The relationship between maternal environmental temperature exposure and preterm birth: A Risk prediction based on machine learning

Abstract

Global warming and the risk of preterm birth are both major factors that impact population health. This study investigated the impact of environmental temperature during different stages of pregnancy on the probability of preterm birth (PTB) in Wuhan, China through 2014 to 2016. The results revealed that temperature exposure throughout the entire pregnancy exhibited a U-shaped relationship with the risk of PTB; when temperature exposure during the entire pregnancy was below 14 °C or above 20 °C, the risk of PTB increased. Early pregnancy exposure to temperatures below 7 °C or above 22 °C, and late pregnancy exposure to temperatures below 7 °C or above 26 °C, were associated with an increased risk of PTB. Additionally, elevated PM_2.5 exposure increased PTB risk, while O₃ exposure exhibited a U-shaped relationship with preterm birth. Compared to non-high-risk pregnancies, high-risk pregnancies exhibited a higher risk of preterm birth across all stages of pregnancy. Notably, when late pregnancy temperature exposure exceeded 28 °C, the risk of PTB rapidly increased for non-high-risk pregnancies. This research has significant implications for improving maternal and new-born health by future sustainable city planning and the optimization of temperature forecast warning systems, particularly under the dual pressures of rapid urbanization and climate change.