A Joint Model Approach for Longitudinal Data with no Time-Zero and Time-To-Event with Competing Risks

Area of Science:

• Biostatistics
• Obstetrics
• Perinatal Care

Background:

• Joint modeling of longitudinal and time-to-event data is crucial in biostatistics.
• Existing models often require a defined time-zero, which is not always present in clinical data.
• Childbirth labor progression, measured by fetal station, lacks a clear start time.

Purpose of the Study:

• To develop a statistical framework for joint models when data lacks a meaningful time-zero.
• To model the relationship between longitudinal fetal station and time-to-delivery.
• To account for competing risks associated with different delivery types.

Main Methods:

• Developed a joint model for longitudinal and time-to-event data without a defined time-zero.
• Utilized shared random effects between survival and longitudinal processes.
• Employed a Bayesian approach for parameter estimation.

Main Results:

• The model successfully analyzed longitudinal station data and its relation to delivery time.
• Assessed the predictive ability of fetal station for delivery type and timing.
• Demonstrated the model's utility with real-world labor data.

Conclusions:

• The proposed joint model is effective for situations with no clear time-zero.
• Fetal station measurements can be valuable predictors of delivery outcomes.
• This framework can improve obstetric management and delivery planning.