|
Activity Number:
|
130
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, August 4, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #301184 |
|
Title:
|
Hierarchical Dynamic Time-to-Event Models for Post-Treatment Preventive Care Data on Breast Cancer Survivors
|
|
Author(s):
|
Freda Cooner*+ and Sudipto Banerjee+
|
|
Companies:
|
U.S. Food and Drug Administration and The University of Minnesota
|
|
Address:
|
1620 E JEFFERSON ST, Rockville, MD, 20852, Biostatistics, Minneapolis, MN, ,
|
|
Keywords:
|
Cure rate models ; Dynamic survival models ; Hierarchical models ; Latent activation schemes ; Markov chain Monte Carlo ; Piecewise exponential distribution
|
|
Abstract:
|
This article considers modeling data in post-treatment preventive care settings, where cancer patients who have undergone the treatment discontinue seeking preventive care services. Clinicians and public health researchers are interested in explaining such behavioral patterns by modeling the time to receiving care while accounting for several patient and treatment attributes. Cure models are often preferred for data where a significant part of the population never experienced the endpoint. Building upon recent work on hierarchical cure model framework we propose modeling a sequence of latent events with a piecewise exponential distribution that remedies oversmoothing encountered in existing models with different latent distributions. We investigate simultaneous regression on the cure fraction and the latent event distribution and derive a flexible class of semiparametric cure models.
|
