主 题: Joint analysis of longitudinal and competing risks survival data
报告人: prof. Gang Li (University of California at Los Angeles)
时 间: 2005-12-28 下午 4:00 - 5:00
地 点: 理科一号楼 1418
In this talk I will discuss joint modeling of repeated measurements
and competing risks failure time data. Our model uses latent random
variables and common covariates to link together the sub-models for the longitudinal measurements and competing risks failure time data. An EM-based algorithm is derived to obtain the parameter estimates. A profile likelihood method is proposed to estimate the standard errors. The significance of the joint modeling approach is at least three-fold. First of all, it enables one to make joint inference on the multiple outcomes of completely different nature, which is often required for analysis of clinical trials. Secondly, it provides a
useful method for analysis of longitudinal data in the presence of
informative dropout that produces non-ignorable missing data and
cannot be appropriately handled by the standard linear mixed effects
models alone. Finally, the joint model utilizes information from both outcomes and thus allows for more efficient analysis than a separate
analysis of the longitudinal or competing risk survival data. The
performance of our method is illustrated and compared with some
separate analyses using both simulated data and a real data from a
clinical trial for scleroderma lung disease.
