|
Activity Number:
|
407
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 9, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #306323 |
|
Title:
|
Random Effects Modeling Approaches for Estimating ROC Curves from Repeated Ordinal Tests without a Gold Standard
|
|
Author(s):
|
Paul S. Albert*+
|
|
Companies:
|
National Cancer Institute
|
|
Address:
|
6130 Executive Blvd., Room 8136, Bethesda, MD, 20892,
|
|
Keywords:
|
diagnostic accuracy ; latent class analysis ; mixture models ; ROC curves
|
|
Abstract:
|
Estimating diagnostic accuracy without a gold standard is an important problem in medical testing. Although there is much literature about this problem for the case of repeated binary tests, there is substantially less work for the case of ordinal tests. A noted exception is the work by Zhou et al. (2005), which proposed methodology for estimating ROC curves without a gold standard from multiple ordinal tests. An assumption in their work was that the test results are independent conditional on the true test result. We propose random effects modeling approaches that incorporate dependence between the ordinal tests. We show, through asymptotics and simulations, the importance of correctly accounting for the dependence between tests. We illustrate these modeling approaches by analyzing the uterine cancer pathology data analyzed in Zhou et al. (2005).
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2006 program |
