Association between Exposure to PM 2·5 and the Date-Based Secondary Attack Rate of COVID-19 in Guangzhou, China: A Retrospective Cohort Study

Background: There is no information on the association between exposure to fine particulate matter with an aerodynamic diameter of 2·5 μm or less (PM2·5) and the date-based secondary attack rate (DBSAR) of COVID-19. Methods: In this retrospective cohort study performed in Guangzhou, China, based on hourly concentration data of PM>sub>2·5, PM10, SO2, NO2, and CO from January 1, 2020 to February 15, 2020 from approximately 1,500 monitoring stations in China and the addresses of the participants, we used the Kriging interpolation method to calculate the moving average (MA) concentrations of the pollutants over the previous 1 to 14 days before symptom onset of secondary cases. After simultaneously adjusting for a series of key covariates, we evaluated the dose-response relationship between PM2·5 and DBSAR using a generalized linear mixed model. Findings: DBSAR fluctuated from 2·12% on January 17 to 6·61% on February 13. The significantly positive relationship between PM2·5 exposure and DBSAR at prior 1 to 14 MA days. For example, the dose-response relationship was as follows, for every 10 μg/m 3 increase in PM2·5, DBSAR increased by 2·77% (95% CI: 1·32% to 4·22%; P < 0·001) at prior 1 MA day and 5·82% (95% CI: 4·00% to 7·65%; P < 0·0001) at prior 14 MA days. The effect of PM2·5 exposure in DBSAR significantly differed by age and exposure time. Interpretation: There was a strong nonlinear dose-response relationship between exposure to PM2·5 and DBSAR at prior 1 to 14 MA days, which provide significant evidence and suggests that transmission by aerosol is possible. Funding Statement: This study was supported by grants from the Health Science and Technology Plan Project of Guangzhou (20191A01158). Declaration of Interests: All authors declare no competing interests. Ethics Approval Statement: This study was approved by the ethics committee of the Guangzhou CDC (Guangzhou, China). Analyses involving personally identifiable data were conducted at the Guangzhou CDC. The analyses were performed with anonymized data.