Investigating links between long-term air pollution exposure and the risk of SARS-CoV-2 infection, COVID-19 hospitalisation and mortality: a systematic review and meta-analysis of cohort studies

Air pollution exposure is suggested to be associated with SARS-CoV-2 infection and COVID-19 outcomes. Available systematic reviews and meta-analyses included studies of various study designs which could be vulnerable to ecological bias. We systematically reviewed the association between particulate matter less than 2.5 aerodynamic diameter (PM2.5), nitrogen dioxide (NO2), and ozone (O3) and the risk of SARS-CoV-2 infection, COVID-19 hospitalisation, and COVID-19 mortality, focusing on cohort studies with individual-level data.

A systematic literature search was conducted on MEDLINE and Scopus in July 2023 and subsequently updated in April 2025. The risk of bias of eligible studies was assessed using a modified Risk of Bias assessment instrument developed by the World Health Organization. Qualitative synthesis was performed on all eligible studies, and random-effects meta-analyses were performed when more than three studies were available for an exposure-outcome pair, after removing studies with overlapping populations.

Long-term PM2.5 exposure was associated with an increased risk of all outcomes investigated (RR for SARS-CoV-2 infection: 1.04 [1.02–1.07], RR for COVID-19 hospitalisation: 1.11 [1.06–1.15], RR for COVID-19 mortality: 1.09 [1.03–1.15], per 1 μg/m3 increase), whereas NO2 exposure was associated with an increased risk of COVID-19 hospitalisation (RR: 1.02 [1.01–1.03], per 1 μg/m3 increase) and COVID-19 mortality (RR: 1.01 [1.01–1.02], per 1 μg/m3 increase). No associations were found for O3 exposure. Univariate meta-regression suggested that country of study accounted for a substantial proportion of the heterogeneity observed in meta-analyses.

This review presents a comprehensive, up-to-date synthesis of the evidence regarding the adverse effects of air pollutant exposure on COVID-19 outcomes based on robustly conducted cohort studies with individual-level information.