Joint modelling of competing risks and current status data: an application to a spontaneous labour study

Area of Science:

• Obstetrics and Gynecology
• Perinatal Medicine
• Biostatistics

Background:

• The second stage of labor, from full dilation to delivery, has recommended time limits to mitigate maternal and neonatal risks.
• Current recommendations may not fully account for varying individual risks of spontaneous vaginal delivery (SVD) and morbidities.
• There is a need for quantitative evidence comparing risks across the duration of the second stage of labor.

Purpose of the Study:

• To quantitatively compare risks for SVD, Caesarean delivery (CD), operative vaginal delivery (OVD), and maternal/neonatal morbidities.
• To develop a statistical model that analyzes the joint distribution of delivery mode and morbidity onset.
• To provide evidence-based insights into the duration of the second stage of labor.

Main Methods:

• Introduction of a joint statistical model combining competing risks data for delivery time and current status data for morbidity.
• The model accounts for individual-specific frailty, assuming dependence between delivery mode and morbidity.
• Numerical studies included simulations and a real-data analysis of nearly 12,000 spontaneous labors.

Main Results:

• The study demonstrates the necessity of incorporating maternal characteristics like age and body mass index.
• These characteristics are crucial for assessing probabilities of SVD, CD, OVD, and morbidity onset.
• Risk assessment varies significantly based on individual patient profiles.

Conclusions:

• Maternal characteristics significantly influence the risks associated with the second stage of labor.
• A nuanced approach considering individual factors is essential for optimizing delivery mode and managing morbidities.
• The developed joint model provides a framework for more personalized risk assessment in obstetrics.