Combined effects of ambient air pollution and temperature on mortality in Thailand

Abstract

This study aimed to investigate the combined effects of air pollution and temperature on mortality in 34 provinces of Thailand by modeling temperature as a confounding factor and effect modifier, estimating the effects of air pollution at low, moderate, and high temperature categories defined by the 1st and 99th province-specific temperature percentiles. When the temperature was modeled as a confounding factor, the relative risk (RR) of mortality associated with a 10 µg/m3 increase in PM10 and PM2.5 (lag 0–2), and a 10-ppb increase in NO2 (lag 0–2) and O3 (lag 0–7) was respectively 1.0096 (95% Confidence Interval (CI): 1.0073, 1.0118), 1.0134 (95% CI: 1.0099, 1.0170), 1.0172 (95% CI: 1.0122, 1.0222), and 1.0164 (95% CI: 1.0093, 1.0236). Regarding temperature as an effect modifier, the combined effects of air pollution and temperature were observed as a U-shaped pattern, where the effects of PM10, PM2.5, NO2, and O3 on mortality were greater at low (< 1st percentile) and high (> 99th percentile) temperature days compared to those at moderate temperature days (1st – 99th percentile). The pattern of combined effects of air pollution and temperature remained robust even when different temperature percentiles were employed, except for that of NO2. Furthermore, the estimated effects of PM10, PM2.5, NO2, and O3 on mortality at high-temperature days were mitigated by high green density. Findings of this study revealed that extreme temperature (both hot and cold) could exacerbate the effect of air pollution on mortality, and higher green density mitigate the combined effects of air pollution and high temperature.