7 Medical Expenditure Panel Survey
7.1 Introduction
Chapter 7 covers the Medical Expenditure Panel Survey (MEPS). The MEPS comprises a series of surveys from families, individuals, employers and medical providers that have been collected by the Agency for Healthcare Research and Quality (AHRQ) since 1996. The purpose of the MEPS is to estimate and monitor national trends in health care use, medical costs and health insurance coverage.1 This chapter includes details on: how data are collected; how data are made publicly available as machine-actionable data files; what variables must be included to address design features of the complex sample; the strengths and limitations of the survey; and practical tips for conducting statistical analysis; and how to answer research questions using a case study. The practical tips provided for analysis of MEPS data are based on the author’s previous experiences analyzing MEPS data from 2001-2017 to answer questions related to associations between predisposing and enabling factors that contribute to morbidity, mortality and health services use. The MEPS case study will explore whether adults who perceive their physician provided quality patient-provider communication (PPC) are more or less likely to receive an annual flu vaccine in comparison those who did not receive quality PPC. The bulk of the chapter will comprise of section 7.6: MEPS Case Study in order to give investigators hands-on practice downloading and cleaning large databases and conducting basic categorical data analysis using PROC SURVEYFREQ and PROC SURVEYLOGISTIC. The syntax provided was created for use with SAS 9.4.
7.2 Data Collection
The MEPS uses a longitudinal panel design collecting data from individuals and families from five rounds of interviews over a two-year period.2 Data can be compiled for cross-sectional analyses or longitudinal, retrospective cohort designs. A subsample of household participants who filled out the previous year’s National Health Interview Survey (NHIS) are selected for each panel. Oversampling of subgroups aligns with the NHIS nationally representative sample design.2 For example, non-Hispanic Blacks and Hispanics were oversampled until 2016. Non-Hispanic Asians were oversampled from 2006-2015. In 2016, the new NHIS design was modified to increase precision of statewide estimates and these changes were reflected in the MEPS beginning in 2017.3 Data for 2018 onward were not available as of this writing. Further details of the MEPS sampling design and data collection methods are reported elsewhere.4
7.3 Data Files
The MEPS is comprised of two main components: 1) household component and 2) medical provider component. Each component includes multiple subsections. The MEPS insurance component also collects health insurance plan offerings from employers in the US on an annual basis.5 Each annual household consolidated and medical condition file represents data from two panels. Each longitudinal file represents data from