Exposure to air pollution is a major public health threat worldwide, with disease burden now comparable to the health risks associated with tobacco smoking and an unhealthy diet.1 Fine particles and gaseous pollutants that infiltrate into the small airways and absorbed into the bloodstream are important risk factors for cardiovascular disease, hypertension, diabetes, lung cancer, respiratory infections especially in children2 and probably asthma and chronic obstructive pulmonary disease (COPD).

While levels of ambient air pollution have declined in developed countries,3 they continue to escalate in countries undergoing rapid urbanization and economic development. Given emerging evidence on the health impacts of air pollution at a much lower thresholds than previously acknowledged, the World Health Organization (WHO) recently revised air quality guidelines to reduce the adverse threshold for the mean annual level of particulate matter of aerodynamic size 2.5 μm (PM2.5) from 10 to 5 μm/m3.1 It is now estimated that 99% of the world's population live in areas that have pollution levels exceeding those recommended by these guidelines.4

There is growing interest in the potential for air pollution to increase the risk for severe outcomes in coronavirus disease 2019 (COVID-19) patients including hospitalization and mortality. The COVID-19 pandemic arising from the widespread dissemination of the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) has resulted in unprecedented increases in acute hospitalizations that, at some point, have overwhelmed healthcare systems in many countries. While some admissions were required to quarantine infected individuals, hospital admissions are clinically indicated for SARS-CoV-2-infected individuals who develop interstitial pneumonia, some of whom develop respiratory failure requiring ventilatory support. Poor prognostic factors include advanced age, co-existent co-morbidities, such as hypertension and diabetes, and reduced baseline oxygenation in COPD and other lung diseases. Understanding the role of these prognostic factors could help identifying disease-modifying strategies.

To date, the findings of studies that investigated the associations between air pollution and COVID-19-related hospitalizations have been conflicting, which might be partly explained by almost all being ecological studies. As the unit of observation is a group of people and not an individual person, by design, ecological studies cannot adequately adjust for confounding variables and there is potential for ecological fallacy which precludes accurate inferences at an individual level. The lack of studies at an individual level was highlighted by the 2021 joint workshop report of the European Respiratory Society (ERS), International Society of Environmental Epidemiology (ISEE), the Health Effects Institute (HEI) and WHO that called for the conduct of carefully designed studies at the individual level, ideally across several countries.5

Responding to this call in a recent publication in Respirology, Mendy et al. have undertaken such an individual-level study.3 The authors used data from the University of Cincinnati Health System which documented COVID-19-related hospitalization and approximately 13.6% of their 14,783 study participants were admitted for ≥24 h between March and September 2020. As exact residential addresses were not known, zonal PM2.5 between 2009 and 2018 was estimated on a 0.01° × 0.01° grid using a validated chemical transport model merged with satellite and ground-based observations over North America,6 that is, prior to the anomalous reductions in air pollution from widespread regional ‘lockdowns’. Geometric mean (SD) of the PM2.5 level in the study was 10.48 (1.12) μg/m3 and the odds for COVID-19-related hospitalization increased by 14%–18% per each 1 μg/m3 increase in PM2.5. Comprehensively adjusting for social disadvantage was a study strength. Interestingly, the authors also adjusted for multiple co-morbidities, which are more likely to be effect modifiers, and interaction analyses could have added further insights into subgroups who were especially at risk. The work complements another US study that similarly investigated this research question but used data from the US Department of Veteran Affairs national health database (N = 169,102) and satellite-based PM2.5 levels from 2018 only.7 This study found that the interquartile increase in the mean annual PM2.5 level between 6.6 and 8.5 μg/m3 (1.9 μg/m3, median 7.4) was associated with a 10% (95% CI: 8%–12%) increase in the relative risk of hospitalization, especially among those socially disadvantaged and of African-American ethnicity. While the exposure assessments and the analyses of these two studies are not directly comparable, the findings are consistent and included a lack of a threshold effect.

Several mechanisms for these observations are possible. Some epidemiological data have supported the possibility that air pollution-related lung injury could manifest as a mixed obstructive–restrictive lung function pattern, especially from higher nitrogen dioxide exposure.8, 9 Thus, air pollution could predispose susceptible individuals to develop lung injury and other cardiometabolic diseases, which could in turn increase the risk of developing more severe infection by SARS-CoV-2. In vitro, animal and human studies suggest that pollution-related impairment of host defences such as increased mucosal permeability, increased oxidative stress and less effective phagocytosis by macrophages could increase one's susceptibility to severe SARS-CoV-2 infection.10 Similar to other respiratory viruses, air pollution can increase protease-induced cleavage of the viral spike protein and enhance SARS-CoV-2 binding to human angiotensin-converting enzyme-2.10, 11 However, ambient particles are unlikely carriers of the aerosolized virus for prolonged periods as they are laden with water, mucus and salts.5, 12

Thus, to further our understanding of the intersection between these two major environmental threats to human health, similar research on the adverse impact of ambient air pollution on COVID-19 hospitalization is needed from countries other than the United States. Nonetheless, the present work highlights the imperative to constantly review global societal priorities5 of transitioning towards reduced energy consumption and adoption of cleaner energy sources.

Jennifer Perret and Shyamali Dharmage hold an investigator-initiated grant supported by GlaxoSmithKline, Australia, for unrelated research.