Online Program Home
My Program

Abstract Details

Activity Number: 370
Type: Contributed
Date/Time: Tuesday, August 2, 2016 : 10:30 AM to 12:20 PM
Sponsor: ENAR
Abstract #318673
Title: Variable Selection for Multistate Models in the Presence of Missing Data
Author(s): Lauren J. Beesley* and Jeremy M. G. Taylor
Companies: University of Michigan and University of Michigan
Keywords: Missing Data ; Multistate models ; Bayesian Variable Selection

In head and neck cancer, it is well established that some patients can be cured of disease. We are interested in identifying factors that are associated with the probability of being cured and the times to cancer recurrence, death, and death after recurrence. We propose a multi-state model with a cure structure to model recurrence and death in head and neck cancer. We use Bayesian methods to fit the model and imputation to handle missing covariate data. Due to the large number of covariates and transitions considered, we explore various Bayesian variable selection approaches and model averaging to identify factors that are most strongly associated with the probability of cure and the transition rates to recurrence and death.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

Copyright © American Statistical Association