|
Activity Number:
|
391
|
|
Type:
|
Invited
|
|
Date/Time:
|
Thursday, August 12, 2004 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #300183 |
|
Title:
|
Development of Statistical Methods to Analyze Complex Health Surveys for Epidemiologic Studies: Some New Methods and Applications
|
|
Author(s):
|
Barry I. Graubard*+ and Thomas R. Fears and Steven B. Cohen and Lester R. Curtin
|
|
Companies:
|
National Cancer Institute and National Cancer Institute and Agency for Healthcare Research and Quality and Centers for Disease Control and Prevention
|
|
Address:
|
Biostatistics Branch, Rockville , MD, 20892,
|
|
Keywords:
|
complex survey ; epidemiology ; population attributable risk ; absolute risk
|
|
Abstract:
|
Over the past 20 years there has been great interest in using surveys to study health and behavioral risk factors for disease. To accomplish this research, surveys have been used to select population controls and cases for case-control studies, for constructing cohorts by following up survey participants, and to conduct cross-sectional studies. With this expanded use of surveys in medical research, statistical methods used in epidemiology have been adapted to account for the complexities of the sample weighting and sample selection of demographic and institutional surveys. We briefly trace the development of statistical methods for analyzing survey data for epidemiological research and discuss recent work of using surveys to estimate absolute risk and population attributable risk and their standard errors. Methods based on influence function theory will used to show how to easily derive Taylor linearized standard errors for these risk estimates. The methods will be illustrated using data from the third National Health and Nutritional Examination Survey and the National Health Interview Survey.
|
