Effect modification of particulate matter air pollution on heat-related mortality in five provinces of Thailand: a two-stage time-series study

Abstract The potential compounding effect between high temperature and air pollution is critical. However, the evidence is limited in tropical climate areas where temperatures hit record highs year after year. This study aimed to examine whether the heat-related mortality is modified by particulate matter (PM) in five urban or industrial provinces of Thailand. We used a two-stage time-series design and collected daily data of temperature, humidity, PM and mortality (non-accidental, cardiovascular and respiratory) from 2015 to 2019. We fitted a generalized linear model with a Poisson distribution and incorporated an interaction term between PM ⩽ 10 μ m (PM 10 ), PM ⩽ 2.5 μ m (PM 2.5 ) in aerodynamic diameter and temperature to examine the effect modification of PM on heat-related mortality during the six warmest months for each province, adjusting for time-varying confounders, such as day of week, relative humidity, long-term and seasonal trend. Relative risks (RR) of mortality for heat were estimated between the 99th and 50th percentiles of temperature. A random-effects meta-analysis was used in second stage to combine the province-level estimates. Causes of death, sex, and age groups (<65, 65–79, and ⩾80 years) were stratified for subgroup analysis. A total of 177 006 deaths were included. We observed a pattern of effect modification, where the risk of mortality for heat increased on days with higher levels of PM. This pattern was more pronounced for respiratory mortality, with the RRs for heat rising from 1.38 (95%CI: 1.05, 1.82) to 1.81 (95%CI: 1.44, 2.28), as PM 2.5 levels increased from 20 μ g m −3 to 55 μ g m −3 . We found that PM air pollution elevated the heat-related mortality. These results underscore the critical need to address the compounding effects of extreme heat and air pollution in mitigation and adaptation strategies to protect public health.