|
Activity Number:
|
437
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #304952 |
|
Title:
|
An Ordinal Logistic Regression Model with Misclassification of the Outcome Variable and a Covariate
|
|
Author(s):
|
Beverly A. Shirkey*+ and Stephen C. Waring and Jay H. Glasser and Wenyaw Chan
|
|
Companies:
|
The University of Texas Health Science Center at Houston and The University of Texas Health Science Center at Houston and The University of Texas Health Science Center at Houston and The University of Texas Health Science Center at Houston
|
|
Address:
|
1701 Upland Dr Apt 10, School of Public Health, Houston, TX, 77043,
|
|
Keywords:
|
Missclassification ; ordinal logistic regression
|
|
Abstract:
|
Disease status---mild, moderate or severe---is a commonly measured outcome in clinical trials of Alzheimer's disease (AD) treatments. An ordinal logistic regression can be used to model disease status, the effect of treatment and other predictive factors of the stage or status of the disease. However, misclassification of the disease status is a potential problem. In addition, predictor variables used in modeling study outcomes can be misclassified. In this paper, we will estimate the extent of misclassification of the ordinal outcome and a predictor variable in the context of an ordinal logistic regression model. The maximum likelihood method will be used to detect the amount of misclassification in simulated data sets, and then applied to an AD data set.
|
