|
Activity Number:
|
278
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Health Policy Statistics
|
| Abstract - #309111 |
|
Title:
|
Multistate Life Table Distributions for Highly Refined Subpopulations from Cross-Sectional Data: A Bayesian Alternative to Sullivan's Method
|
|
Author(s):
|
Scott Lynch*+
|
|
Companies:
|
Princeton University
|
|
Address:
|
Department of Sociology, Princeton, NJ, 08544,
|
|
Keywords:
|
Gibbs sampling ; multistate life tables ; demography ; Sullivan's method
|
|
Abstract:
|
Sullivan's method is the most commonly used method for estimating healthy life expectancy (HLE). Although the data requirements for the method are minimal, the method is limited in its ability to produce HLE estimates for subpopulations due to limited disaggregation of data in cross-sectional mortality files and small cell sizes in aggregated micro survey health data. We develop a method that adapts Sullivan's approach to allow the inclusion of covariates in estimation of HLE and to generate interval estimates of HLE for any desired subpopulation that can be specified in the health data. The method involves (1) using Gibbs sampling to sample parameters from a hazard model; (2) using ecological inference to produce transition probability matrices from the Gibbs samples; and (3) applying standard multistate calculations to the transition probability matrices.
|
