Mediation of Fine Particulate Matter on the Association Between Daily Temperature and Mortality

Background: Global warming has raised the earth’s temperature approximately 1°C compared to the pre-industrial level. Particulate matter is one pathway through which climate change affects health. This study aimed to estimate the attribution of PM_2.5 on mortality associated with increased temperatures in seven major Korean cities. Methods: Our data includes daily non-accidental mortality counts, mean humidity, mean PM_2.5 concentration, and mean temperature from 2010 to 2019. Data on daily mean temperature was obtained from the public data portal of the Korea Meteorological Administration. Because the monitoring of PM_2.5 concentration in South Korea started in 2015, we utilized modeling data of the daily mean PM_2.5 concentration from 2010 to 2014. For the period between 2015 and 2019, we used monitoring stations to obtain the daily mean PM_2.5 concentration. Non-accidental mortality counts (ICD-10: A00-R99) were extracted from the cause-of-death statistics in Korea. To measure the attribution of PM_2.5, we conducted mediation analysis. Using the mediation analysis results, we estimated attributable fraction (AF) of increase in temperature compared to climate normal on mortality. Results: The mediating effects of PM_2.5 were observed on days warmer than minimum mortality temperature. The pattern of this mediation effect was similarly observed among those aged ≥ 65 years. In 2019, the AF of the total effect of temperature was from 3.15% to 4.25% in these cities. The AF of indirect effect through PM_2.5 was up to 11% of that of the total effect. Conclusion: We observed that part of the health risk from increasing temperature is attributable to PM_2.5. These results suggest that the health impacts of climate change can be mitigated by regulating particulate matter pollution as well as greenhouse gas emissions.