The Medical Expenditure Panel Survey (MEPS) is a vital national resource for assessing the impacts of the 2010 Patient Protection and Affordable Care Act on mental health care expenditures. However, the MEPS is known to under-represent mental health expenditures, in general, and evidence suggests the magnitude of bias could be larger for persons with high expenditures for mental health care. This study addresses long- standing questions surrounding the accuracy of MEPS expenditure estimates for persons with mental illness. It also will provide information about biases in MEPS expenditure data for other hard-to-reach populations, such as low-income persons with HIV and persons with substance use disorders. The study's specific aims are: 1) to quantify, for subgroups defined by health condition category, insurance coverage, and age, the bias in MEPS estimates of annual expenditures for emergency department visits and inpatient care;and 2) to demonstrate an estimation method that will achieve a more representative health care expenditure distribution for subgroups of emergency department patients and inpatients that are underrepresented in MEPS. The State Inpatient Databases (SID) and State Emergency Department Databases (SEDD) contain a complete census of all inpatient stays and emergency department visits at community hospitals. Data from the SID and SEDD and data from the MEPS will be used to construct annual cross-sectional samples of persons who had at least one qualifying emergency department visit or inpatient admission during a calendar year. MEPS estimates of expenditures for emergency department and inpatient hospital services use and expenditures in the MEPS will then be compared to these same quantities developed from the SID/SEDD. We will examine expenditure differences especially in relation to mental disorder diagnoses, substance use disorder diagnoses, HIV, Hepatitis C, insurance coverage, and age group, using moment condition constraints developed from SID/SEDD. The study's results will provide insight into the importance of MEPS expenditure biases for subpopulations that may be hard-to-reach in a household survey, and will provide preliminary evidence for a future study of how the PPACA impacts health expenditures for persons with mental illness.

Public Health Relevance

National household survey data on expenditures for health care services are a vital national resource for policy analyses and for microsimulation studies of the impacts of national health care reforms. This study will provide new information about the accuracy of expenditures data in the Medical Expenditure Panel Survey, and will examine an innovative methodology for correcting biases in health expenditures data.

National Institute of Health (NIH)
National Institute of Mental Health (NIMH)
Exploratory/Developmental Grants (R21)
Project #
Application #
Study Section
Special Emphasis Panel (ZRG1-HDM-C (51))
Program Officer
Rupp, Agnes
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Maryland Baltimore
Schools of Medicine
United States
Zip Code
Slade, Eric P; Goldman, Howard H; Dixon, Lisa B et al. (2015) Assessing the Representativeness of Medical Expenditure Panel Survey Inpatient Utilization Data for Individuals With Psychiatric and Nonpsychiatric Conditions. Med Care Res Rev 72:736-55
Slade, Eric P; Goldman, Howard H (2015) The dynamics of psychiatric bed use in general hospitals. Adm Policy Ment Health 42:139-46