The Medical Expenditure Panel Survey (MEPS) is a nationally representative survey that provides detailed information on healthcare. MEPS collects data on healthcare use, expenditures, sources of payment, and health insurance coverage as well as information on respondents' health status, employment status, access to healthcare, satisfaction with healthcare, and demographic characteristics. These data are sourced from large-scale surveys of individuals, families, their medical providers (doctors, hospitals, pharmacies, etc.), and employers across the United States. The MEPS data and methodologies are summarized in the table below.
The data are sourced from three separate components: the Household Component (HC); the Insurance Component (IC); and the Medical Provider Component (MPC).
The HC collects data from a subsample of families and individuals participating in the previous year's National Health Interview Survey (NHIS), which is conducted by the National Center for Health Statistics in communities it selects. The data for HC are collected via a contract with Westat, Inc.
The IC data are based on health insurance plans offered by a sample of private and public sector employers. The data for the IC are collected by the U.S. Census Bureau.
The MPC data are collected from a sample of providers who administered medical care to MEPS household-component respondents. The data for the MPC are collected via a contract with the Research Triangle Institute.
The data units are family or household units, called Reporting Units (RU). An RU is a group related by some family association within a physical structure known as a Dwelling Unit (DU) as defined by the NHIS. MEPS interviews each RU as a single unit.
The HC survey panel design is composed of five interview rounds over a two calendar year period. It provides data for analyzing person or event level changes in selected variables such as expenditures, health insurance coverage, and health status. Even though the HC was initiated in 1996, it is possible to analyze long term trends because the data collected are comparable to earlier surveys. Each annual MEPS HC sample size is about 15,000 households. Data can be analyzed at either the person or event level and must be weighted to produce national estimates. The set of households selected for each panel of the MEPS HC is a subsample of households participating in the previous year's National Health Interview Survey (NHIS) conducted by the National Center for Health Statistics. The NHIS sampling frame provides a nationally representative sample of the U.S. civilian noninstitutionalized population. The linkage of the MEPS to the previous year's NHIS provides additional data for longitudinal analytic purposes.
The IC surveys private and public sector employers, and the MPC samples providers, such as physicians, hospitals, and pharmacies. Because the IC and the MPC surveys are not comparable to the CE survey in scope or design, no further discussion of these components is warranted.
Notable Sample Exclusions
|The MEPS sample excludes RUs consisting of students living in student housing, or military personnel only, with no civilians present.|
The Consumer Expenditure Surveys (CE) consist of two separate nationwide surveys: The Interview Survey and the Diary Survey.1 MEPS is also a nationwide survey, and has three components: HC, IC, and MPC. (For more information, see "Data Source" in the table above). Since the HC is the only household-based survey in MEPS, it is the only source compared with the CE.
Chart 1 compares CE and MEPS estimated expenditures for Total healthcare; Prescription drugs; and Dental services. These components of healthcare are chosen for comparison because they are the most direct matches between the two estimates.
Charts 1 and 2 show a consistent gap between CE and MEPS, with the CE-to-MEPS ratio being roughly 0.43 over the time series.2 This can be explained by the differences between the CE and MEPS approaches. For example, while CE total healthcare expenditures include nonprescription drugs, MEPS excludes these. MEPS also excludes indirect payments such as the Medicaid Disproportionate Share and Medicare Direct Medical Education subsidies, and any charges associated with uncollected liability, bad debt, and charitable care (unless provided by a public clinic or hospital).
There are also differences in the way prescription drug data are collected between CE and MEPS. For example, MEPS data include diabetic supplies and equipment, such as syringes and test strips; however, CE counts those as medical supplies. In addition, while CE and MEPS both ask survey participants about their prescription drugs, MEPS asks their permission to collect more detailed information from their pharmacies, including type, dosage, and payment for each filled prescription, while the CE only asks for out of pocket expenditures. MEPS is designed to collect the full cost of drugs, while CE is designed to collect only the portion of that cost for which the consumer pays directly. Even though the dental services categories are similar across CE and MEPS, the ratios range from 0.36 to 0.40 as the samples and survey questions differ.
Although both CE and MEPS make estimates for the same Census regions, it mirrors the observations above that CE has consistently lower estimates, which is shown in Chart 2, with ratios between 0.59 and 0.88.
The CE and MEPS data comparison was developed utilizing comparable annual average and total healthcare expenditure categories with a supporting analysis conducted on a regional scale.
1 In BLS publications of CE data, most healthcare expenditures are taken from the Interview Survey. The exceptions are Nonprescription drugs, Nonprescription vitamins, and Topicals and dressings, which are taken from the Diary Survey.
2 That is, the average of CE-to-MEPS ratios from 2010-2018 for the three categories considered is 0.43. (See the CE data comparison Excel file in Supporting Documentation)
Last Modified Date: April 28, 2022