|
Activity Number:
|
248
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
IMS
|
| Abstract - #309989 |
|
Title:
|
Robust Methods for Estimation of State Occupancy Probabilities with Interval-Censored Multistate Data
|
|
Author(s):
|
David Tolusso*+ and Richard J. Cook
|
|
Companies:
|
University of Waterloo and University of Waterloo
|
|
Address:
|
Statistics and Actuarial Science, Waterloo, ON, N2J 4H2, Canada
|
|
Keywords:
|
|
|
Abstract:
|
Chronic disease processes are often conveniently modeled using multistate models. In many settings however, patients are only observed at periodic assessment times in which case transition times are interval censored. In this talk we consider robust methods for estimating state occupancy probabilities based on interval-censored multistate data. Methods are described based on marginal failure time models and Markov models fitted via local likelihood. Data from a study of patients with psoriatic arthritis are used for illustration.
|
