Researchers have developed a practical prediction model to identify newborns at risk of severe complications at the start of the second stage of labor. In a retrospective study of over 22,000 term, singleton pregnancies, the model used electronic fetal monitoring (EFM) data alongside maternal and labor characteristics.
The study compared three approaches—logistic regression, LASSO, and extreme gradient boosting—with logistic regression showing the best performance. Babies were stratified into risk groups, with adverse outcomes occurring in 2.6% of the low-risk group, 6.7% of the higher-risk group, and 10.3% of the highest-risk group. Key predictors included the presence of meconium, fetal tachycardia within two hours of the second stage, and the number of prior deliveries.
According to the researchers, this model could help clinicians better anticipate complications during labor and guide timely interventions, potentially improving neonatal outcomes.
Please login to comment on this article