Data on Health and Well-being of American Indians, Alaska Natives, and Other Native Americans, Data Catalog

Medical Expenditure Panel Survey (MEPS)

[ Main Page of Report | Contents of Report ]

Sponsor: U.S. Department of Health and Human Services (DHHS)/Agency for Health Care Research and Quality (AHRQ) and National Center for Health Statistics (NCHS)
Description: The Medical Expenditure Panel Survey (MEPS) is conducted to provide nationally representative estimates of health care use, health care expenditures, sources of payment, health insurance coverage and health status for the U.S. civilian noninstitutionalized population. MEPS data can be used to estimate the impact of changes in sources of payment and insurance coverage on different economic groups or special populations of interest, such as the poor, elderly, families, veterans, the uninsured, and racial and ethnic minorities.

MEPS is comprised of three component surveys: the Household Component (HC), the Medical Provider Component (MPC), and the Insurance Component (IC). The HC is the core survey, and it forms the basis of the MPC sample. Together these surveys yield comprehensive data that provide national estimates of the level and distribution of health care use and expenditures, support health services research, and can be used to assess health care policy implications.

Relevant Policy Issues: Measurement of Health Status, Disease-specific Measurements, Health Disparities, Factors Contributing to Measured Health Disparities, Measures of Well-being for Families/Households, Factors Contributing to Well-being Disparities of Families, Measures of Well-being for Children, Measures of Well-being for Elders, and Factors Contributing to Well-being Disparities of Elders.
Data Type(s): Survey
Unit of Analysis: Data can be analyzed at the person level, the event level, the family level, or the health insurance unit level.
Identification of AI/AN/NA: Race categories include (2002 and beyond):
  • White alone
  • Black alone
  • American Indian or Alaska Native (AI/AN) alone
  • Asian alone
  • Native Hawaiian or Pacific Islander (NH/PI) alone
  • Multiple Race
AI/AN/NA Population in Data Set: Based on 2004 MEPS data, out of 34,403 total records, 293 are coded AI/AN alone and 150 are coded NH/PI alone. Multiple years of data can be pooled to increase the sample size. Aggregation requires at least 100 unweighted cases to support national estimates.
Geographic Scope: The geographic scope of the MEPS is national. The MEPS is designed to support national and regional estimates. Due to small sample sizes, state estimates of the AI/AN/NA population are not possible.
Date or Frequency: MEPS was initiated in 1996 and is a continuous ongoing survey. MEPS predecessor surveys were conducted in 1987 and 1977.
Data Collection Methodology: The MEPS HC uses an overlapping panel design in which data are collected through a series of five rounds of interviews over a 2-1/2 year period using computer-assisted personal interviewing technology to collect information on all household members. This series of data collection rounds is launched each subsequent year on a new sample of households to provide overlapping panels of survey data that, when combined, will provide continuous and current estimates of health care expenditures. In 2000, an annual fielding of a self-administered questionnaire (SAQ) was introduced into the process as well.

For the MPC, a sample of medical providers are contacted by telephone to provide information that the household respondents cannot accurately provide (after obtaining permission from the HC respondents). The IC is an annual panel survey that collects data on health insurance plans obtained through employers, unions, and other sources of private health insurance. Data are collected by the Census Bureau from the sampled organizations through a prescreening telephone interview, a mailed questionnaire, and a telephone follow-up for non-respondents.

Participation: Optional, with incentives ($25.00) for each of the five rounds of interviews completed and $5.00 for each self-administered or child questionnaire completed.
Response Rate: The full-year HC response rate has generally ranged from about 65 to 71 percent. Conditional response rates for Rounds 2-5 are always over 90 percent.
Sampling Methodology: The MEPS-HC uses the National Health Interview Survey (NHIS) conducted by the National Center for Health Statistics as its sampling frame. The NHIS sampling frame provides a nationally representative sample of the U.S. civilian non-institutionalized population and reflects an over-sampling of Blacks and Hispanics. In certain years MEPS over samples additional policy relevant sub-groups. This design allows linkage back to the previous year’s NHIS for analytic purposes.
Analysis: As with most surveys, participating individuals represent only a fraction of the overall population the survey is intended to reflect. In order to calculate estimates representing the overall population, responses from surveyed individuals must be weighted by the proportion of the population they represent. In addition, adjustments must be made to account for non-response. Each MEPS file contains appropriate weight variables that can be applied to the data to generate national estimates of the civilian non-institutionalized population. A detailed description of the weighting process and how weights are applied to estimates can be found in the weighting and estimation section of the online workbook under the workshop and events section of the MEPS web site.

Because MEPS relies on a complex sampling design rather than simple random sampling, it is important to take into account reductions in the accuracy of calculated estimates (size of standard errors). Statistical software programs such as SAS (version 8.2 or higher), SUDAAN, STATA, and SPSS (version 12.0 or higher), are available to accommodate the complex design and calculate robust standard errors. A full description of how to compute standard errors for MEPS can be found on the MEPS web site (

Strengths: MEPS facilitates research on relationships between individual characteristics and health care utilization. It provides national and regional estimates of health care use and expenditures. It contains two full years’ worth of data for each panel. Each panel also can be linked to the previous year’s NHIS for a third point in time. This design facilitates analysis of change over time. MEPS data can be used to estimate the impact of changes in sources of payment and insurance coverage on different populations of interest such as the AI/AN/NA population, or to evaluate the impact of an intervention or treatment on health status over time. It is the only source of actual sources of payments and amounts paid including out-of-pocket expenditures for health care visits.
Limitations: The MEPS was not designed to produce state-level estimates. While aggregate estimates for a selected number of large states may be possible, small sample sizes preclude making such estimates for the AI/AN/NA populations. Even after pooling several years of MEPS data, sample size limitations and confidentiality restrictions make MEPS data unsuitable for certain types of analysis. For example, the MEPS data do not support research on rare conditions. Moreover, information on conditions is household reported and not verified by clinical records. All MEPS data are reported by one designated household respondent. Reporting detailed information on other household members can sometimes be problematic.
Access Requirements and Use Restrictions: MEPS HC data releases, including documentation and codebooks, are available free to the public on the Internet (via the MEPS web site). MEPS IC data are published in tabular format on a yearly schedule. MEPS data (HC and IC data) are also available via MEPSnet, an on-line, interactive, statistical tool developed to give users the ability to analyze MEPS data in real-time. Access to the MEPS IC full data set is only available in a Census Bureau Research Data Center.

Many of the MEPS databases include considerably more data than can be made available to the general public because of the constraints of confidentiality guidelines. In order to facilitate the use of such data, while maintaining the confidentiality, AHRQ developed a Data Center (a physical space at AHRQ in Rockville, Maryland) where researchers with approved projects can access data files not available for public dissemination. See the MEPS web site for details.

Contact Information: The MEPS website is

By e-mail:
By phone/fax: MEPS Information Coordinator (301) 427-1656
CFACT General Information: (301) 427-1406
CFACT Fax: (301) 427-1276
By mail: Project Director
Center for Financing, Access and Cost Trends
Medical Expenditure Panel Survey
Agency for Healthcare Research and Quality
540 Gaither Road
Rockville, MD 20850
Telephone: (301) 427-1656

Reports of Interest For copies of data products and reports, see the MEPS web site. Selected MEPS data products are available from:
AHRQ Publications Clearinghouse
P.O. Box 8547
Silver Spring, MD 20907
(1-800) 358-9295
(703) 437-2078 outside the U.S.
TDD for the hearing impaired, toll free: (888) 586-6340

Where to?

Top of Page

Main Page of Report | Contents of Report

Home Pages:
Human Services Policy (HSP)
Assistant Secretary for Planning and Evaluation (ASPE)
U.S. Department of Health and Human Services (HHS)