A1456
Title: Time-to-event analysis with unknown time origins via longitudinal biomarker registration
Authors: Wensheng Guo - University of Pennsylvania (United States) [presenting]
Abstract: In observational studies, the time origin of interest for time-to-event analysis is often unknown, such as the time of disease onset. Using the study entry as time zero often leads to misleading results. Existing approaches to estimating the time origins are commonly built on extrapolating a parametric longitudinal model, which relies on rigid assumptions that can lead to biased inferences. A flexible semiparametric curve registration model is introduced. It assumes the longitudinal trajectories follow a flexible common shape function with a person-specific disease progression pattern characterized by a random curve registration function, which is further used to model the unknown time origin as a random start time. This random time is used as a link to jointly model the longitudinal and survival data where the unknown time origins are integrated into the joint likelihood function, which facilitates unbiased and consistent estimation. Simulation studies and two real data applications demonstrate the effectiveness of this new approach.
