People with chronic obstructive pulmonary disease, cardiovascular disease, or hypertension have a high risk of developing severe coronavirus disease 2019 (COVID-19) and of COVID-19 mortality . However, the association between long-term exposure to air pollutants, which increases cardiopulmonary damage, and vulnerability to COVID-19 has not yet been fully established . We collected data of confirmed COVID-19 cases during the first wave of the epidemic in mainland China . We fitted a generalized linear model using city-level COVID-19 cases and severe cases as the outcome, and long-term average air pollutant levels as the exposure . Our analysis was adjusted using several variables, including a mobile phone dataset, covering human movement from Wuhan before the travel ban and movements within each city during the period of the emergency response . Other variables included smoking prevalence, climate data, socioeconomic data, education level, and number of hospital beds for 324 cities in China . After adjusting for human mobility and socioeconomic factors, we found an increase of 37.8% (95% confidence interval [CI]: 23.8% –52.0 %), 32.3% (95% CI : 22.5% –42.4 %), and 14.2% (7.9% –20.5 %) in the number of COVID-19 cases for every 10-μg/m3 increase in long-term exposure to NO2, PM2.5, and PM10, respectively . However, when stratifying the data according to population size, the association became non-significant . The present results are derived from a large, newly compiled and geocoded repository of population and epidemiological data relevant to COVID-19 . The findings suggested that air pollution may be related to population vulnerability to COVID-19 infection, although the extent to which this relationship is confounded by city population density needs further exploration.