Wu et al. found a strong positive association between cumulative daily county-level COVID-19 mortality and long-term average PM2.5 concentrations for data up until September 2020. We replicated the results of Wu et al. and extended the analysis up until May 2022. The association between PM2.5 concentration and cumulative COVID-19 mortality fell sharply after September 2020. Using the data available from Wu et al.'s "updated_data" branch up until May 2022, we found that the effect of a 1 μg/m3 increase in PM2.5 was associated with only a +0.603% mortality difference. The 95% CI of this difference was between -0.560% and +1.78%, narrow bounds that include zero, with the upper bound far below the Wu et al. estimate. Short-term trends in the initial spread of COVID-19, not a long-term epidemiologic association, caused an early correlation between air pollution and COVID-19 mortality.
Keywords: COVID-19; PM2.5; air pollution; epidemiology; mortality; multiple regression.