Temporal Trends of Associations Between Short-Term Exposure to Fine Particulate Matter (PM2.5) and Risk of Hospitalizations in Understudied Populations, with Effect Modification by Sex and Urbanicity in U.S. Counties

C. Chen,A. Chan,F. Dominici,R.D. Peng,B. Sabath,Q. Di,J. Schwartz,M.L. Bell

DOI: https://doi.org/10.1289/isee.2020.virtual.p-1252

2020-01-01

Abstract:Background: In a previous study, we identified evidence of change in the association between short-term exposure to fine particulate matter (PM2.5) and risk of hospitalization over time in U.S. urban counties. Whether this temporal change exists in less densely populated area or varies across individual-level characteristics is unclear. Methods: We constructed a daily time-series dataset of 968 continental U.S. counties with cause-specific hospitalization from Medicare fee-for-service data and PM2.5 estimated concentration from a hybrid model incorporating data from multiple sources for the period 2000 to 2016. Using a modified Bayesian hierarchical model, we evaluated whether disparity exists in the association between PM2.5 and risk of hospitalizations and its temporal trend across characteristics like sex and urbanicity. Urbanicity was categorized into three levels by county-specific percentage of urban population. Preliminary results: We found positive associations between circulatory and respiratory hospitalization and short-term exposure to PM2.5, with higher effect estimates towards the end of study period. The association for circulatory hospitalization decreases as urbanicity level decreases but the shape of temporal trend persists across urbanicity. The counties with low urbanicity demonstrated a 0.43%(95% posterior credible interval: 0.27 to 0.59%) increase in risk of circulatory hospitalization per 10μg/m3 increase in PM2.5 concentration for the moving average of the same day and previous two days, which is significantly lower than the 0.8%(0.54 to 1.05%) increase in counties with high urbanicity. We did not observe significant patterns in association for respiratory hospitalization or its temporal trend across urbanicity. We found no statistically significant difference in associations or their temporal trends between sexes. Conclusion: This study provides evidence of deleterious association between PM2.5 and hospitalization in understudied areas with less urban population, with different impacts by urbanicity. The association between PM2.5 and circulatory hospitalizations appears to be increasing over time with similar patterns across urbanicity.