|
Activity Number:
|
380
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #304116 |
|
Title:
|
Informative Retesting
|
|
Author(s):
|
Christopher R. Bilder*+ and Joshua M. Tebbs and Peng Chen
|
|
Companies:
|
University of Nebraska-Lincoln and University of South Carolina and Takeda Pharmaceuticals
|
|
Address:
|
Department of Statistics, Lincoln, NE, 68583,
|
|
Keywords:
|
binary response ; chlamydia ; gonorrhea ; group testing ; pooled testing ; retesting
|
|
Abstract:
|
Group testing identification algorithms traditionally have been applied to populations where each individual is considered to have an equal probability of being diseased. However, in most real problems, covariate information is available for each individual and can be used to report covariate-adjusted individual probability estimates. We propose the first methodology that directly incorporates this additional covariate information into the identification process. These informative retesting procedures are shown to reduce the number of tests needed to decode a positive pool when compared to the corresponding non-informative procedures. We illustrate this new methodology by using it for the Infertility Prevention Program in its identification of individuals with Chlamydia and gonorrhea.
|
