Association between air quality, meteorological factors and COVID-19 infection case numbers

The coronavirus disease (COVID-19) has become a global pandemic affecting many countries, including Singapore. Previous studies have investigated the relationship of air pollutant levels and meteorological factors with respiratory disease risk and hospital admission rates. However, associations between air pollutant concentrations and meteorological factors with COVID-19 infection have been equivocal. This study aimed to assess the association between core air pollutant concentrations, meteorological variables and daily confirmed COVID-19 case numbers in Singapore. Data on air pollutant levels (particulate matter [PM(2.5), PM(10)], ozone [O(3)], carbon monoxide [CO], nitrogen dioxide [NO(2)], sulphur dioxide [SO(2)], pollutant standards index [PSI]) and meteorological factors (rainfall, humidity, temperature) was obtained from the Singapore National Environment Agency (NEA) from January 23, 2020 to April 6, 2020. The daily reported COVID-19 case numbers were retrieved from the Singapore Ministry of Health (MOH). Generalized linear models with Poisson family distribution and log-link were used to estimate the model coefficients and 95% confidence intervals (CIs) for the association between air pollutant concentrations and meteorological factors (8-day and 15-day moving averages (MA)) with COVID-19 case numbers, adjusting for humidity, rainfall and day of week. We observed significantly positive associations between NO(2), PSI, PM(2.5) and temperature with COVID-19 case numbers. Every 1-unit increase (15-day MA) in PSI, 1 μg/m(3) increase (15-day MA) in PM(2.5), NO(2) and 0.1 °C increase in temperature were significantly associated with a 35.0% (95% CI: 29.7%–40.5%), 22.6% (95% CI: 12.0%–34.3%), 34.8% (95% CI: 29.3%–40.4%) and 28.6% (95% CI: 25.0%–32.4%) increase in the average daily number of COVID-19 cases respectively. On the contrary, PM(10), O(3), SO(2), CO, rainfall and humidity were significantly associated with lower average daily numbers of confirmed COVID-19 cases. Similar associations were observed for the 8-day MAs. Future studies could explore the long-term consequences of the air pollutants on COVID-19 infection and recovery.