Individualized assessment of preterm birth risk using two modified prediction models

Area of Science:

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

Background:

• Preterm birth remains a leading cause of neonatal morbidity and mortality.
• Accurate prediction of preterm delivery is crucial for timely intervention and improved outcomes.
• Existing models for preterm birth risk assessment require refinement for better individualization.

Purpose of the Study:

• To develop and validate two novel prediction models for individualized risk assessment of preterm delivery within 48 hours.
• To assess the risk of preterm delivery before 32 weeks of gestation.
• To evaluate the validity of modified and previously published prediction models.

Main Methods:

• Analysis of data from 617 women with preterm labor between 22 and 32 weeks gestation.
• Multivariate analyses to identify key predictive variables for delivery within 48 hours and before 32 weeks.
• Development and internal validation of two modified prediction models incorporating selected variables.

Main Results:

• Two validated models were developed, utilizing four and six variables respectively, including sonographic cervical length, fetal fibronectin, and C-reactive protein.
• The models demonstrated excellent calibration, with high concordance-indices of 0.80 for the 48-hour model and 0.85 for the 32-week model.
• Internal validation confirmed the high accuracy and reliability of the developed prediction models.

Conclusions:

• Two modified prediction models for assessing preterm birth risk have been successfully constructed and validated.
• These models enable more accurate, individualized risk assessment for preterm birth compared to single-factor assessments.
• An online risk-calculator is available for clinical use, facilitating practical application of the models.