Prior literature has found that extreme temperature exposure is associated with preterm birth (PTB). However, current evidence provides heterogeneous conclusions, and data on extreme cold and across different pre-pregnancy body mass index (BMI) statuses are limited.
We conducted a population-based retrospective cohort of 251,257 women between 2014 and 2017 in Guangdong, China, to evaluate whether the association between extreme temperature exposure and PTB varied in pre-pregnancy BMI status. Participants were divided into three categories based on pre-pregnancy BMI: underweight (BMI < 18.5 kg/m2), normal weight (18.5–23.9 kg/m2), overweight or obesity (≥ 24.0 kg/m2). We fitted Cox proportional hazards models to assess the association between daily mean temperature and PTB at each trimester for each BMI category separately. The hazard ratios (HRs) at the 5th and 95th percentiles of temperature (defined as low and high temperatures respectively) were provided using the median temperature at each trimester as a reference.
58,220 (23.2%) were underweight, and 27,865 (11.1%) were overweight or obese. Of the 251,257 women, 18,612 (7.41%) had PTB delivery. Both low-and high-temperature exposure increased the risk of PTB in the third trimester, while cold exposure mostly mitigated the risk for the first and second trimesters. The association with low temperature was the strongest in the third trimester, especially for underweight women (HR: 1.825; 95%CI: 1.529 ~ 2.179), while the association with high temperature was the strongest also in the third trimester, especially for obese or overweight women (HR:1.825; 95%CI:1.502 ~ 2.218). Furthermore, the attributable fractions of PTB risk in the third trimester were estimated as 5.59% (95% CI: 3.58, 7.98%) for cold exposure among underweight women and 3.31% (95% CI: 2.01, 4.88%) for hot exposure among overweight or obese women.
Exposure to either low temperature in the third trimester or high temperature during pregnancy was associated with a higher risk of PTB. Moreover, pre-pregnancy BMI status might affect the susceptibility of pregnant women. Such findings would be useful to develop targeted measures for vulnerable populations.