Association between ambient particulate matter and preterm birth stratified by temperature: A population-based pregnancy cohort study

A growing body of literature reports associations between exposure to particulate matter with aerodynamic diameters <= 2.5 mu m (PM2.5) and 2.5-10 mu m (PM10-2.5) during pregnancy and preterm birth (PTB). However, the role of ambient temperature in PM-PTB associations was rarely investigated. In Israel, we used Maccabi Healthcare Services data to establish a population-based cohort of 381,265 singleton births reaching 24-42 weeks' gestation and birth weight of 500-5000 g (2004-2015). Daily PM and ambient temperature predictions from a satellite-based spatiotemporal model, at a 1 x 1 km spatial resolution, were linked to the date of birth and maternal residence. Mixed effects Cox regression models, adjusted for covariates, with a random intercept at the mother level were used to assess associations between mean exposure during pregnancy and PTB. We found that exposure to PM2.5 was positively associated with PTB when the average exposure during pregnancy was either low (first quintile) or high (fifth quintile), compared to exposure in the 2nd-4th quintiles, with hazard ratios (HRs) 1.18 (95% confidence interval [CI], 1.13-1.24) and 1.07 (95% CI, 1.02-1.12), respectively. The results revealed effect modification of temperature. For mothers exposed to low (below median) average temperature during pregnancy, HRs of PTB were 0.93 (95% CI, 0.87-1.00) and 1.21 (95% CI, 1.14-1.29) for the first and fifth PM2.5 quintiles, respectively, when compared to the 2nd-4th quintiles. However, a reverse trend was indicated for high-temperature pregnancies, where the corresponding HRs were 1.48 (95% CI, 1.39-1.58) and 0.92, (95% CI, 0.96-0.98). In conclusion, consideration of climatic factors can provide new insights into the risk of PTB as a result of exposure to PM2.5 during pregnancy.