Joint modeling time-to-event and longitudinal data with skewness and covariate measurement errors: a robust Bayesian approach
Abstract In the development of joint models for timeto-event and longitudinal data, observed variables are typically assumed to have a normal distribution in both Bayesian and classic models. However, a serious departure from normality may cause lack of robustness and subsequently lead to invalid inference and unreasonable estimates. Covariates are usually introduced in the models to partially explain intersubject variations, but some covariates may be often measured with substantial errors. In many practical situations, the time-to-event of interest is likely associated with the longitudinal response and covariate processes. In this article, our objectives are to investigate joint models in general forms that allow longitudinal variables to have skew distributions in the presence of covariate measurement error and time-to-event variable to have a nonparametric prior distribution, which follows a Dirichlet process prior. A Bayesian joint modeling approach is offered to estimate parameters simultaneously. An example using marker data from an AIDS study is given to illustrate the methodology by jointly modeling viral response dynamics, CD4 covariate process with measurement error and time to decrease in CD4/CD8 ratio, and to compare potential models with various scenarios and different distribution specifications. The results from this study suggest that the time-varying CD4 covariate has a significantly positive effect on the first-phase viral decay rate; the results also indicate that the time to CD4/CD8 decrease is not highly associated with either the two viral decay rates or the CD4 changing rates over time, which is different from what was anticipated. These findings may be important for HIV/AIDS studies in providing quantitative guidance to better understand the virologic responses to antiretroviral treatment.
Bayesian analysis Covariate measurement errors Dirichlet process Mixed-effects joint models Skew distributions Time-to-event
Yangxin Huang
Department of Epidemiology & Biostatistics,College of Public Health,University of South Florida,Tampa,FL 33612,USA
国际会议
Second Joint Biostatistics Symposium(第二届生物统计国际研讨会2012)
北京
英文
281-306
2012-07-08（万方平台首次上网日期，不代表论文的发表时间）