Correlates of Nonresponse in the 2010 and 2012 Medical Expenditure Panel Survey

• Activity Number: 472 - Imputation and Nonresponse Bias
• Type: Contributed
• Date/Time: Wednesday, August 2, 2017 : 8:30 AM to 10:20 AM
• Sponsor: Survey Research Methods Section
• Author(s): Frances Chevarley* and William Mosher
• Companies: Agency for Healthcare Research and Quality and Johns Hopkins University (formerly NCHS)
• Keywords: Medical Expenditure Panel Survey (MEPS) ; National Health Interview Survey (NHIS) ; nonresponse

Abstract

This paper analyzes estimates from the Household Component of the Medical Expenditure Panel Survey (MEPS) matched with the National Health Interview Survey (NHIS) and uses practical tools to inform MEPS nonresponse estimates. MEPS is a nationally representative panel survey studying health care use, access, expenditures, source of payment, insurance coverage, and quality of care sponsored by the Agency for Healthcare Research and Quality. Each year a new panel begins and each panel has 5 rounds of data collection over 2 ½ years. The MEPS sample is a subsample of the previous year's NHIS sample; NHIS is sponsored by the National Center for Health Statistics, Centers for Disease Control and Prevention (CDC). The goal of this paper is to explain MEPS nonresponse rates for 2010 to 2012. We use demographic and other personal characteristics from NHIS and MEPS to predict MEPS nonresponse rates. Data used are from the 2009 and 2011 NHIS matched with the 2010 and 2012 MEPS files along with additional paradata.