PLEASE NOTE CHANGE OF LOCATION FOR WINTER QUARTER
M112 Alway Building, Medical Center
(next to the Dean's courtyard)
|DATE:||January 12, 2017|
|TIME:||1:30 - 2:50 pm|
|TITLE:||Joint Bayesian Semiparametric Regression Analysis of Recurrent Adverse Events and Survival|
|SPEAKER:||Juhee Lee, Assistant Professor
Department of Applied Mathematics and Statistics
University of California, Santa Cruz
We propose a Bayesian semiparametric joint regression model for a recurrent event process and survival time. Assuming independent latent subject frailties, marginal models for each subject are defined for the recurrent event process and survival distribution as functions of the subject’s frailty and covariates. Averaging over the frailty distribution yields a joint model. A robust Bayesian model is obtained by assuming a Dirichlet process for the frailty distribution. The model is applied to analyze an observational dataset from esophageal cancer patients treated with radiation therapy, including recurrent effusions of fluid to the heart or lungs. A simulation study is presented that compares posterior estimates under the joint model to simpler Bayesian models that ignore either the recurrent event process or survival time, and also to a frequentist joint model. The simulations show that the proposed Bayesian joint model does a good job of correcting for treatment assignment bias, is robust to a mild violation of the frailty model assumption, and has favorable estimation reliability and accuracy compared with each of the Bayesian sub-models and the frequentist joint model.
This is joint work with Peter F. Thall, Department of Biostatistics, M.D. Anderson, and Steven H. Lin, Department of Radiation Oncology, M.D. Anderson, Houston, TX
Brown, E. R. and Ibrahim, J. G. (2003). A Bayesian semiparametric joint hierarchical model for longitudinal and survival data. Biometrics 59, 2, 221–228.
Xu, G., Chiou, S. H., Chiug-Yu Huang, M.-C. W., and Yan, J. (2016). Joint scale-change models for recurrent events and failure time. Journal of the American Statistical Association , just-accepted.
Ghosh, S. K. and Ghosal, S. (2006). Semiparametric accelerated failure time models for censored data. Bayesian Statistics and its Applications 15, 213–229.