Preeclampsia poses significant risks to both mothers and babies. Many prediction models for preeclampsia have emerged in recent years. Using repeated measurements along with maternal factors has proven to be more effective in screening for preeclampsia than models that only consider maternal risk factors. However, traditional methods may introduce bias due to competitive events, and there is currently no preeclampsia prediction model in ALSPAC that considers competing-risk events.

Moreover, while numerous prediction models involve complex variables or models, cost-effectiveness must be taken into consideration. It is important to customize prediction models to local populations to effectively apply them. Relying on variables that are not available in local antenatal care can restrict their usefulness. It is suggested that localization should prioritize non-invasive indicators that are easily obtainable in clinical practice. An example of this is continuous blood pressure monitoring (CBP), which was recommended by the U.S. Preventive Services Task Force in 2017 as a PE screening method until a proven one is developed. However, it is not being utilized enough.

This project aims to develop a multivariate prediction model for preeclampsia by considering competing risk events and clinically accessible repeated measurements.
Before putting the prediction model into clinical practice, it will undergo essential validation across multiple datasets externally. The project’s benefits will be two-fold: first, shedding light on preeclampsia onset determinants in line with clinical practice, and second, improving the identification of women of high risk for preeclampsia.