Abstract Body (Do not enter title and authors here): Past epidemiological evidence for the causal link between cardiovascular disease (CVD) and fine particulate matter (PM_2.5) particles have focused on overall mass concentration, rather than source and composition-specific effect estimates. Interventions targeting point-source industrial emission were thus usually faced with challenge of accurately assessing the benefits from source-specific PM reductions. This study took advantage of the January 2016 closure of the Shenango coke plant in Pennsylvania (Pa), US, to assess the association between source-specific air pollution and the cardiovascular health of residents. We hypothesized that exposure to PM_2.5 from different sources exerted different CVD outcomes.
PM_2.5 trace element monitoring data in 2013-2018 (i.e., three years before vs. after the shutdown) were acquired from the US EPA Chemical Speciation Network monitors. Source apportionment via Factor Analysis and Absolute Principal Component Analysis was performed to identify and estimate the source-specific PM_2.5 levels. Emergency department (ED) visits were acquired from the Pa Department of Health. A quasi-Poisson generalized additive model adjusting for seasonality was used to assess the associations between each source component and ED visit rate.
We found a decrease in coal-related PM_2.5 after the plant closure, but not for other source categories identified (Traffic, Biomass, Crustal, Oil, and Salt). This decrease was concurrent with a decrease in CVD ED visits in the community residing near the coke plant. The reduction in coal-related PM_2.5 was associated with a statistically significant decrease of CVD ED visit risk (IRR = 1.14 per IQR change). Exposure to traffic-related PM_2.5 was associated with increased CVD ED risk (IRR = 1.21 per IQR increase). Coal and traffic emissions are major source contributors to fossil-fuel-related PM particles. We did not observe statistically significant associations between CVD risk and exposures to crustal, oil, or salt PM_2.5. Biomass-related PM_2.5 increased over the period of study and was negatively associated with CVD ED visits.
This study highlights the importance of understanding the distinct CVD effects of different PM sources and components, especially fossil-fuel-related PM exposures. We recommend considering the source-type specific PM health effects in future policymaking and clinical practice to improve the estimates of health benefits and efficacy of health and emissions interventions.