Extreme urban temperature exposure and preterm birth: Spatial-temporal risk zone prediction using machine learning...

Area of Science:

• Environmental epidemiology
• Public health
• Climate change impacts

Background:

• Preterm birth (PTB) is a major global health concern.
• Environmental factors, particularly temperature, are increasingly recognized as potential contributors to PTB.
• Understanding the spatiotemporal relationship between temperature and PTB is crucial for targeted interventions.

Purpose of the Study:

• To investigate the impact of ambient temperature on preterm birth (PTB) risk.
• To identify specific temperature thresholds and variability associated with increased PTB.
• To map spatial patterns of temperature-related PTB risk.

Main Methods:

• Utilized GPS coordinates of 311,972 pregnant women in Wuhan, China.
• Integrated daily environmental data with residential addresses.
• Developed and applied a machine learning model for exposure-outcome analysis.
• Conducted spatial analysis to identify risk zones.

Main Results:

• PTB risk elevated with temperatures <14°C or >21°C.
• Excessive temperature variability and acute extreme weather exposure increased PTB risk.
• Heat-related PTB risk concentrated in urban heat islands; cold-related risks were more widespread.

Conclusions:

• Temperature extremes and variability are significant risk factors for preterm birth.
• Spatial disparities in temperature-related PTB risk exist, influenced by urban heat island effects.
• Findings support evidence-based urban planning and climate adaptation strategies to mitigate PTB.