Short-term exposure to ambient air pollution and individual emergency department visits for COVID-19: a case-crossover study in Canada

Background Ambient air pollution is thought to contribute to increased risk of COVID-19, but the evidence is controversial. Objective To evaluate the associations between short-term variations in outdoor concentrations of ambient air pollution and COVID-19 emergency department (ED) visits. Methods We conducted a case-crossover study of 78 255 COVID-19 ED visits in Alberta and Ontario, Canada between 1 March 2020 and 31 March 2021. Daily air pollution data (ie, fine particulate matter with diameter less than 2.5 µm (PM2.5), nitrogen dioxide (NO2) and ozone were assigned to individual case of COVID-19 in 10 km × 10 km grid resolution. Conditional logistic regression was used to estimate associations between air pollution and ED visits for COVID-19. Results Cumulative ambient exposure over 0–3 days to PM2.5 (OR 1.010; 95% CI 1.004 to 1.015, per 6.2 µg/m3) and NO2 (OR 1.021; 95% CI 1.015 to 1.028, per 7.7 ppb) concentrations were associated with ED visits for COVID-19. We found that the association between PM2.5 and COVID-19 ED visits was stronger among those hospitalised following an ED visit, as a measure of disease severity, (OR 1.023; 95% CI 1.015 to 1.031) compared with those not hospitalised (OR 0.992; 95% CI 0.980 to 1.004) (p value for effect modification=0.04). Conclusions We found associations between short-term exposure to ambient air pollutants and COVID-19 ED visits. Exposure to air pollution may also lead to more severe COVID-19 disease.

Models represent pooled health region-specific estimates derived using two-stage random effects meta-analysis and meta-regression. Models adjusted for daily mean ambient temperature, relative humidity, the effective reproduction number, the OxCGRT Government Response Index and population density and percentage of the population self-identified as Black as meta-predictors. 2 The variance due to heterogeneity estimated by the I²-statistic.  Table S4. Second-stage random-effects meta-analysis and meta-regression models for the associations between PM2.5 (per 6.2 µg/m 3 ) and COVID-19 ED visits: multivariate Wald test on significance of each meta-predictor in explaining variation in overall associations, Cochran Q test for heterogeneity and I 2 statistics for residual heterogeneity. Different meta-regression models are being presented: base model (i.e. only including pooled ORs) and models with different meta-predictors. Random-effects multivariate meta-regression models were used to test potential effect modification by between-city differences in meta-predictors. The outcome variables in the meta-regression models in this study were the pooled estimates and the explanatory variables (i.e. potential effect modifiers) were the continuous variables at the health region level. Effect modification was considered statistically significant if the effect modifier's p-value was less than 0.05.

Meta
BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)  Table S5. Ssecond-stage random-effects meta-analysis and meta-regression models for the associations between NO2 (per 7.7 ppb) and COVID-19 ED visits: multivariate Wald test on significance of each meta-predictor in explaining variation in overall associations, Cochran Q test for heterogeneity and I 2 statistics for residual heterogeneity. Different meta-regression models are being presented: base model (i.e. only including pooled ORs) and models with different meta-predictors. Random-effects multivariate meta-regression models were used to test potential effect modification by between-city differences in meta-predictors. The outcome variables in the meta-regression models in this study were the pooled estimates and the explanatory variables (i.e. potential effect modifiers) were the continuous variables at the health region level. Effect modification was considered statistically significant if the effect modifier's p-value was less than 0.05.

Meta-predictors
BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)  Table S6. Ssecond-stage random-effects meta-analysis and meta-regression models for the associations between O3 (per 10.8 ppb) and COVID-19 ED visits: multivariate Wald test on significance of each meta-predictor in explaining variation in overall associations, Cochran Q test for heterogeneity and I 2 statistics for residual heterogeneity. Different meta-regression models are being presented: base model (i.e. only including pooled ORs) and models with different meta-predictors. Random-effects multivariate meta-regression models were used to test potential effect modification by between-city differences in meta-predictors. The outcome variables in the meta-regression models in this study were the pooled estimates and the explanatory variables (i.e. potential effect modifiers) were the continuous variables at the health region level. Effect modification was considered statistically significant if the effect modifier's p-value was less than 0.05.

Meta-predictors
BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)  Table S7. Odds ratios 1 (ORs) and 95% CIs for associations between the cumulative effects of ambient air pollutants over 0 to 3 days (per interquartile range increase) and emergency department visits for COVID-19, stratified by whether patients came from institutional settings and by time period of the study. ORs reflect a 6.2 µg/m 3 change in PM2.5, a 7.7 ppb change in NO2 and a 10.8 ppb change in O3. Models represent pooled health region-specific estimates derived using two-stage random effects meta-analysis and meta-regression. 0.994 (0.988 -1.000) I 2 = 0.0% (0.50) P value for effect modification 0.55 0.03 (<0.01) I 2 (P value for heterogeneity) 2 I 2 = 5.2% (0.37) I 2 = 43,6% (<0.01) I 2 = 71.3% (<0.01) 1 Models represent pooled health region-specific estimates derived using two-stage random effects meta-analysis and meta-regression. ORs reflect a 6.2 µg/m 3 change in PM2.5, a 7.7 ppb change in NO2 and a 10.8 ppb change in O3. Models adjusted for daily mean ambient temperature, relative humidity, the effective reproduction number, the OxCGRT Government Response Index and population density and percentage of the population self-identified as Black as meta-predictors. I 2 : The variance due to heterogeneity estimated by the I²-statistic for the strata models and the models when calculating the p value for effect modification. In parentheses, the p values for the statistical significance of heterogeneity are reported.

Characteristics
BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)  Table S8. Odds ratios 1 (ORs) and 95% CIs for associations between acute exposure to ambient air pollutants and emergency department visits for myocardial infarction. ORs reflect a 1.8 µg/m 3 change in PM2.5, a 2.3 ppb change in NO2 and a 11.7 ppb change in O3 (N = 26,437 2.5% (0.95) 0.0% (0.99) 0.0% (0.98) 1 Models represent pooled health region-specific estimates derived using two-stage random effects meta-analysis and meta-regression. Models adjusted for daily mean ambient temperature, relative humidity and the OxCGRT Government Response Index 2 The variance due to heterogeneity estimated by the I²-statistic.
BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)  Figure S1. Odds ratios 1 (ORs) and 95% CIs for associations between PM2.5 (per 6.2 µg/m 3 ) and emergency department visits for COVID-19 for lags 0 to 21 days. Models represent pooled health region-specific estimates derived using two-stage random effects meta-analysis and meta-regression. Models adjusted for daily mean ambient temperature, relative humidity, the effective reproduction number, the OxCGRT Government Response Index and population density and percentage of the population self-identified as Black as meta-predictors.
BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)  Figure S2. Directed acyclic graph for estimating the direct effect of ambient air pollution exposure on COVID-19 ED visits. Parameters in red are potential confounding factors. Green line: causal path. According to the DAG, the minimal sufficient adjustment for estimating the total effect of ambient air pollution on COVID-19 ED visits is: ambient temperature, Government Stringency Index, Relative humidity, Rt BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)