Is there an association between exposure to air pollution and severity of COVID-19 infection?

April 29, 2020

Uy Hoang, Nicholas R. Jones

 On behalf of the Oxford COVID-19 Evidence Service Team
Centre for Evidence-Based Medicine, Nuffield Department of Primary Care Health Sciences
University of Oxford

Correspondence to uy.hoang@phc.ox.ac.uk

VERDICT

• There is very limited data to date, and we found only one study which had adjusted for confounders, but emerging evidence suggests there may be a positive association between long-term exposure to ambient air pollution and COVID-19 mortality.
• These data might be of particular importance as international lockdown measures are eased, given the restrictions have caused a considerable fall in levels of air pollution.
• Public health measures to reduce air pollution can also prevent many avoidable deaths from non COVID-19 causes and so help lower the baseline demand on health services during the current pandemic.

BACKGROUND
Over the past few decades, considerable research has been undertaken into the adverse health effects of ambient (outdoor) air pollution and potential mitigation measures.¹ However ambient air pollution remains a major environmental health hazard, responsible for an estimated 8.8 million premature deaths per year (95%CI 7.11 to 10.41) and loss of life expectancy of 2.9 years (95%CI 2.3 to 3.5 years).² These figures now exceed that of tobacco smoking.

Air pollution, and particularly high nitrogen dioxide (NO₂) concentration, may be of particular importance in the context of respiratory tract infections. Both household and ambient air pollution have been linked to an increased risk of hospitalisation and death due to pneumonia.³ An observational analysis during the 2003 Severe Acute Respiratory Syndrome (SARS) epidemic in China found mortality rates were higher in urban regions with high-levels of ambient air pollution compared to low pollution areas, as measured by the Air Pollution Index (API), although these results were not adjusted for important confounders, such as age, sex and co-morbidities.⁴

A number of risk factors have been identified for developing the COVID-19 infection associated with the SARS-CoV-2 virus, including population density.⁵ This may be because the virus is transmitted more rapidly between people living in the close proximity of urban compared to rural areas. However, there may also be an effect from ambient air pollution, given this too is strongly associated with population density.⁶

In this review we aim to determine if there is an association between ambient air pollution and severity of COVID-19 infection. This information is important to inform potential public health measures to improve inhaled air quality.

CURRENT EVIDENCE
We searched MEDLINE from 1946 to present, Embase from 2000 to 2020, and the Cochrane Library of systematic reviews up to 20^th April 2020 for studies looking at the association between coronavirus infection and ambient air pollution. The following key words were used – air pollution, coronavirus, coronavirinae, coronavirus infections.

We screened 46 research articles at title and abstract, of which seven were relevant to infection with COVID-19. Of these seven, we included the three studies that reported the relationship between ambient atmospheric pollution and severity of COVID-19.^7-9 Two were observational analyses and one a narrative review. The other four studies we excluded as not being relevant to exposure to air pollution (one described public health measures to prevent COVID-19,¹⁰ one described the airborne transmission of COVID-19¹¹ and a further two articles described the effects of public health measures on ambient air pollution levels^{12 13}).

All three studies included concluded that there was a likely association between regions with high ambient air pollution and case fatality from COVID-19. Wu et al compared historical levels of ambient air pollution with COVID-19 death rates (n=45,817) across 3,000 counties in the United States of America.⁹ Air pollution was calculated in terms of fine particulate matter (PM_2.5), estimated using a combination of satellite imaging, monitored PM_2.5 data and modelling. Results were adjusted for 20 potential confounders including age, ethnicity, household income, population density, smoking, weather and local health resources. In the adjusted analysis, an increase of 1𝜇g/m3 in long-term exposure to PM_2.5 was associated with an 8% increased risk of COVID-19 mortality (Estimated mortality rate ratio 1.08, 95%CI 1.02 to 1.15). It is important to note this is a pre-print publication that has not been peer-reviewed.

The study by Ogen combined data on NO₂ concentration from the Sentinel-5 Precursor space-borne satellite as part of the European ‘Copernicus’ programme with atmospheric conditions (provided by the NOAA/OAR/ESRL PSD, Boulder, Colorado, USA (https://psl.noaa.gov/)) and COVID-19 fatality rates across 66 regions in France, Germany, Italy and Spain.⁸ Across these 66 regions there were 4,443 reported fatalities due to COVID-19 as of March 19, 2020. Of these fatalities, 83% (n=3,701) occurred in the regions with the highest NO₂ concentration and just 1.5% (n=51) in the regions with the lowest NO₂ concentration.⁸ Mortality rates are not reported with reference to population size.

The review article reported data showing case fatality for COVID-19 was 12% in the dense urban areas of Lombardy and Emilia Romagna compared to the Italian national average of 4.5%.⁷ The authors postulate this may be because long-term exposure to ambient air pollution promotes a chronic inflammatory state and increased risk of chronic respiratory disease, both of which have been linked to an increased risk of death in COVID-19. However, the review did not include any studies that had analysed these associations directly.

Quality assessment

We used the STROBE checklist to assess quality of reporting for the observational studies¹⁴ with additional questions for ecological studies, as suggested by Dufault and Klar.¹⁵ The study by Wu et al was well conducted, with a range of important confounders considered and adjusted for. Sensitivity analysis was conducted to check the influence of potential outliers, such as New York metropolitan area, where the death rate was considerably higher than the national average.⁹ Nonetheless, unmeasured confounding may have influenced these results and the analyses, which were done at county level, are not directly applicable to individuals.⁹ We found that the quality of evidence from the other two studies was weak and neither reported any quantifiable correlation or relative risks.^{7 8} The findings of these two studies are limited by the cross-sectional, ecological study designs and the lack of data to account for sources of potential bias or imprecision. Death rates were not adjusted for population density, nor for relevant population characteristics, such as age, sex or deprivation profiles, which are known to be linked to increased risk of COVID-19.

Air pollution and other coronavirus infections

Evidence of the effect of ambient air pollution from the previous coronavirus epidemic of SARS in 2003 is also limited.⁴ One cross-sectional, ecological study assessed the risk of death from SARS in relation to ambient air pollution, as measured by the API across different areas of China. The study reported a positive association between API and risk of mortality, with the relative risk of death for people living in the most polluted regions, compared to least polluted of 2.18 (95% CI: 1.31–3.65). However, the results were not adjusted for important confounders, such as age, sex and co-morbidities.

Currently the only published cluster randomised controlled trial into the effects of air pollution and coronavirus studied the effect of indoor household pollution on acute lower respiratory tract infections caused by coronavirus and other pathogens (both bacterial and viral).¹⁶ The study, conducted in Ghana, showed no link between levels of household air pollution and viral nasal carriage.¹⁶ Evidence for the effectiveness of interventions to limit indoor household air pollution on health is generally limited.^17-19

Implications for practice and public health policy

The global burden of disease study suggests that interventions to reduce household air pollution, ambient particulate matter pollution, and second hand smoking are important to reduce mortality from lower respiratory tract infection in both adults and children.²⁰ These interventions to improve air quality may also have an effect in lowering the baseline of demand on health services during the COVID pandemic.²¹

We have found emerging evidence to suggest increased levels of ambient air pollution or NO₂ concentration may be linked to adverse outcomes in COVID-19. However, the nature and strength of any direct association remains uncertain. The current lockdown measures have led to dramatic reductions in levels of urban air pollution, with the most significant reductions in black carbon and NO₂ (45 to -51% decrease).¹² Further research is needed into changing levels of air pollution and the potential link with COVID-19, which can inform public health policy in the context of easing lockdown measures.

CONCLUSIONS

• Emerging evidence suggests there may be a positive association between long-term exposure to ambient air pollution and severity of COVID-19 infection.
• Overall the current evidence is limited, and more research is required to categorise the importance of exposure timing and pollution level, especially from studies with robust designs and adjusted for important confounders.
• However, the global burden of disease study suggests that interventions to reduce household air pollution, ambient particulate matter pollution, and passive smoking are still important public health measures and could prevent many avoidable deaths. Whilst it is not certain whether these interventions to improve air quality could reduce coronavirus disease severity, they will have an important impact in lowering the baseline demand on health services during the COVID pandemic.

Disclaimer:  the article has not been peer-reviewed; it should not replace individual clinical judgement and the sources cited should be checked. The views expressed in this commentary represent the views of the authors and not necessarily those of the host institution, the NHS, the NIHR, or the Department of Health and Social Care. The views are not a substitute for professional medical advice.

AUTHORS
Dr Uy Hoang is a research fellow working within the Royal College of General Practitioners Research and Surveillance Centre in the Nuffield Department of Primary Care Health Sciences. His role within the research group is to provide epidemiological support and a public health overview to the work on chronic and infectious diseases. Alongside specialist medical training in public health at the Oxford Deanery, he also has a doctorate in clinical epidemiology from the Institute of Psychiatry at King’s College and a Masters in Public Health from Yale University.
Dr Nicholas Jones is a GP and Wellcome Trust Doctoral Research Fellow at the University of Oxford.

SEARCH TERMS
Population – General population, Intervention – air pollution, Comparison – None, Outcome – COVID-19.

We searched MEDLINE from 1946 to present, Embase from 2000 to 2020, and the Cochrane Library of systematic reviews  date of search 20^th April 2020 for studies looking at the association between coronavirus infection and ambient air pollution. The following key words were used – air pollution, coronavirus, coronavirinae, coronavirus infections.

REFERENCES

1. Sweileh W, Al-Jabi S, Zyoud S, et al. Outdoor air pollution and respiratory health: a bibliometric analysis of publications in peer-reviewed journals (1900 – 2017). Multidisciplinary Respiratory Medicine 2018;13(1):15. doi: 10.1186/s40248-018-0128-5
2. Lelieveld J, Pozzer A, Poschl U, et al. Loss of life expectancy from air pollution compared to other risk factors: a worldwide perspective. Cardiovasc Res 2020 doi: 10.1093/cvr/cvaa025 [published Online First: 2020/03/04]
3. Nhung N, Amini H, Schindler C, et al. Short-term association between ambient air pollution and pneumonia in children: A systematic review and meta-analysis of time-series and case-crossover studies. Environ Pollut 2017;230:1000-08. doi: 10.1016/j.envpol.2017.07.063 [published Online First: 2017/08/03]
4. Cui Y, Zhang Z, Froines J, et al. Air pollution and case fatality of SARS in the People’s Republic of China: an ecologic study. Environ Health 2003;2(1):15. doi: 10.1186/1476-069x-2-15 [published Online First: 2003/11/25]
5. Pedrosa RHL. The dynamics of Covid-19: weather, demographics and infection timeline. MedRxiv 2020 [published Online First: 27.4.20]
6. Chen B, Kan H. Air pollution and population health: a global challenge. Environ Health Prev Med 2008;13(2):94-101. doi: 10.1007/s12199-007-0018-5 [published Online First: 2008/03/01]
7. Conticini E, Frediani B, Caro D. Can atmospheric pollution be considered a co-factor in extremely high level of SARS-CoV-2 lethality in Northern Italy? Environmental pollution (Barking, Essex : 1987) 2020:114465-65. doi: 10.1016/j.envpol.2020.114465
8. Ogen Y. Assessing nitrogen dioxide (NO2) levels as a contributing factor to coronavirus (COVID-19) fatality. Sci Total Environ 2020;726:138605. doi: 10.1016/j.scitotenv.2020.138605 [published Online First: 2020/04/18]
9. Wu X, Nethery RC, Sabath BM, et al. Exposure to air pollution and COVID-19 mortality in the United States: A nationwide cross-sectional study. medRxiv 2020:2020.04.05.20054502. doi: 10.1101/2020.04.05.20054502
10. Cheng V, Wong S, Chen J, et al. Escalating infection control response to the rapidly evolving epidemiology of the coronavirus disease 2019 (COVID-19) due to SARS-CoV-2 in Hong Kong. Infect Control Hosp Epidemiol 2020:1-6. doi: 10.1017/ice.2020.58 [published Online First: 2020/03/07]
11. Faridi S, Niazi S, Sadeghi K, et al. A field indoor air measurement of SARS-CoV-2 in the patient rooms of the largest hospital in Iran. Sci Total Environ 2020;725:138401. doi: 10.1016/j.scitotenv.2020.138401 [published Online First: 2020/04/14]
12. Tobias A, Carnerero C, Reche C, et al. Changes in air quality during the lockdown in Barcelona (Spain) one month into the SARS-CoV-2 epidemic. Sci Total Environ 2020;726:138540. doi: 10.1016/j.scitotenv.2020.138540 [published Online First: 2020/04/18]
13. Dutheil F, Baker J, Navel V. COVID-19 as a factor influencing air pollution? Environ Pollut 2020;263(Pt A):114466-66. doi: 10.1016/j.envpol.2020.114466
14. von Elm E, Altman D, Egger M, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Bmj 2007;335(7624):806-8. doi: 10.1136/bmj.39335.541782.AD [published Online First: 2007/10/20]
15. Dufault B, Klar N. The quality of modern cross-sectional ecologic studies: a bibliometric review. Am J Epidemiol 2011;174(10):1101-7. doi: 10.1093/aje/kwr241 [published Online First: 2011/09/24]
16. Carrion D, Kaali S, Kinney P, et al. Examining the relationship between household air pollution and infant microbial nasal carriage in a Ghanaian cohort. Environ Int 2019;133(Pt A):105150. doi: 10.1016/j.envint.2019.105150 [published Online First: 2019/09/14]
17. Mortimer K, Ndamala C, Naunje A, et al. A cleaner burning biomass-fuelled cookstove intervention to prevent pneumonia in children under 5 years old in rural Malawi (the Cooking and Pneumonia Study): a cluster randomised controlled trial. Lancet 2017;389(10065):167-75. doi: 10.1016/s0140-6736(16)32507-7 [published Online First: 2016/12/13]
18. Kirby M, Nagel C, Rosa G, et al. Effects of a large-scale distribution of water filters and natural draft rocket-style cookstoves on diarrhea and acute respiratory infection: A cluster-randomized controlled trial in Western Province, Rwanda. PLoS Med 2019;16(6):e1002812. doi: 10.1371/journal.pmed.1002812 [published Online First: 2019/06/04]
19. Jary H, Simpson H, Havens D, et al. Household Air Pollution and Acute Lower Respiratory Infections in Adults: A Systematic Review. PLoS One 2016;11(12):e0167656-e56. doi: 10.1371/journal.pone.0167656
20. Collaborators GBDLRI. Estimates of the global, regional, and national morbidity, mortality, and aetiologies of lower respiratory infections in 195 countries, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Infect Dis 2018;18(11):1191-210. doi: 10.1016/S1473-3099(18)30310-4 [published Online First: 2018/09/19]
21. Bhopal S, Boa G, Hughes R, et al. Can we improve the NHS’s ability to tackle covid-19 through emergency public health interventions? : BMJ; 2020 [Available from: https://blogs.bmj.com/bmj/2020/03/24/can-we-improve-the-nhss-ability-to-tackle-covid-19-through-emergency-public-health-interventions/ accessed 3rd April 2020.