Dynamic Prediction of Joint Longitudinal-Survival Models Using a Similarity-Based Approach

Longitudinal and time-to-event data are often analyzed in biomarker research to study the association between the longitudinal biomarker measurements and the event-time outcome, in which the longitudinal information contributes to the probability of the outcome of interest. An attractive nature of fitting a joint model on this type of data is that we can dynamically predict the survival probability as additional longitudinal information becomes available. We propose a new similarity-based method for the dynamic prediction of joint models where we consider training the model on only a targeted subset of the data to obtain an improved outcome prediction. Through comprehensive simulation study and an application to intensive care unit data, we demonstrate that the predictive performance of the dynamic prediction of joint models can be improved with our proposed similarity-based approach.