A0216
Title: Bayesian multivariate joint model of longitudinal, recurrent, and competing risk terminal events
Authors: Qi Qian - University of California, Los Angeles (United States) [presenting]
Abstract: The longitudinal decline of kidney function is intricately related to hospitalizations due to cardiovascular disease (CVD) and eventual "terminal" kidney failure and mortality in patients with chronic kidney disease (CKD). To better understand the mechanism and risk factors underlying these interdependent processes, as well as to tailor decision making to the needs of individual patients, a novel Bayesian joint model is developed for the interdependent outcomes of kidney function, recurrent cardiovascular events, and competing-risk terminal events (kidney failure and death). The proposed joint modelling not only allows the study of the risk factors associated with each outcome but also facilitates the dynamic updating of cumulative incidence probabilities of each competing risk for future subjects based on their profiles of previous longitudinal measurements and recurrent events. Efficient and flexible estimation and prediction procedures are proposed within a Bayesian framework using Markov Chain Monte Carlo, and the predictive ability is assessed via the dynamic area under the receiver operating characteristic (ROC) curves (AUC) and expected Brier score (BS). The efficacy of the proposed joint model and prediction procedure is shown via extensive simulations. The proposed methodology is also applied to data from the Chronic Renal Insufficiency Cohort (CRIC) study.
