of these pollutants, linking them to the Nurses’ Health Study II (NHS II), a nationwide longitudinal cohort. We will model UFPs and PM2.5 chemical components as a mixture, using methods such as weighted quantile sums (WQS), quantile g- computation, and other machine learning methodologies to account for correlations and identify especially harmful constituents of the mixture as well as its joint effects on incident hypertension (Aim 1) and incident CVD (encompassing coronary heart disease and stroke) as well as CV mortality and all-cause mortality (Aim 2). Finally, we will leverage a quasi-experimental design to investigate changes in the exposure mixture among the subset of nurses who moved residential addresses during follow-up, and those changes’ association with hypertension, CVD, and CV and all-cause mortality (Aim 3). As concentrations and sources of PM change in the U.S. and abroad, understanding which constituents of air pollution are most harmful will allow for informed response and targeted reductions of specific pollutants. In this research training plan, I will receive extensive training in multiple aspects of exploring the impacts of PM constituents on CV health and mortality. This will position me as a future leader in the field of air pollution epidemiology through high-quality mentorship and extensive professional development opportunities.
