The Synergistic Effect of Short and Long Term Exposure to PM2.5 and Temperature on Hospital Admissions in New England

Abstract

Background/Aim: Particulate-pollution and heat are strong predictors of morbidity, yet studies that assessed their joint effect on health, while assessing both short and long-term exposure, are scarce. We aimed to investigate the joint effect of short-term exposure to PM2.5 and temperature on hospital admissions, accounting for the long-term exposures. Methods: We included respiratory, ischemic-heart-disease (IHD) and ischemic stroke admissions of residents (age≥ 65) across New-England between 2004-2011 and constructed daily counts per zip code. We obtained PM2.5 and temperature exposure estimates from a model incorporating daily satellite remote sensing data at 1km spatial resolution. We defined short-term exposures as the difference between the two day average of exposures (lags 0-1) and the zip code mean exposure across the study period, and long term exposures as annual moving averages. We performed separate Poisson regressions with a random intercept per zip code for each admission-cause on short-term exposures (including their interaction) and long-term exposures, with adjustment for time and area-level socioeconomic factors. Results are presented as Relative Risks for IQR increase of PM2.5 (6.1µg/m3) or temperature (90C). Results: We found a significant interaction between short-term PM2.5 and temperature for respiratory admissions, where the effect of PM2.5 was larger in the 75th percentile of temperature (1.010[1.005;1.015]) compared to the 25th percentile (1.004[0.999;1.008]). Increase in long term exposure to PM2.5 was associated with 1.077[1.033;1.123] increase in admissions. Increases in short and long-term exposures to PM2.5 were associated with increases in admissions for IHD (1.009[1.003;1.016] and 1.082[1.018;1.150],respectively) and stroke (1.013[1.005;1.021] and 1.304[1.208;1.419],respectively). For temperature, long-term exposure was associated with increases of 2.229[1.737;2.859] and 1.191[0.984;1.442] in IHD and stroke admissions. Conclusions: We observed a synergistic effect of short-term exposures to PM2.5 and temperature and respiratory admissions; with higher PM2.5 effects on warmer days. Long-term exposures were translated into larger increases in hospital admissions compared to short-term exposures.