|
Activity Number:
|
374
|
|
Type:
|
Invited
|
|
Date/Time:
|
Wednesday, August 6, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Social Statistics Section
|
| Abstract - #300252 |
|
Title:
|
Application of the Peters-Belson Method for Estimating Disparities
|
|
Author(s):
|
Barry I. Graubard*+ and Sowmya R. Rao and Joseph Gastwirth
|
|
Companies:
|
National Cancer Institute and MGH Biostatistics Center/The Institute for Health Policy and The George Washington University
|
|
Address:
|
6120 Executive Blvd, Room 8024, Bethesda , MD, 20892,
|
|
Keywords:
|
health disparities ; regression ; complex samples ; cancer screening ; cancer incidence
|
|
Abstract:
|
The Peters-Belson (PB) method is used to estimate wage discrimination and disparities in cancer screening between an advantaged group (e.g., Caucasians) and a disadvantaged group (e.g., Hispanics). The PB method uses regression to estimate the proportion of the observed disparity that is not explained by the covariates used in the regression. This method first fits a regression model with individual-level covariates to the majority group and then uses the fitted model to estimate the expected values of the minority-group members had they been members of the majority group. We extend the PB method to using a sample with measured covariates from an external population. We illustrate the methods with cancer screening and cancer incidence data from the National Health Interview Survey and the National Health and Nutrition Examination Survey that have weighted multistage cluster samples.
|
