LAUSR

Background and Objectives To characterize the association of ambient particulate matter (PM) pollution of different sizes (PM ≤1 µm in aerodynamic diameter [PM1], PM2.5, and PM10) with in-hospital case fatality among patients with stroke in China. Methods We collected hospitalizations due to stroke in 4 provinces in China from 2013 to 2019. Seven-day and annual averages of PM prior to hospitalization were estimated using bilinear interpolation and residential addresses. Associations with in-hospital case fatality were estimated using random-effects logistic regression models. Potential reducible fraction and the number of fatalities attributed to PM were estimated using a counterfactual approach. Results Among 3,109,634 stroke hospitalizations (mean age 67.23 years [SD 12.22]; 1,765,644 [56.78%] male), we identified 32,140 in-hospital stroke fatalities (case fatality rate 1.03%). Each 10 µg/m3 increase in 7-day average (short-term) exposure to PM was associated with increased in-hospital case fatality: odds ratios (ORs) were 1.058 (95% CI 1.047–1.068) for PM1, 1.037 (95% CI 1.031–1.043) for PM2.5, and 1.025 (95% CI 1.021–1.029) for PM10. Similar but larger ORs were observed for annual averages (long-term): 1.240 (95% CI 1.217–1.265) for PM1, 1.105 (95% CI 1.094–1.116) for PM2.5, and 1.090 (95% CI 1.082–1.099) for PM10. In counterfactual analyses, PM10 was associated with the largest potential reducible fraction in in-hospital case fatality (10% [95% CI 8.3–11.7] for short-term exposure and 21.1% [19.1%–23%] for long-term exposure), followed by PM1 and PM2.5. Discussion PM pollution is a risk factor for in-hospital stroke-related deaths. Strategies that target reducing PM pollution may improve the health outcomes of patients with stroke.