Imputation of Medicaid Beneficiaries' Expenditures for Physician Visits and Hospital Care in the Medical Expenditure Panel Survey

• Activity Number: 548
• Type: Topic Contributed
• Date/Time: Wednesday, August 6, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Health Policy Statistics Section
• Author(s): Jeffrey Gonzalez*+ and Lisa Mirel and Steven R. Machlin
• Companies: AHRQ and and AHRQ
• Keywords: Underreporting ; Data quality ; Measurement error ; Total survey error ; Imputation ; Data linkage

Abstract

The Medical Expenditure Panel Survey (MEPS) collects data on health care expenditures for the US civilian noninstitutionalized population. MEPS includes two key components: the household (HC) and the medical provider component (MPC). The HC was designed to enumerate all health care service events over a two-year period and their associated charges and payments for all sample persons via a series of five personal-visit interviews. However, the household is not always the best source of information on medical expenditures. This is particularly true for enrollees in the Medicaid program, where financial transactions occur between the medical provider/plan and the state Medicaid agency. The MPC was designed to supplement household reported data with health care provider data for a subset of the household reported events (e.g., physician visits and hospital care). In this paper, we describe the methodology used to impute missing expenditure information for Medicaid beneficiaries in MEPS. We then examine the extent to which Medicaid and out-of-pocket payments for reported health care service events of Medicaid beneficiaries are obtained via the household, medical providers, or imputation. Finally, we identify areas of future research.