Medical Expenditure Panel Survey (MEPS)

Medical Expenditure Panel Survey (MEPS, 1996 present) The Medical Expenditure Panel Survey is a nationally representative, longitudinal data system that captures detailed information on health care use, expenditures, insurance coverage, and health status for the civilian non institutionalized U.S. population. Initiated in 1996 and administered annually by the Agency for Healthcare Research and Quality (AHRQ), MEPS consists of three inter linked components the Household Component (HC), the Medical Provider Component (MPC), and the Insurance/Employer Component each surveyed over five interview rounds spanning two years. Purpose use cases MEPS was created to estimate and monitor national trends in health care utilization, costs, and insurance coverage. Researchers employ the data to evaluate policy impacts, model health care spending, study disparities in access and outcomes, assess prescription drug use, and develop cost effectiveness analyses. Key features / unique aspects - Longitudinal panel design: the same individuals are followed across five interview waves, enabling both cross sectional and cohort analyses. - Comprehensive variable set: demographics, socioeconomic status, health conditions, service events (inpatient, outpatient, emergency, prescription fills), detailed expenditure amounts, source of payment, and insurance plan characteristics. - Linkage to the National Health Interview Survey (NHIS), providing a probability based sample with oversampling of subpopulations. - Public use files and extensive documentation (codebooks, variable explorer) are freely downloadable from the AHRQ MEPS website, while restricted components (e.g., 1996 Nursing Home data) are available through Federal Statistical Research Data Centers. These attributes make MEPS a cornerstone resource for health services research, cost containment studies, and federal policy evaluation.

Data and Resources

Additional Info

Field Value
Last Updated February 2, 2026, 19:23 (UTC)
Created February 2, 2026, 19:23 (UTC)
contextual_insight_id 7b7578f9-cbdf-4db0-b8b4-65b0ea134123