Form Submission: Participation Entry

Background/Aim Previous studies show that fine particulate matter (PM2.5) concentrations are associated with increase in short-term risk of adverse health outcomes. However, since the U.S. established many policies to control PM2.5 in the past several decades, leading to changes in PM2.5 total mass concentration and composition, the short-term association of PM2.5 and adverse health outcomes may have changed as well. This study aims to investigate whether a temporal trend exists for this association. Methods We utilized hospital admission data of U.S. Medicare beneficiaries (>65y) and EPA PM2.5 monitoring data between 1999 and 2013, and analysed cardiovascular outcomes and respiratory outcome separately. We employed two-stage Bayesian hierarchical models to estimate the county-level and nation-wide temporal trend of association between hospital admission rates and PM2.5 concentration. To ensure internal validity of statistical model, we conducted extensive sensitivity analyses with the hypotheses tested stated a priori. We also summarized policies targeting PM2.5 between 1999 and 2013. Results We observed a 1.77% (95% CI: 0.92 to 2.63%) decrease in the percentage change in respiratory hospital admission rate with a 10 μg/m3 increase in same day PM2.5 over the study period when assuming a linear temporal trend, while no statistical significant trend was observed for cardiovascular hospital admission rates. Based on previous literatures and biological plausibility, we evaluated four possible explanations for the observed temporal trend. Conclusions Results indicate that the health impact of PM2.5 on respiratory admissions has declined over time. Based on qualitative evidence, changes in population susceptibility towards acute exposure to PM2.5 on respiratory adverse health outcomes, and combination of changes in multiple chemical components are the likely explanations for the observed temporal trend.