Leveraging Administrative Health Records and Machine Learning for Population-Level Prediction of Preterm Birth

The authors aimed to develop and evaluate machine learning models for predicting preterm birth at 26 weeks of gestation using routinely collected administrative health records, addressing the limitations of existing prediction tools that rely on clinical data. Their study found that a gradient-boosted tree model provided the best performance in predicting preterm birth, demonstrating the potential of using administrative data for early population-level risk stratification. This approach could enhance the identification of pregnancies needing closer monitoring and preventive care.