A Multvariate Frailty Model for Disease Recurrences and Survival

• Activity Number: 336
• Type: Topic Contributed
• Date/Time: Tuesday, August 3, 2010 : 10:30 AM to 12:20 PM
• Sponsor: ENAR
• Author(s): Sijin Wen*+ and Xuelin Huang and Ralph F. Frankowski
• Companies: MD Anderson Cancer Center and MD Anderson Cancer Center and The University of Texas School of Public Health
• Keywords: disease recurrence ; survival ; frailty model ; Markov chain Monte Carlo ; hybrid Monte Carlo method

Abstract

Multiple disease recurrences may be observed for each patient and they increase the risk of death. A multivariate frailty model is established for joint-modeling of three correlated outcomes: local recurrence, distant recurrence, and overall survival. The dependence is modeled by conditioning on a shared frailty that is included in the hazard functions. Independent variables such as treatment and age can be included in the model as covariates. Markov chain Monte Carlo methods are utilized to estimate the posterior distributions of parameters. The algorithm used in present application is the hybrid Monte Carlo method, which updates parameters with evaluations of gradient of log posterior density. The performance of this approach is examined based on simulation studies. We apply the proposed methods to a study of patients with soft tissue sarcoma at MD Anderson Cancer Center.