Recorded presentation

Jennifer D. Parker, PhD, Office of Analysis and Epidemiology, Centers for Disease Control and Prevention, National Center for Health Statistics, 3311 Toledo Road, Hyattsville, MD 20782, 301 458-4419, jdp3@cdc.gov

Early in 2003, the National Center for Health Statistics (NCHS) released a set of population estimate files for 2000 and 2001, based on the Census Modified Race Summary File, where the multiple race counts on the Census file were distributed amongst the corresponding single-race counts. The objective of this talk is to describe the initial models used to distribute the multiple race counts, briefly discuss some statistical and policy issues associated with the original models, and present preliminary findings from ongoing model development that may be used for post-censal population counts. The National Health Interview Survey (NHIS) has allowed multiple race responses for many years. In addition, the NHIS asks multiple race respondents to identify one of their responses as a primary race. Using the NHIS for 1997-2000, for each multiple race group, the individual primary race responses were used as the outcome variable in the original categorical regression models; individual-level and county-level contextual variables were included as covariates. Models and predictions differed markedly betweeen multiple-race groups. Recent revisions to these models include the use additional years of the NHIS, additional contextual variables recently available from the Census, relax the constraint of model-similarity across multiple-race groups, and examine the inclusion of time as a predictor for post-censal estimates

Learning Objectives:

• Learn how race-bridging models were developed to obtain single race population estimates, the considerations underlying orignial and revised models, and see effect of different models for race-bridging on some race-specific health statistics

Related Web page: www.cdc.gov/nchs/about/major/dvs/popbridge/popbridge.htm

Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.

Recorded presentation