Re-Weighting the National Health and Nutrition Examination Survey Linked to Medicare Administrative Records

• Activity Number: 287
• Type: Topic Contributed
• Date/Time: Tuesday, August 2, 2011 : 8:30 AM to 10:20 AM
• Sponsor: Section on Government Statistics
• Author(s): Lisa Beth Mirel*+ and Jennifer D. Parker
• Companies: Centers for Disease Control and Prevention/NCHS and Centers for Disease Control and Prevention
• Keywords: National Health and Nutrition Examination Survey ; NHANES ; Centers for Medicare and Medicaid Services ; CMS ; Statistical weights ; Non-response adjustment

Abstract

The National Health and Nutrition Examination Survey (NHANES) was recently linked to Centers for Medicare and Medicaid Services (CMS) Medicare data. Non-response bias from refusal or insufficient information for linkage needs to be considered when analyzing these data. We compared statistical weight adjustments for non-response, based on the 1999-2004 NHANES respondents who could be linked and had data from the 2007 Medicare Denominator file. Different weights were calculated using SUDAAN's WTADJUST procedure. Summary statistics for the different weighting methods are presented in addition to the estimated percent of beneficiaries enrolled in Medicare Advantage, using the different weights. The weights and estimates did not vary among weighting approaches. However the comparison of the different weighting approaches may differ by analysis variables, and respondent characteristics.