|
Activity Number:
|
458
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Health Policy Statistics
|
| Abstract - #304544 |
|
Title:
|
Using the Census Bureau's Surname List to Improve Estimates of Race/Ethnicity and Associated Disparities
|
|
Author(s):
|
Marc N. Elliott*+ and Peter A. Morrison and Allen Fremont and Daniel F. McCaffrey and Philip Pantoja and Nicole Lurie
|
|
Companies:
|
RAND Corporation and RAND Corporation and RAND Corporation and RAND Corporation and RAND Corporation and RAND Corporation
|
|
Address:
|
1776 Main Street, Santa Monica, CA, 90407,
|
|
Keywords:
|
Bayesian Inference ; health disparities ; race & ethnicity ; health insurance
|
|
Abstract:
|
Many health plans lack member racial/ethnic information for measuring disparities. To impute self-reported race/ethnicity, we use the U.S. Census Bureau's latest surname list to refine an existing Bayesian method for integrating surname and geocoded information. We validate this approach with self-reported racial/ethnic data from ~2 million enrollees of a national health plan. This approach has an overall concordance statistic (area under the Receiver Operating Curve) of 0.93. The largest improvements were in areas where prior performance was weakest (for Blacks and Asians). The new Census surname list accounts for about three-fourths of the variance explained in the new estimates. Unfortunately, imputing Native American and multiracial identities from surname and residence remains challenging.
|
