Recent Advances in Joint Models for Longitudinal and Time-to-Event Data

• Activity Number: 379
• Type: Roundtables
• Date/Time: Tuesday, August 6, 2013 : 12:30 PM to 1:50 PM
• Sponsor: Biometrics Section
• Author(s): Dimitris Rizopoulos*+
• Companies: Erasmus MC
• Keywords: Survival Analysis ; Mixed Effects Models ; Risk Prediction ; Dynamic Predictions ; Personalized Medicine

Abstract

Joint models for longitudinal and survival data have become a useful and popular tool in current (bio)statistics research. These models are applicable in mainly two settings. First, when focus is in the survival outcome and we wish to account for the effect of an endogenous (i.e., internal) time-dependent covariate measured with error. Second, when focus is in the longitudinal outcome and we wish to correct for nonrandom dropout. The aim of this roundtable is to bring together (potentially new) researchers in this field and discuss uses and advances of joint models. Particular attention will be given in a relatively new application of the joint modeling framework for deriving dynamically updated individualized predictions for either the longitudinal or event time outcomes. This is motivated by current trends toward personalized medicine that aim to provide physicians with up-to-date accurate predictions that use all available patient data (including both baseline information and accumulated longitudinal responses).