|
Activity Number:
|
382
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 11, 2004 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #301734 |
|
Title:
|
Odds Ratio Estimates for Case-control Studies Subject to Informatively Missing and Misclassified Exposure
|
|
Author(s):
|
Robert H. Lyles*+ and Andrew S. Allen and W.D. Dana Flanders and Lawrence L. Kupper and Deborah L. Christensen
|
|
Companies:
|
Emory University and Duke University and Emory University and University of North Carolina, Chapel Hill and Emory University
|
|
Address:
|
Dept. of Biostatistics, Atlanta, GA, 30322,
|
|
Keywords:
|
bias ; maximum likelihood ; reassessment ; validation
|
|
Abstract:
|
In case-control studies, it is common for a categorical exposure variable to be misclassified. It is also common for exposure status to be informatively missing for some individuals, in that the probability of missingness may be related to exposure. Procedures for addressing the bias due to misclassification via validation data have been extensively studied, and related methods have been proposed for dealing with informative missingness based on supplemental ("reassessment") sampling of some of those with missing data. We discuss study designs and analytic methods for dealing with both problems simultaneously in a 2X2 analysis. The motivating example comes from a case-control study of the association between low birthweight and the diagnosis of breast cancer later in life, where self-reported birthweight for some women is supplemented by accurate information from birth certificates.
|
