Association between exposure to air pollutants and the risk of hospitalization for pulmonary embolism in Beijing, China: A case-crossover design using a distributed lag nonlinear model

Pulmonary embolism (PE) is a life-threatening condition. Few studies have evaluated the relationship between air pollution and PE, and these results have been inconsistent. Therefore, our study aimed to investigate the association between air pollutant exposure and the risk of hospitalization due to PE. Daily PE admissions, meteorological data, and ambient pollution data from January 1, 2015, to December 31, 2018, were collected in Beijing. A quasi-Poisson regression model combined with time-stratified case-crossover design and a distributed lag nonlinear model was used to determine the effect of air pollutant exposure on PE admission. To examine the stability of air pollutants' effects, multi-pollutant analyses were performed. Stratified analyses by age and sex were further conducted. There were 5060 PE admissions during the study period, with an estimated incidence of 6.5 per 100,000. PM2.5, PM10, SO2, O3 and CO exposures were significantly associated with elevated risk of PE hospitalization. The highest cumulative risks were observed at a lag of 0–28 days for PM2.5 (relative risk [RR] = 1.056, 95% confidence intervals [CI]: 1.015–1.098), PM10 (RR = 1.042, 95%CI: 1.010–1.075), and CO (RR = 1.466, 95%CI: 1.127–1.906), at a lag of 0–27 days for SO2 (RR = 1.674, 95%CI: 1.200–2.335), and at a lag of 0–4 days for O3 (RR = 1.019, 95%CI: 1.001–1.038). All associations mentioned above except O3 remained significant in multi-pollutant models. Stratified analyses showed that women and those aged ≥65 years people were more sensitive to PM10 and CO exposure than men and those aged