Exposures and Health Effects of Ambient Particle Radioactivity

Abstract

While the effects of particulate matter (PM) on morbidity and mortality are well-established, the specific properties of PM responsible for these effects are still not completely understood. However, identifying the toxic properties of particles is critical for developing cost-effective air quality standards and ultimately protecting public health. This dissertation investigates the potential health effects of particle radioactivity, a property of airborne particles. We hypothesize that radionuclides attached to PM may have direct health effects by releasing ionizing radiation in the lungs after inhalation and deposition. In our first study, we assessed potential effect modification of radon on PM2.5-associated daily mortality in 108 U.S. cities. First, we estimated city- and season-specific PM2.5 mortality risks using time-series models, and then we regressed these effect estimates against city-level radon concentrations to estimate potential modification by radon. We found that higher mean city-level radon concentrations increased the PM2.5-mortality association in the spring and fall, suggesting that radon may enhance PM2.5 mortality. In our second study, we used time-varying measurements of gross beta radiation to directly estimate the independent health effects of particle radioactivity. Specifically, we evaluated whether short- and intermediate-exposures to particle radioactivity were associated with biomarkers of inflammation and endothelial function in an elderly male cohort. We observed associations between particle radioactivity on C-reactive protein, intercellular adhesion molecule-1 and vascular cell adhesion moleducle-1, suggesting a potential pathway for radiation-induced cardiovascular effects. Finally, we investigated the spatial and temporal variability of particle radioactivity using particle activity concentrations measured by RadNet monitors located in and around Massachusetts. We found a distinct seasonal trend at all monitors, with the highest concentrations occurring in the winter. Using a back-trajectory and clustering analysis, we found that air masses arriving from the west and southwest had the highest concentrations of beta radiation. Prevailing air mass trajectories appeared to strongly influence seasonal trends in particle radioactivity. We also investigated the role of different meteorological predictors on beta concentrations using mixed-effects models. Finally, we imputed missing beta concentrations at RadNet monitors using random forest models for use in future health studies.