Differential effects of air pollution on ischemic stroke and ischemic heart disease by ethnicity in a nationwide cohort in the Netherlands

Abstract

Introduction Air pollution is a major risk factor for cardiovascular diseases and exposure is unequally distributed among ethnic groups. Knowledge on whether the effect of air pollution on cardiovascular disease differs between ethnic groups can provide insight into the mechanisms leading to ethnic inequalities in disease outcomes. We explored differences in associations between air pollution and ischemic stroke and ischemic heart disease (IHD) for the six largest ethnic groups in the Netherlands. Methods For this nationwide analysis between 2014 and 2019, we linked residential-address concentrations of NO2 and PM2.5 were to individual-level hospital discharge and mortality data centralized by Statistics Netherlands. To evaluate the outcome incident ischemic stroke (ICD10 codes I63, G45), we created a cohort of residents 30 years and free of ischemic stroke and for incident IHD (ICD10 codes I20-I25) we created a cohort free of IHD at baseline. We performed Cox proportional hazard survival analyses in each cohort with the 2014 annual average concentrations of PM2.5 or NO2 as determinants, standardized by their interquartile ranges (IQR). Analyses were stratified by ethnicity (Dutch, German, Indonesian, Surinamese, Moroccan, Turkish) and adjusted for age, sex, urbanicity, socioeconomic indicators and geographical region. Results Both cohorts consisted of >9.5 million people. During follow-up, 127,673 (1.3%) people developed ischemic stroke. For ischemic stroke, the p-values of the likelihood ratio test comparing models with and without an interaction term between the air pollutant and ethnicity were 0.057 for NO2 and 0.055 for PM2.5. The hazard ratio of 1 IQR increase (6.42 µg/m3) of NO2 for ischemic stroke was lowest for Moroccans (fully corrected model: 0.92 [0.84-1.02], p-value=0.032 difference with Dutch) and highest for Turks (fully corrected model: 1.09 [1.00-1.18], p-value=0.157 difference with Dutch). Results of PM2.5 exposure were similar to those of NO2. During follow-up, 156,517 (1.6%) people developed IHD. The associations of air pollution with IHD were in unexpected directions, such that higher exposure was associated with lower IHD incidence. For IHD, the p-values of the interaction between air pollution and ethnicity were 1.75*10-5 for NO2 and 1.06*10-3 for PM2.5. The hazard ratio’s for IHD were lowest for Turks (NO2: 0.88 [0.83-0.92], p-value=2.0*10-4 difference with Dutch, PM2.5: 0.86 [0.82-0.91], p-value=1.3*10-4 difference with Dutch) and highest for Surinamese (NO2: 1.02 [0.97-1.07], p-value=0.014 difference with Dutch) and Dutch (PM2.5: 0.96 [0.94-0.98]) in the fully corrected models. Conclusion The associations between air pollutants and ischemic stroke or IHD differ notably between ethnic groups in the Netherlands. Air pollution lowering measures and policies to prevent CVD should target populations that are vulnerable to air pollution exposure and have a high cardiovascular disease risk.