News & Events


Speaker:Prof. Chia-Hui Huang (Department of Statistics, NTPU)

Topic:Estimating treatment effects for semicompeting risks data with treatment switching
Speaker:Speaker:Prof. Chia-Hui Huang (Department of Statistics, NTPU)
Date Time:Fri. Sep 21,2018, 10:40 AM -11:30 AM
Place:4F-427, Assembly Building I


In clinical trials, it often arises that some patients randomized to control group switch to the alternative treatment after tumor progression. In such a study design, it is very likely that the patient's initiating the alternative treatment is related to his/her prognosis and may change the hazards of death. In this work, we propose a class of semicompeting risks models to estimating the effects of the combination treatments. The framework consists of a copula model for predicting the survival times of the two events at the first stage, and a conditional hazard model for predicting the death after progression with treatment switching. The maximum likelihood inference procedure is developed for the proposed models. Simulation studies and application to a clinical study are conducted to demonstrate the utility of the proposed methods. This is a joint work with Dr. Yi-Hau Chen. 
Last modification time:2018-09-21 AM 11:49

  • recruiting animation-EN
  • NCTU
cron web_use_log