A Joint Modeling Approach for Right-Censored Multivariate Longitudinal Data

• Activity Number: 186
• Type: Contributed
• Date/Time: Monday, August 4, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Miran Jaffa*+ and Ayad A. Jaffa and Mulugeta Gebregziabher
• Companies: American University of Beirut and American University of Beirut and Medical University of South Carolina
• Keywords: High dimensional data ; Informative right censoring ; Kidney failure ; Multivariate longitudinal outcomes ; Random effect ; Slope estimation

Abstract

Longitudinal Data Analysis is complicated by informative right censoring that Here, we propose a novel likelihood based approach wherein we jointly model the censoring process along with the slopes of the multivariate outcomes in the same likelihood function. We used pseudo likelihood function to generate parameter estimates for the population slopes and Empirical Bayes estimates for the individual slopes. The performance of this approach was assessed using simulated and real data and compared to an existing bivariate approach. Our simulation study results suggested that there were 40% and 25% reductions in bias and mean squared errors (MSEs)respectively associated with the joint model compared to the pairwise bivariate model. Furthermore, our simulation study demonstrated that high dimensional multivariate outcomes (10 outcomes) can be jointly modeled in the likelihood function and corresponding slopes estimates were successfully acquired with mimumum bias and MSEs. The proposed multivariate approach was applied to jointly model longitudinal measures of markers of kidney function in a cohort of renal transplant patients followed from kidney transplant to kidney failure.