Ambient temperature and the risk of preterm birth: A national birth cohort study in the mainland China

Background: Little is known about the effect of ambient temperature on preterm birth, especially for the trimester-specific effects. Objective: To evaluate whether exposure to relatively low or high temperature during pregnancy is associated with increasing risk of preterm birth or not. Method: We analysed the data of a birth cohort with 1,281,859 singleton pregnancies during 2013-2014 and matched the home address of each pregnant women to the model based daily meteorological and air pollution data. Then we used the Cox proportional hazard regression models with random effect to estimate the non-linear associations between exposure to relatively low or high temperature at each trimester of pregnancy and the risk of preterm birth, after controlling for air pollution and individual-level covariates. Finding: The overall preterm birth rate was 8.1% (104,493 preterm births). Exposure to relatively low or high temperatures during the entire pregnancy significantly increase the risk of preterm birth, with hazard ratios (HRs) [95% confidence intervals (CIs)] of 1.03 (95%CI: 1.02, 1.04) for relatively low (9.1 degrees C, the 5th percentile) temperature and 1.55 (95%CI: 1.48, 1.61) for relatively high (23.0 degrees C, the 95th percentile) temperature in comparison with the thresholds (12.0.). Pregnant women at the early pregnancy (the 1st and 2nd trimester) are more susceptible to high temperatures while pregnant women at the late pregnancy (the 3rd trimester) are more susceptible to low temperatures. Conclusion: These findings provide new evidence that exposure to relatively low or high temperatures during pregnancy increases the risk of preterm birth, which can serve as scientific evidence for prevention of preterm birth.