Long-term exposure to air pollution and COVID-19 mortality and morbidity in Denmark: Who is most susceptible? (AIRCODEN)

This study will investigate whether exposure to air pollution increases the risk of severe COVID-19 outcomes, and identify the most susceptible groups by socioeconomic status, ethnicity, and comorbidities. Between March and December 2020, the investigators followed more than 3 million adults in Denmark. They will evaluate whether long- (40 years) and short- (1 year) term exposure to several common air pollutants increased the risk of COVID-19 hospitalization and mortality.

Poster abstract for HEI Annual Conference 2022

Long-term exposure to air pollution and COVID-19 mortality and morbidity in Denmark: Who is most susceptible? (AIRCODEN)

Zorana Jovanovic Andersen,¹ Jeanette Therming Jørgensen,¹ Amar Mehta,¹ Theis Lange,¹ Rudi Westendorp,¹ Tom Cole-Hunter,¹ Shadi Azam,¹ Jørgen Brandt,² Thea Kølsher Fisher,³ Steffen Loft¹, Gerard Hoek,⁴ Youn-Hee Lim¹

¹University of Copenhagen, Copenhagen, Denmark; ²Aarhus University, Roskilde, Denmark; ³Nordsjællands Hospital, Hillerød, Denmark; ⁴Utrecht University, Utrecht, the Netherlands

Background: Long-term exposure to air pollution increases the risk of respiratory and cardio metabolic diseases, which increases the risk of death from COVID-19, leading to the hypothesis that air pollution may increase susceptibility to mortality and morbidity from COVID-19. Early results from a handful of individual-level cohort studies suggest associations between long-term exposure to air pollution and COVID-19 mortality and morbidity, but it remains unclear who is most susceptible. We aim to examine whether long-term exposure to particulate matter < 2.5 µg/m³ in diameter (PM_2.5), particulate matter < 10 µg/m³ in diameter (PM₁₀), nitrogen dioxide (NO₂), black carbon (BC), and ozone (O₃) are associated with COVID-19 related mortality (Aim 1), hospitalizations (Aim 2) and incidence (Aim 3), and to identify the most susceptible groups by socio-economic status (SES), ethnicity, and co-morbidities.

Methods: In Aim 1, a cohort of 5,600,000 subjects residing in Denmark on March 1^st, 2020, will be followed until the date of first COVID-19 infection, death from other reasons, emigration, or the end of follow-up on 30.4.2021. In Aims 2 and 3, 3,056,854 Danes age 30 years or older and residing in Denmark on March 1^st, 2020, will be followed until the date of first COVID-19 hospitalization/death, death from other reasons, emigration, or the end of follow-up on 30.4.2021. Air pollution will be assigned to AIRCODEN cohort subjects at the residential address in 2020 and with complete moving history available from 1979-2019, by: 1) the Danish integrated DEHM/UBM model providing annual means of NO₂, PM_2.5, PM₁₀, BC, O₃, from 1979 until 2019; 2) the European-wide hybrid land use regression model developed within the Effects of Low-Level Air Pollution: A Study in Europe (ELAPSE) project, providing annual concentrations of PM_2.5, NO₂, BC, and O₃ for year 2010. We will estimate the association between long-term exposure to air pollution (40-year (1979-2019) and 1-year (2019)) residential mean levels of PM_2.5, PM₁₀, NO₂, BC, and O₃ and COVID-19 incidence/mortality/hospitalizations, using Cox regression models, adjusting for individual-level (education, income, employment status, housing, marital status, and country of origin) and area-level (parish) SES factors (population density, mean income, % unemployment, % primary or lower education, access to healthcare). Effect modification by individual-level SES, ethnicity and co-morbidities with cardiovascular disease, respiratory diseases, diabetes, lung cancer and dementia, and ethnicity will be evaluated by entering interaction terms into the model. Separate analyses will be performed for different time-periods, defined by stages of lockdown as well as increasing capacity of testing strategies.

Conclusions. Our findings will provide important new data to elucidate whether long-term exposure to ambient air pollution contributes to COVID-19 morbidity and mortality, identify relevant pollutants, as well as who is most suscpetible.