Comparison of National Factor-Based Models for Preeclampsia Screening

Area of Science:

• Obstetrics and Gynecology
• Maternal-Fetal Medicine
• Clinical Prediction Models

Background:

• Preeclampsia (PE) is a significant cause of maternal and neonatal morbidity.
• Accurate screening for PE is crucial for timely intervention and improved outcomes.
• Existing factor-based models, including those from ACOG, NICE, and SOGC, vary in their predictive performance.

Purpose of the Study:

• To compare the predictive accuracy of factor-based preeclampsia screening models developed by ACOG, NICE, and SOGC.
• To evaluate the performance of these models in predicting both preeclampsia and preterm preeclampsia.

Main Methods:

• Secondary analysis of maternal and birth data from 32 hospitals.
• Calculation of PE risk using ACOG, NICE, and SOGC factor-based models for 130,939 deliveries.
• Assessment of detection rates (DR), false positive rates (FPR), positive predictive values (PPV), and negative predictive values (NPV) using ROC curves.

Main Results:

• The ACOG model showed a 43.6% DR for PE and 50.3% for preterm PE with a 15.6% FPR.
• The ACOG model's PPV for PE (9.3%) and preterm PE (1.9%) were comparable or superior to NICE and SOGC models.
• Area under ROC curves indicated ACOG superiority over NICE for PE and preterm PE prediction, and over SOGC for preterm PE prediction.

Conclusions:

• The ACOG factor-based model, excluding race, is superior to NICE and SOGC models for predicting PE and preterm PE.
• Clinical factor-based models demonstrate moderate predictive capabilities for PE, with approximately 44% detection at a 16% FPR.
• Factor-based models may perform differently in parous versus nulliparous populations.