Regional Difference in the Association Between Long-Term PM and Cardiovascular Disease Incidence and Potential Determinants of the Difference

While short-term studies of the adverse effects of air pollution have found regional differences, there has been insufficient evidence from long-term studies. This study investigated the spatial variation of the association between long-term exposure to particulate matter (PM) air pollution and cardiovascular disease (CVD) incidence and their relevant regional characteristics. We selected 155,017 adults who have lived over 13 years in each of the 16 regions of South Korea from the National Health Insurance Service-National Sample Cohort. We assessed the individual long-term exposure as a 5-year average concentration at district-level residential addresses estimated by previously validated exposure prediction models. We applied time-dependent Cox-proportional hazard models to explore the spatial variation of long-term PM exposure's association with CVD incidence. After adjusting for individual-level characteristics, we estimated the region-specific hazard ratios (HRs) of incident CVD per 10 μg/m³ increase in PM₁₀ and PM_2.5. Then, we calculated the correlation coefficients between region-specific HRs and 80 regional attributes in 13 categories across 16 regions to identify the related regional characteristics. Region-specific HRs of CVD incidence for PM showed considerable variation in magnitude and direction across 16 regions. HRs of PM₁₀ and PM_2.5 were higher in major urban areas and lower in rural areas. Urbanicity and emission sources exhibited significant correlations with HRs of overall CVD. Our findings offer an opportunity to explore regional characteristics that derive the difference in air pollution-associated health effects and help us provide guidance to target specific characteristics to avoid the adverse health effects of air pollution.