Association between short-term exposure to fine particulate matter components and cause-specific mortality in Korea: a multi-city time-series study

Abstract

Background/Aim: Many studies have reported fine particulate matter (PM2.5) has various health effects. The effects of PM2.5 have been shown to vary in space and time. One of the underlying explanations was the variability in the distribution of chemical components in PM2.5. Therefore, identifying whether the health effects of each component of PM2.5 differ can provide evidence for the heterogeneous mortality risk of PM2.5. Our study aims to estimate the associations between short-term exposure to components and cause-specific mortality across six major cities in Korea. Methods: Daily mass concentration of PM2.5 and 25 components of PM2.5 were obtained from the National Institute of Environmental Research for 2012-2019. We used six intensive air monitoring stations in each of the six major cities: Seoul, Incheon, Daejeon, Gwangju, Ulsan, and Jeju. We employed a two-stage analysis. In the first stage, we conducted generalized additive model with quasi-Poisson regression to estimate city-specific associations of mortality with PM2.5 and its components on the same day. We excluded Asian dust days and adjusted the model for temperature, relative humidity, and time trend. In the second stage, we used fixed-effects meta-analysis to pool the city-specific estimates. Results: We could not find statistically significant association of PM2.5 and all components with all-cause mortality. Potassium was associated with 2.2% increment (95% CI:0.4%, 4.0%) in cardiovascular mortality per IQR increase (0.24μg/m3). Magnesium ion was associated with 0.7% increment (95% CI: 0.1%, 1.4%) in respiratory mortality per IQR increase (0.01μg/m3). Conclusion: This study suggests potential adverse health effects of particles produced by dust, salt, or traffic. In contrast to previous studies, our results showed no association of elemental carbon or nitrate with any mortality. Further study would be needed to account for the joint and lagged effect of components and meteorological variables. Keyword: fine particulate matter, components, multi-city analysis, epidemiology, health