Machine Learning to Predict Fever After Labor Analgesia

Post-labor analgesia intrapartum fever (PAIF) is a serious complication that increases the risk of cesarean delivery, postpartum bleeding, and health problems for newborns. Currently, doctors lack a simple tool to predict which mothers will develop a fever before they start labor analgesia. This study aims to develop and test a computer prediction tool (using machine learning) to estimate the risk of PAIF. The investigators will review the medical records of mothers who received labor analgesia and delivered at term in three hospitals in China. Information that is available before the mothers receive analgesia, such as maternal age, gestational age, blood test results, and labor progress, will be collected. Using artificial intelligence, the investigators will build a prediction model. This model will then be tested on a separate group of women from a different hospital to ensure it works well in different settings. Finally, an open-access, interactive online calculator will be developed for clinicians. The goal of this study is to provide doctors with a practical tool to quickly assess a mother's risk level before starting analgesia. This will help doctors monitor high-risk mothers more closely and prepare for possible complications. This study only uses existing medical records and will not change the routine care of any patients. Further studies will be needed to see if using this tool improves the health of mothers and babies.

Inclusion Criteria: * Parturients with singleton pregnancies. * Opted for and received labor analgesia. * Delivered at term (37-42 weeks of gestation). * Admitted to the participating study hospitals. Exclusion Criteria: * Diagnosed with active infections at the onset of labor. * Axillary temperature ≥ 37.3°C after entering labor but before receiving labor analgesia. * Missing data on key variables for the prediction model.