Title: Conditional modeling of survival data with semi-competing risks
Authors: Tonghui Yu - Nanyang Technological University, Singapore (Singapore) [presenting]
Abstract: In recent years, various approaches have been developed for analysis of survival data in the presence of semi-competing risks arising in many clinical studies, where individuals are likely to experience two types of events: nonterminal and terminal. Most existing approaches use copula models to capture the dependence structure between both events. However, predicting overall survival (i.e., terminal event time) and accounting for the progression of the intermediate nonterminal event is often of interest in practice. It endeavors to aggregate the censored nonterminal event time with other risk factors for better modeling the overall survival. We propose a new semiparametric model for the terminal event time based on proportional hazards regression conditioning over the nonterminal event time. We develop a two-stage estimation procedure and establish asymptotic properties of the proposed estimator. Simulation studies and a real data example are used to evaluate the performance of the proposed model.