MEPS-HC Sample Design
The Medical Expenditure Panel Survey Household Component (hereafter, MEPS-HC) is a nationally-representative, longitudinal survey of the civilian non-institutionalized U.S. population. The MEPS-HC sample includes an oversample of Blacks and Hispanics. The MEPS-HC, administered by the Agency for Healthcare Research and Quality (AHRQ), collects information about families and individuals in five interviews conducted over a two-year period. For most MEPS-HC panels, the selected sample represents about three-eighths of the households responding to the previous year's National Health Interview Survey (NHIS) collected by the National Center for Health Statistics. The linkage of the MEPS-HC to the previous year's NHIS provides additional data for longitudinal analytic purposes.
Typically, sample allocations across sample domains change from one MEPS-HC panel to another. The sample domains used may also (but don't necessarily) vary by panel. Within each domain (sample stratum) MEPS selected systematic samples of the MEPS-HC-eligible households from among the NHIS household respondents made available by NCHS.
Differences between the MEPS-HC and NHIS samples
Once the MEPS-HC sample is selected from among the NHIS households characterized as NHIS respondents, Responding Units (RUs) representing students living in student housing or consisting entirely of military personnel are deleted from the sample. For the NHIS, college students living in student housing are sampled independently from their families. For MEPS-HC, such students are identified through the sample selection of their parents' RU. These college students living in student housing are removed from the MEPS-HC samples in order to eliminate the opportunity of multiple chances of selection for MEPS-HC for these students. Military personnel not living in the same RU as civilians are ineligible for MEPS-HC. After such exclusions, all RUs associated with households selected from among those identified as NHIS responding households are then fielded in the first round of MEPS-HC.
MEPS-HC subsampling has to be done soon after NHIS responding households are identified, causing a small percentage of the NHIS households initially characterized as NHIS respondents to later be classified as nonrespondents for the purpose of NHIS data analysis. This means that some MEPS-HC respondents are not actually represented in the NHIS dataset. It is not possible to link these MEPS-HC respondents to their corresponding NHIS data. When these households respond to MEPS but are nonresponding in NHIS, there is a slight increase in the overall MEPS-HC response rate.
If an individual was in a NHIS household, but moved to a new household at any point between their NHIS interview and before the conclusion of their MEPS-HC panel, MEPS includes this new household in their sample. The exception to this is new households created after the NHIS interviews for the respective panels and consisting exclusively of persons who entered the target population after the associated NHIS survey. Neither are previously out-of-scope persons who joined existing households but are unrelated to the current household residents. Because of these excpetions some members of the following groups are not covered by the MEPS-HC: immigrants, persons leaving the military, U.S. citizens returning from residence in another country, and persons leaving institutions. Those not covered represent only a small proportion of the MEPS-HC target population.
For more information about persons in the MEPS-HC who are in-scope for only part of the year, please refer to this report by Agency for Healthcare Research and Quality staff.
Oversampling is a feature of the MEPS-HC sample design, improving the precision of estimates for some subgroups of interest. The oversampling of Hispanic and Black households for the NHIS carries over to MEPS-HC. Under the NHIS sample design during the 1995-2005 period, Hispanic households were oversampled at a rate of 2 to 1. The oversampling of black households was roughly 1.5 to 1. Under a new NHIS sample design initiated in 2006, Asian households were also oversampled. The oversampling rates for these three minority groups under the new sample design, implemented by NHIS in 2016, have not yet been reported. For recent panels, all households in the Asian, Hispanic, and Black domains were sampled with certainty (i.e. all households assigned to those domains were included in the MEPS-HC). Sample weights ensure that population estimates are not distorted by a disproportionate contribution from oversampled subgroups.Note: This user note is based on the documentation accompanying the 2014 Medical Expenditure Panel Survey Full Year Consolidated File (H-171), created by staff at the Agency for Healthcare Research and Quality, Center for Financing, Access, and Cost Trends.