|
Activity Number:
|
92
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Health Policy Statistics
|
| Abstract - #308361 |
|
Title:
|
Testing for Trends in a Two-State Markov Model with Applications in Smoking Cessation Studies
|
|
Author(s):
|
Charles Minard*+ and Wenyaw Chan and Carol J. Etzel and David Wetter
|
|
Companies:
|
The University of Texas M.D. Anderson Cancer Center and The University of Texas at Houston and The University of Texas M.D. Anderson Cancer Center and The University of Texas M.D. Anderson Cancer Center
|
|
Address:
|
1155 Pressler Blvd, Houston, TX, 77030,
|
|
Keywords:
|
binary ; longitudinal ; GEE ; Markov chain ; smoking cessation
|
|
Abstract:
|
Intervention trials may observe participants alternating between two states over time. The generalized estimating equations (GEE) method is commonly used to analyze binary, longitudinal data in the context of independent variables. A trend may be evaluated by including an interaction term in the GEE model. However, the sequence of observations may also follow a Markov chain with stationary transition probabilities. Assuming a log-transformed trend parameter, determinants of a trend may be evaluated by maximizing the likelihood function. New methodology is presented here to test for the presence and determinants of a trend in binary, longitudinal observations. Empirical studies are evaluated, and comparisons are made with respect to the GEE approach. Practical application is made to data available from a smoking cessation study.
|
