|
Activity Number:
|
242
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 5, 2008 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Health Policy Statistics
|
| Abstract - #301623 |
|
Title:
|
Fitting a Semi-Markov Process Model to Data on Transitions Between Health States in the Presence of Left Censoring
|
|
Author(s):
|
Nathaniel Schenker*+ and Liming Cai and James Lubitz
|
|
Companies:
|
National Center for Health Statistics and National Center for Health Statistics and National Center for Health Statistics (retired)
|
|
Address:
|
National Center for Health Statistics, Hyattsville, MD, 20782,
|
|
Keywords:
|
Aging ; Disability ; Life Expectancy ; Missing Data ; Multi-State Life Table ; Stochastic EM Algorithm
|
|
Abstract:
|
This presentation describes a method for fitting a semi-Markov Process model to data on transitions between health states (e.g., good health, disability, death) in the older population, when the data are left-censored in the sense that the time of entrance into the initial state observed for each person in the study is unknown. An analog to the stochastic EM algorithm is used to address this missing-data problem. The approach is applied to data from the Medicare Current Beneficiary Survey and the Cardiovascular Health Study, and comparisons are made to other techniques for handling left-censoring as well as to estimates based on a traditional multi-state life table model.
|
