Modeling the Type and Timing of Consecutive Events: Application to Predicting Preterm Birth in Repeated Pregnancies

• Activity Number: 630
• Type: Topic Contributed
• Date/Time: Thursday, August 7, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Joanna Shih*+ and Paul Albert and Pauline Mendola and Katherine Laughon
• Companies: National Cancer Institute and Eunice Kennedy Shriver National Institute of Child Health and Human Development and National Institute of Child Health and Human Development and National Institute of Child Health and Human Development
• Keywords: adverse pregnancy outcome ; repeated pregnancies ; polychotomous random effects logistic model ; normal copula

Abstract

Predicting the occurrence and timing of adverse pregnancy events such as preterm birth is an important analytical challenge in obstetrical practice. Often adverse pregnancy outcomes are subject to competing events. In this article, we propose modeling the type and timing of adverse outcomes in repeated pregnancies. We formulate a joint model, where types of adverse outcomes across repeated pregnancies are modeled with a polychotomous logistic regression model with random effects, and gestational ages at delivery are modeled conditional on the types of adverse outcomes. The correlation between the types of adverse outcomes in repeated pregnancies is induced by type-specific random effects, and the correlation between gestational ages conditional on the adverse pregnancies is modeled by the semi-parametric normal copula function. We present an estimation method and develop the asymptotic theory for the proposed estimators. The proposed model and estimation procedure are applied to the NICHD Consecutive Pregnancy Study data and evaluated by simulations.