MEPS Home Medical Expenditure Panel Survey
Font Size:
Contact MEPS FAQ Site Map  

Update Notes

MEPS HC-001: 1996 Panel Round 1 Population Characteristics

Update #2: 11/22/13

SPSS and STATA Programming Statements have been added.

Update #1: 01/03/97

Estimates Derived from Public Use Files

The Household Component (HC) of the Medical Expenditure Panel Survey (MEPS) collects detailed information on the health status, health care use and expenses, and health insurance coverage of individuals and families in the United States. MEPS estimates of the uninsured differ somewhat from those of the Census Bureau's Current Population Survey (CPS) and of the National Center for Health Statistics' National Health Interview Survey (NHIS) because of definitional and methodological differences.

  • The surveys differ in the "reference period" for the estimate of uninsurance. For example, the CPS counts persons that are uninsured for a full year, and NHIS measures persons that lack insurance at a given point in time -- the month before the interview. For this first release, MEPS considers persons uninsured if they lack insurance from 1/1/96 until the date of the first interview - a period averaging 5 months. When full-year MEPS data is available, AHRQ will be able to construct measures of uninsurance comparable to both CPS and NHIS.

  • The surveys differ in how respondents are contacted. The CPS is conducted annually, with a recall period of the entire year. The NHIS collects data on a continuous basis each year, with a recall period for insurance of a month. MEPS respondents are contacted three times to provide a years' worth of insurance information, and the recall period is variable -- the time between 1/1/96 and their first interview, or for later contacts, the time between interviews. These recall periods vary but average 4-5 months.

  • The surveys differ in their definition of "insurance". The CPS counts military veterans whose source of health care is the Department of Veterans Affairs as insured. CPS also counts the children of adults covered by Medicaid as insured. For these preliminary estimates, MEPS does not consider these children insured unless their families report them as such. AHRQ data analysts project that full year estimates from MEPS will show the percentage of the population uninsured to be somewhat lower than these part year estimates because some people will have obtained coverage. In addition, later data releases will incorporate information from payment sources for health care use in assigning insurance coverage to individuals. This additional information will also likely reduce the number of persons categorized as uninsured.

Finally, analysts should also consider standard errors for population estimates when comparing MEPS data to other data sources.

Information for PC-SAS Users:

Note: When doing PC-SAS conversion of ASCII data files, users must specify the LRECL in the INFILE statement, due to the fact that PC-SAS uses a default LRECL value of 256. Below is a sample INFILE statement to be used with the data file contained on the MEPS CD-ROM HC-001:

For the data file HC001.DAT: 'INFILE IN1 LRECL=264 RECFM=V;'


Back to topGo back to top
Get Social
Agency for Healthcare Research and Quality
5600 Fishers Lane
Rockville, MD 20857
Telephone: (301) 427-1364
Improving the Quality, Safety, Efficiency, and Effectiveness of Health Care For All Americans