|
Activity Number:
|
249
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
WNAR
|
| Abstract - #302839 |
|
Title:
|
Joint Models for Longitudinal and Survival Data
|
|
Author(s):
|
Jeremy Taylor*+
|
|
Companies:
|
University of Michigan
|
|
Address:
|
Department of Biostatistics, Ann Arbor, MI, 48109,
|
|
Keywords:
|
Joint models ; prostate cancer ; prediction ; proportional hazards
|
|
Abstract:
|
In many medical studies there are two types of response variables, a longitudinal variable and an event time, both of which may measure the progression of a disease or the response to an intervention. The longitudinal variable is typically modeled using a random effect model, and the event time is typically modeled as a time dependent proportional hazards model. In this talk I will focus on using a joint model to assist with individual prediction of future event times for censored subjects. The model and methods are developed in the context of a prostate cancer application where the longitudinal variable is PSA and the event time is recurrence of the cancer following treatment with radiation therapy. The model is fit using MCMC techniques. An algorithm is developed to give predictions for subjects who were not part of the original data from which the model was developed.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2009 program |
