Department of Statistics Seminar
North Carolina State University

presents

Dr. Brad Carlin*

University of Minnesota

"Bayesian versus Likelihood Joint Modeling of Longitudinal and Event Time AIDS Clinical Trial Data"

ABSTRACT

Many clinical trials generate both longitudinal (repeated measurement) and survival (time to event) data. Many well-established methods exist for analyzing such data separately, but these may be inappropriate when the longitudinal variable is correlated with patient health status, hence the survival endpoint (as well as the possibility of study dropout). To remedy this, Henderson et al. (2002) propose a joint model for longitudinal and survival data, obtaining maximum likelihood estimates via the EM algorithm. Here the longitudinal and survival responses are assumed independent given a linking latent bivariate Gaussian process and available covariates. We develop a fully Bayesian version of this approach, implemented via Markov chain Monte Carlo (MCMC) methods, and investigate possible advantages that accrue over traditional likelihood methods. We then use the approach to jointly model the longitudinal and survival data from an AIDS clinical trial comparing two treatments, didanosine (ddI) and zalcitabine (ddC). Despite the complexity of the model, we find it to be relatively straightforward to implement and understand using the WinBUGS software.

*Friday lunch seminar: 12:00-1:00, Patterson Hall
Slides for Hierarchical Models for Spatio-Temporally Correlated Public Health Data
(hosted by Montserrat Fuentes)

Friday, October, 03, 2003

3:35 - 4:35 pm

206 Cox Hall

Refreshments will be served on the second floor of Dabney Hall (left of Room 222) at 3:00 pm.