Effect of Air Pollution Reductions on Mortality During the COVID-19 Lockdowns in Early 2020

Abstract

Introduction: COVID-19 lockdowns led to considerable reductions in air pollutant emissions worldwide, providing a unique opportunity to examine the impacts of reduced air pollution on mortality. This project aimed to quantify changes in nitrogen dioxide (NO₂) and fine particulate matter (PM_2.5) concentrations due to COVID-19 lockdowns, estimate associations between short-term exposures to these air pollutants and mortality rates, and calculate the attributable changes in mortality in four regions that implemented lockdowns but were mildly affected by the pandemic in early 2020, including Jiangsu Province, China; California, USA; Central and Southern Italy; and Germany.

Methods: To account for meteorological impacts and air pollution time trends, we used a machine learning-based meteorological normalization technique and the difference-in-differences approach to quantify changes in NO₂ and PM_2.5 concentrations due to lockdowns in early 2020. Using daily air pollution and mortality data from 2015 to 2019, we applied interactive fixed effects models (a causal modeling approach) to estimate associations between day-to-day changes in PM_2.5 and NO₂ concentrations and all-cause, natural-cause, and cardiovascular mortality rates before the pandemic in each region. Finally, using the quantified air pollution changes and the estimated air pollution-mortality relationships, we calculated the changes in mortality that were attributable to air pollution changes due to the lockdowns.

Results: We found that meaningful improvements in air quality occurred during the lockdowns in early 2020 in Jiangsu, China; California, USA; and Central and Southern Italy, with smaller magnitudes of reduction in PM_2.5 compared to NO₂. We observed no significant reduction in NO₂ and a small increase in PM_2.5 in Germany. After controlling for unmeasured spatial and temporal confounders, we detected statistically significant associations between short-term increases in PM_2.5 and NO₂ concentrations and increases in daily all-cause, natural-cause, and cardiovascular mortality rates in all four study regions from 2015 to 2019. Specifically, we determined that lockdown-induced reductions in NO₂ resulted in avoiding 1.41 (95% empirical confidence interval [eCI]: 0.94-1.88), 0.44 (95% eCI: 0.17-0.71), and 4.66 (95% eCI: 2.03-7.44) deaths per 100,000 people in Jiangsu, China; California, USA; and Central and Southern Italy, respectively. Mortality benefits attributable to PM_2.5 reductions in these regions also were statistically significant, albeit of a smaller magnitude, and resulted in avoiding 0.16 (95% eCI: 0.04-0.29), 0.23 (95% eCI: 0.03-0.43), and 0.91 (95% eCI: 0.09-1.78) deaths per 100,000 people in Jiangsu, China; California, USA; and Central and Southern Italy, respectively. In Germany, the mortality benefits attributable to NO₂ changes were not statistically significant (mortality change of -0.11; 95% eCI: -0.25 to 0.03 deaths per 100,000 people), and an observed increase in PM_2.5 was associated with an increase in mortality of 0.35 (95% eCI: 0.22-0.48) deaths per 100,000 people during the lockdown.

Conclusions: Using a causal modeling approach, this study contributes to the growing body of evidence that short-term exposures to PM_2.5 and NO₂ are associated with increased all-cause and cause-specific mortality rates. In areas mildly affected by the COVID-19 pandemic, lockdowns in early 2020 generally improved air quality and led to health benefits, especially in association with NO₂ reductions, with notable heterogeneity across regions. This study underscores the importance of accounting for local characteristics when policymakers adapt successful emission control strategies from other regions.

Statement Figure.

Changes in concentrations of nitrogen dioxide and fine particulate matter during COVID-19 lockdowns in early 2020 compared to prelockdown reference periods in four regions, after removing the influence of weather, seasonality, and year-to-year trends.

Figure 1.

Assumed causal relationships in Aim 2. This directed acyclic graph displays the assumed causal relationships among factors considered in estimating air pollution–mortality relationships in Aim 2.

Figure 2.

Daily mean PM_2.5 and NO₂ concentrations in each spatial unit, 2015–2019. These maps display the daily average PM_2.5 and NO₂ concentrations in each county or municipality in in (A) Jiangsu, China; (B) California, USA; (C) Central and Southern Italy; and (D) Germany. The unshaded areas in California, USA, represent counties without air quality monitoring stations for both PM_2.5 and NO₂; the unshaded areas in Central and Southern Italy represent municipalities with a population <10,000.

Figure 3.

Observed and deweathered daily NO₂ and PM_2.5 concentrations in Jiangsu, China (A); California, USA (B); Central and Southern Italy (C); and Germany (D) from January to May of 2020 versus 2015–2019. This figure displays the time series of the observed and deweathered air pollution concentrations from January to May 2020 and the average concentrations during the same calendar period from 2015 to 2019 in all four study regions. For Jiangsu, China, we used Chinese lunar calendar dates to account for the Chinese New Year holiday.

Figure 4.

Distribution of spatial unit–level changes in mean daily NO₂ and PM_2.5 concentrations due to the COVID-19 lockdown in each region. These box plots show the distribution of spatial unit–level changes in mean daily NO₂ and PM_2.5 concentrations due to the COVID-19 lockdown in each study region. Lower and upper box boundaries represent the 25th and 75th percentiles of the distribution; lower and upper error lines represent the 1.5 interquartile range below the third quartile and above the first quartile; the horizontal lines and the triangles inside boxes represent median and mean values, respectively; and the black dots represent outliers.

Figure 5.

Spatial distribution of the changes in mean daily NO₂ and PM_2.5 concentrations due to the COVID-19 lockdown in each study region. These maps display the spatial distribution of the changes in NO₂ and PM_2.5 concentrations due to the lockdown in each spatial unit in each study region. Unshaded areas represent spatial units that were excluded from the analysis because they did not have air quality monitoring sites.

Figure 6.

Estimated change in daily all-cause mortality rate (per 100,000 people) in 2015–2019 associated with a 10-μg/m³ increase in PM_2.5 or NO₂ concentration. This figure shows the estimated change in all-cause (ICD-9 codes 001–999; ICD-10 codes A00–Z99) daily mortality rates (per 100,000 people) per 10-μg/m³ increase in PM_2.5 or NO₂ concentration, based on single- and two-pollutant models, on different lag days in Jiangsu, China; California, USA; Central and Southern Italy; and Germany. Error bars represent 95% CIs.

Figure 7.

Estimated change in daily natural-cause and cardiovascular mortality rates (per 100,000 people) associated with a 10-μg/m³ increase in PM_2.5 or NO₂ concentration. This figure shows the estimated change in daily natural-cause (ICD-9 codes 001–799; ICD-10 codes A00–R99) and cardiovascular (ICD-9 codes 390–459; ICD-10 codes I00–I99) mortality rates (per 100,000 people) per 10-μg/m³ increase in PM_2.5 or NO₂ concentration, based on single- and two-pollutant models, on different lag days in Jiangsu, China; California, USA; Lazio, Italy; and Germany. Error bars represent 95% Cis; lag02 stands for the cumulative lag of 0–2 days after exposure; lag37 stands for the cumulative lag of 3–7 days after exposure; and lag07 stands for the cumulative lag of 0–7 days after exposure.

Figure 8.

All-cause mortality changes (per 100,000 people) attributable to air pollution changes due to the COVID-19 lockdown in each region. The box plot (upper figure) displays the distribution of spatial unit–level, mean, all-cause mortality changes (per 100,000 people) attributable to air pollution changes due to the COVID-19 lockdown in each region. Green represents NO₂, and blue represents PM_2.5. Left and right box boundaries represent the 25th and 75th percentiles of the distribution, respectively; left and right error lines represent the 1.5 interquartile range below the third quartile and above the first quartile, respectively; the vertical lines and triangles inside boxes represent median and mean values, respectively; and the black dots represent outliers. The bar plot (lower figure) presents the total estimated changes in all-cause mortality (per 100,000 people) attributable to NO₂ and PM_2.5 changes due to the lockdown, summed across all spatial units in each study region. Green represents NO₂, and blue represents PM_2.5. Error bars represent 95% CIs.

Figure 9.

Spatial distribution of all-cause mortality changes (per 100,000 people) attributable to air pollution changes due to the COVID-19 lockdown in each study region. These maps display the spatial distribution of the changes in all-cause mortality (per 100,000 people) attributable to NO₂ and PM_2.5 changes due to the lockdown in each spatial unit in each study region. Unshaded areas represent spatial units that were excluded from the analysis because they did not have air quality monitoring sites.

Figure 10.

Cause-specific mortality changes (per 100,000 people) attributable to air pollution changes due to the COVID-19 lockdown in each region. This bar plot presents the estimated changes in natural-cause and cardiovascular mortality (per 100,000 people) attributable to NO₂ and PM_2.5 changes due to the lockdown in each study region. Error bars represent 95% CIs.

Commentary Figure 1.

Policy-based lockdown periods (used in main analyses) and mobility-based lockdown periods (used in sensitivity analyses) in the four study regions during early 2020 and during earlier years for comparison.

Commentary Figure 2.

Mean changes in concentrations of NO₂ and PM_2.5 during COVID-19 lockdowns in early 2020 compared to prelockdown reference periods in California, USA; Central and Southern Italy; Germany; and Jiangsu, China, after removing the influence of weather, seasonality, and year-to-year trends. Error bars show variation (95% empirical confidence intervals) among municipalities or counties.

Commentary Figure 3.

Estimated change in daily all-cause mortality rate (per 100,000 people) associated with a 10-μg/m³ increase in NO₂ or PM_2.5 concentration during the prepandemic period (2015–2019). Data are presented with 95% confidence intervals and for main lags only (cumulative 0- to 1-day lag for associations with NO₂ in Germany and with PM_2.5 in both Germany and Jiangsu; cumulative 0- to 2-day lag for all other associations).