Modeling Repeated Labor Curves in Consecutive Pregnancies: A Latent Process Approach That Characterizes Autoregressive Dependence Across Pregnancies

• Activity Number: 475
• Type: Topic Contributed
• Date/Time: Wednesday, August 12, 2015 : 8:30 AM to 10:20 AM
• Sponsor: ENAR
• Author(s): Paul S. Albert* and Olive Buhule
• Companies: Eunice Kennedy Shriver National Institute of Child Health and Human Development and Eunice Kennedy Shriver National Institute of Child Health and Human Development
• Keywords: autoregressive, nested time scales, change points

Abstract

Measuring cervical dilation in the late stage of pregnancy is a commonly used technique for monitoring the progression of labor. Recent statistical methodology has been developed to address the analytical challenges for such data when only a single labor curve is observed on each woman (McLain and Albert, 2014, Biometrics). These challenges include conducting valid inference and prediction when there is not a time zero (i.e., when women enter the hospital at different stages of their labor). Motivated by the NICHD Consecutive Pregnancy Study (CPS), a unique cohort study that collected repeat labor data on over 50,000 women, we propose new methodology for analyzing labor curves across multiple pregnancies. We propose a latent modeling approach that characterizes serial correlation in the labor process across multiple pregnancies using an Ornstein-Uhlenbeck process. We employ Bayesian methodology (MCMC) for parameter estimation and prediction. The methodology was used in analyzing the CPS data, and in developing a predictor for labor progression that can be used in clinical practice.