≤ 2.5 μ m (PM 2.5 ) and ≤ 10 μ m (PM 10 ), ambient temperature (AT), absolute humidity (AH) and migration scale index (MSI) in 72 cities of China (excluded Wuhan city) on a daily basis, each of whichconfirmed more than 50 cases from January 20th to March 2nd, 2020. We applied a two-stage analysis. Generalized additive models with quasi-Poisson distribution was first fitted to estimate city-specific effects of PM 10 and PM 2.5 on daily confirmed COVID-19 cases while controlling AT, AH and MSI. Then, we used meta-analysis to generate the pooled effect estimates from city-specific results.Results: During the study period, there were a total of 24 939 COVID-19 cases, most of which were reported in Hubei Province. In our meta-analysis, we found each 10 μ g/m 3 increase in concentration of PM 2.5 and PM 10 in single day lag (from lag 0 to lag 7 and lag 14) were positively associated with confirmed cases of COVID-19, not including PM 10 at lag 5, lag 6 and lag 7, and PM 2.5 at lag 5, lag 6. Similar trend was also found in different cumulative lag days (from lag 01 to lag 07 and lag 014). The effects of PM 2.5 and PM 10 on daily COVID-19 confirmed cases are statistically significant for three cumulative lag periods over 3, 7 and 14 days with the greatest effect over 14 days. The estimated RRs of which were 1.64 (95% CIs: 1.47, 1.82) and 1.47 (95% CIs: 1.34, 1.61) with each 10 μ g/m 3 increase in concentrations of PM 2.5 and PM 10 , respectively. In addition, we found that the effects of PM 2.5 on daily confirmed cases were greater than PM 10 in all included lag days.
Conclusions:This nationwide study suggests that airborne PM pollution likely increases the risk of getting COVID-19 in China.