Methodology Report #22:
Sample Design of the Medical Expenditure Panel Survey Household Component, 1998–2007
Trena M. Ezzati-Rice, Frederick Rohde, Janet Greenblatt
Table of Contents
_._ Abstract
_._ The Medical Expenditure Panel Survey (MEPS)
_._ Background
_._ Sample Design
_._ Sample Yields and Survey Response Rates
_._ References
_._ Tables
_._ Table 1. Key Sample Design Features of the Medical Expenditure Panel Survey Household Component: 1996-2007
_._ Table 2a. Number of responding dwelling units, families, and persons, by year and panel
_._ Table 2b. Number of responding dwelling units, families, and persons, by year and panel
_._ Table 3a. Number of completed person-level interviews by age, sex, race/ethnicity, region, and MSA status (based on
MEPS Point in Time files), MEPS 1996-2005
_._ Table 3b. Number of completed person-level interviews by age, sex, race/ethnicity, region, and MSA status (based on
MEPS Full Year Files): MEPS-HC 1996-2005
_._ Table 4. Example sample size and response rates - Full Year (Panel 7, Rounds 1-3 combined with Panel 6, Rounds 3-5):
MEPS 2002
_._ Table 5. MEPS Individual Panel and Combined Annual Response Rates
Abstract
This report describes the sample design for the
1998−2007 Medical Expenditure Panel Survey Household Component (MEPS-HC)
and provides updates to the earlier descriptions of the 1996 and 1997
MEPS-HC sample designs. Since the sample of households selected for each
panel of MEPS-HC is a subsample of households participating in the
previous year’s National Health Interview (NHIS) conducted by the
National Center for Health Statistics of the Centers for Disease Control
and Prevention, the report includes a summary of the key features of the
NHIS sample design. This report also includes the target sample sizes,
number of sampled units, number of completed interviews, and response
rates for the MEPS-HC.
The estimates in this report are based on the most
recent data available at the time the report was written. However,
selected elements of MEPS data may be revised on the basis of additional
analyses, which could result in slightly different estimates from those
shown here. Please check the MEPS Web site for the most current file
releases.
Center for Financing, Access, and Cost Trends
Agency for Healthcare Research and Quality
540 Gaither Road
Rockville, MD 20850
http://www.meps.ahrq.gov/_
Return to Table of Contents
The Medical Expenditure Panel Survey (MEPS)
Household Component
The Medical Expenditure Panel Survey (MEPS) provides
nationally representative estimates of health care use, expenditures,
sources of payment, and health insurance coverage for the U.S. civilian
non-institutionalized population. The MEPS Household Component (HC) also
provides estimates of respondents' health status, demographic and
socio-economic characteristics, employment, access to care, and
satisfaction with health care. Estimates can be produced for
individuals, families, and selected population subgroups. The panel
design of the survey, which includes five rounds of interviews covering
two full calendar years, provides data for examining person level
changes in selected variables such as expenditures, health insurance
coverage, and health status. Using computer assisted personal
interviewing (CAPI) technology, information about each household member
is collected, and the survey builds on this information from interview
to interview. All data for a sampled household are reported by a single
household respondent.
The MEPS-HC was initiated in 1996. Each year a new panel
of sample households is selected. Because the data collected are
comparable to those from earlier medical expenditure surveys conducted
in 1977 and 1987, it is possible to analyze long-term trends. Each
annual MEPS-HC sample size is about 15,000 households. Data can be
analyzed at either the person or event level. Data must be weighted to
produce national estimates.
The set of households selected for each panel of the
MEPS-HC is a subsample of households participating in the previous
year's National Health Interview Survey (NHIS) conducted by the National
Center for Health Statistics of the Centers for Disease and Control
Prevention. The NHIS sampling frame provides a nationally representative
sample of the U.S. civilian non-institutionalized population and
reflects an oversample of blacks, Hispanics, and starting in 2006,
Asians. MEPS oversamples additional policy relevant sub-groups such as
low income households. The linkage of the MEPS to the previous year's
NHIS provides additional data for longitudinal analytic purposes.
Medical Provider Component
Upon completion of the household CAPI interview and
obtaining permission from the household survey respondents, a sample of
medical providers are contacted by telephone to obtain information that
household respondents can not accurately provide. This part of the MEPS
is called the Medical Provider Component (MPC) and information is
collected on dates of visit, diagnosis and procedure codes, charges and
payments. The Pharmacy Component (PC), a subcomponent of the MPC, does
not collect charges or diagnosis and procedure codes but does collect
drug detail information, including National Drug Code (NDC) and medicine
name, as well as date filled and sources and amounts of payment. The MPC
is not designed to yield national estimates. It is primarily used as an
imputation source to supplement/replace household reported expenditure
information.
Survey Management
MEPS-HC and MPC data are collected under the authority
of the Public Health Service Act. Data are collected under contract with
Westat. Data sets and summary statistics are edited and published in
accordance with the confidentiality provisions of the Public Health
Service Act and the Privacy Act. The National Center for Health
Statistics (NCHS) of the Centers for Disease Control and Prevention
provides consultation and technical assistance.
As soon as data collection and editing are completed,
the MEPS survey data are released to the public in staged releases of
summary reports, micro data files, and tables via the MEPS Web site: http://www.meps.ahrq.gov/.
Selected data can be analyzed through MEPSnet, an on-line interactive
tool designed to give data users the capability to statistically analyze
MEPS data in a menu-driven environment.
Additional information on MEPS is available from the
MEPS project manager or the MEPS public use data manager at the Center
for Financing Access and Cost Trends, Agency for Healthcare Research and
Quality, 540 Gaither Road, Rockville, MD 20850; 301-427-1406.
AHRQ Publications Clearinghouse
Attn: (publication number)
P.O. Box 8547 Silver Spring, MD 20907
800-358-9295
703-437-2078 (callers outside the United States only)
888-586-6340 (toll-free TDD service; hearing impaired only)
To order online, send an e-mail to: ahrqpubs@ahrq.gov.
Be sure to specify the AHRQ number of the document or
CD-ROM you are requesting. Selected electronic files are available
through the Internet on the MEPS Web site:
http://www.meps.ahrq.gov/
For more information, visit the MEPS Web site or e-mail
mepspd@ahrq.gov.
Return to Table of Contents
Background
The Medical Expenditure Panel Survey (MEPS) Household
Component, a nationally representative sample of the U.S. civilian
non-institutionalized population, has been conducted on an annual basis
since 1996 by the Agency for Healthcare Research and Quality (AHRQ),
U.S. Department of Health and Human Services (DHHS). The MEPS provides
national estimates of health care use, expenditures, sources of payment,
and health insurance coverage as well as information on respondents'
health status, demographic/socio-economic characteristics, employment
status, access to health care, and satisfaction with health care. The
survey reports estimates for persons and families as well as subgroups
of the population.
This report describes the sample design of the MEPS
Household Component (MEPS-HC) for 1998 through 2007 and updates
descriptions of the earlier MEPS sample designs. MEPS Methodology Report
2 details the sample design of the 1996 MEPS-HC and MEPS Methodology
Report 11 describes the 1997 MEPS (Cohen SB, 1997; Cohen SB, 2000). An
additional report provides an overview of the core components of the
MEPS data collection and the statistical features of the survey (Cohen
SB, 2003). This updated report also includes the target sample sizes,
number of sampled units, number of completed interviews and response
rates for the recent panels of the MEPS.
Return to Table of Contents
Sample Design
Summary of Sample Design
The MEPS-HC is a complex national probability survey of
the U.S. civilian noninstitutionalized population. Each year a new panel
of households is selected from among those households that participated
in the previous year’s National Health Interview Survey (NHIS), another
large ongoing Federal health survey conducted by the National Center for
Health Statistics of the Centers for Disease Control and Prevention,
DHHS. Each new MEPS annual sample is referred to as a panel. In the MEPS,
for each new annual sample, data are collected through a series of five
rounds of computer-assisted personal interviews (CAPI) over 30 months to
yield annual data for two full calendar years.
As illustrated in Figure 1, a new MEPS panel of
households has been selected and fielded each year since 1996. During
each calendar year (with the exception of 1996) data are collected
simultaneously for two MEPS panels. One panel is in its first year of
interviews (e.g., in the year 2001, Rounds 1, 2, and 3 of Panel 6),
while the prior year’s panel is in its second year of data collection
(e.g., in 2001, Rounds 3, 4, and 5 of Panel 5).
Figure 1. MEPS Household Component
Overlapping Panel Design MEPS Panel
It should be noted that Round 3 for each MEPS panel
overlaps two calendar years. In 1996, the first year of MEPS, only one
panel (Panel 1) was fielded, thus the annual data for 1996 were based on
this single panel of data. However, starting in 1997, to increase
statistical power of annual estimates produced from MEPS, data are
combined across two distinct nationally representative samples, making
use of the MEPS overlapping panel design. More specifically, annual
estimates are made by combining data from the panel in its first year of
data collection and the panel in its second year of data collection. For
example, 2001 annual estimates are represented by data collected for the
second year of Panel 5 and data collected in year one of Panel 6.
In addition to annual estimates, the MEPS design
structure permits longitudinal estimates over two consecutive calendar
years, thus allowing examination of person-level changes in selected
variables over a two year period. For example, research analysts can
assess the persistence of high health care expenditures by examining
whether individuals with high expenditures in one year have high
expenditures in the subsequent year or shift to a higher or lower
expenditure level (Cohen SB and Ezzati-Rice TM, 2006).
Return to Table of Contents
MEPS Linked to the National Health Interview Survey (NHIS)
The set of households selected for each panel of the
MEPS-HC is a subsample of households participating in the previous
year's NHIS conducted by the National Center for Health Statistics of
the Centers for Disease Control and Prevention. The NHIS sampling frame
is a nationally representative sample of the U.S. civilian
noninstitutionalized population. The use of a subsample of NHIS
households provides budgetary savings by eliminating the need to
independently list and screen households to locate selected
policy-relevant subgroups of the population. The linkage also provides
an additional data point for enhanced longitudinal analyses (Cohen SB,
2003; Cohen SB, Makuc DM, Ezzati-Rice TM, 2007).
Since the MEPS sample of households is subsampled from
the NHIS, some knowledge of the NHIS sample design is needed in order to
understand the MEPS sample design. The NHIS has been in continuous
operation since 1957. Every ten years the NHIS sample design is updated
to reflect the changes to the U.S. population. Detailed information
about the NHIS sample designs is available from the NCHS Web site. 1 For
example, the NCHS Series 2, Number 130 report describes the sample
design of the 1995−2004 NHIS. The subsample of households selected for
each MEPS panel from 1996−2006 was based on the 1995−2004 NHIS sample
design.
Return to Table of Contents
Overview of 1995-2004 NHIS Sample Design
The 1995−2004 NHIS was based on a stratified multi-stage
sample design. A brief and simplified description of the NHIS design
follows. The first stage of sample selection was an area sample of
Primary Sampling Units (PSUs), where PSUs generally consisted of one or
more counties. Many PSUs were selected with certainty, i.e.,
"self-representing" PSUs. Within each PSU, density strata were formed
using 1990 Census population distributions of Hispanic persons and black
persons for single or groups of blocks or block equivalents. Within each
density stratum, "supersegments" were formed, consisting of clusters of
housing units. Samples of supersegments were selected for use over a
10-year data collection period for the NHIS. Reserve samples for two
additional years were also selected. Households within supersegments
were subsequently selected for each calendar year the NHIS was carried
out. Households containing Hispanics and blacks were oversampled at
rates of approximately 2 and 1.5 times, respectively, the rate of
remaining households.
The complete NHIS sample for each year of the 1995−2004
NHIS consisted of 358 PSUs. Each year a sample of approximately 76,000
households was selected and approximately 40,000 households were
interviewed. The annual NHIS sample of households is partitioned into
four subdesigns, referred to as "panels", each with approximately the
same number of households (NHIS, Series 2 report). (The terminology
Panel for the NHIS has been used by the Census Bureau, the NHIS data
collection agent, for the 1985−1994 and 1995−2004 NHIS, and this term
"Panel" should not be confused with the term’s frequent, but different
usage in the context of longitudinal surveys like the MEPS.) In the NHIS
Panel context, the NHIS survey sample is considered an all-area sampling
frame. In parts of the country where local governments issue building
permits, the area sample is supplemented with a sample of permits for
residential housing units built after the decennial census. Within each
sample PSU, the survey uses an area frame and in some PSUs a permit
frame is also used. The area sample and permit frame listings define the
Second Stage Units (SSUs) from which the NHIS household samples are
taken over the life of the design, usually a 10 year period. (Note: The
1995 NHIS sample design was used for an 11 year period, i.e.,
1995−2005). The SSUs for the NHIS are partitioned into four subdesigns
or Panels, and are identified by Panel labels 1, 2, 3, or 4. Typically,
an SSU is assigned a panel label which remains fixed for the life of the
survey.
There are two main objectives of the NHIS Panel
subdesign structure. The first objective is to provide nationally
representative subdesigns with "similar" features but with smaller
sample sizes as contingencies to deal with any potential NCHS budget
exigencies for the NHIS. The second objective is to provide a subsample
for use as a sampling frame for a smaller "follow-on" survey (i.e., a
survey whose sample design is then said to be linked with that of the
NHIS). Panels can be further sub-divided by sample assignment weeks
(e.g., calendar quarters) to provide even smaller surveys. Since 1996,
two panels of the NHIS have been reserved for use by AHRQ for the MEPS.
Each MEPS panel can be linked back to the previous year’s NHIS public
use data files. For information on obtaining MEPS/NHIS link files,
please see www.meps.ahrq.gov/data_stats/download_data_files.jsp.
Return to Table of Contents
1998−2006 MEPS Sample Design (MEPS Panels 3 through 11)
This section documents the 1998−2006 MEPS sample
designs. Earlier reports provide the details for the 1996 and 1997
designs (Cohen SB, 1997; Cohen SB, 2000).
Target Population and Sampling Frame
The target population for the MEPS consists of all
persons who are members of the civilian noninstitutionalized (e.g., not
in prisons or nursing homes) population at any time during the year and
living in the 50 States or the District of Columbia. In addition, the
sample is designed to meet a specified level of precision in estimates
for certain subgroups of the population, including persons who have a
family income less than 200% of the Federal poverty level and selected
racial and ethnic groups.
The NHIS serves as the sampling frame for the MEPS. In
most years, the MEPS annual household panel sample is selected from
responding households in two of the four NHIS panels during calendar
quarters 1−3 of the previous year. The NHIS quarter 4 is not processed
soon enough for use in selecting the following year’s MEPS sample as
each new MEPS panel must be fielded in January. Thus, a sample
representing about three-eighths of the NHIS responding households is
generally made available for use in MEPS.
Analytical goals, precision requirements, and sample size targets
The broad sample design goals for the MEPS include:
- A sample that will provide unbiased
national and regional estimates (four Census regions) of health care
expenditure estimates and other health parameters with targeted
precision, and
- A sample that will meet targeted
precision requirements for policy relevant subgroups of the
population.
Based on varying DHHS objectives coupled with the MEPS
budget resources, the sample size and subdomains oversampled for MEPS
can vary from year to year. The overall target precision requirement for
the current MEPS-HC (2001 and forward) is an average design effect of
1.6 for key survey estimates for policy relevant population subgroups
(Cohen SB, 2003). The MEPS person-level precision requirements are
specified for national estimates derived from individuals that are
considered full year respondents (individuals with responses for their
entire period of living in the civilian noninstitutionalized
population). Consequently, in the determination of sample sizes
necessary to achieve the precision requirements, adjustments must be
made for dwelling unit nonresponse and survey attrition to determine the
required number of initial sample units per year. Starting with calendar
year 2002, the target sample size for producing annual estimates for the
MEPS-HC is approximately 15,000 families (reporting units) or about
37,000 persons. While these target precision levels and samples sizes
have been specified, the targets are sometimes modified based on
analytical objectives and AHRQ budget resources.
Return to Table of Contents
Sample Selection Criteria and Oversampling
The initial MEPS-HC sample consists of a subsample of
households that responded to the prior year’s NHIS. After selection of
the NHIS households (occupied dwelling units), "reporting units" are
formed based on information collected in NHIS and for fielding of the
MEPS sample. In brief, a household may contain one or more family units,
each with one or more individuals. The NHIS family units become
reporting units (RUs) for MEPS. A more detailed discussion of the
definition of RUs for MEPS is found in the section "Sampling Unit
Definitions and Eligibility Criteria".
The initial sample size for each panel of the MEPS is
determined on the basis of the budget resources available at the time of
sample selection and on the eligible sample available from NCHS, thus
the sample sizes have varied from year to year. A review of the 1996 and
1997 MEPS is first provided followed by more specific details for the
1998−2006 MEPS. The 1995 NHIS subsample selected for the 1996 MEPS
consisted of 195 PSUs. An initial subsample of 10,597 households was
selected from NHIS Panels 1 and 3 in two targeted quarters (2 and 3) of
the NHIS. (Table 1) The 1997 MEPS panel (Panel 2) sample of 6,300
households was again selected in the same 195 PSUs (as in MEPS 1996) and
was selected as a subsample of households responding to the 1996 NHIS,
Panels 1 and 3 in three targeted NHIS quarters (1, 2, and 3). Both the
1996 and 1997 MEPS reflected an oversample of Hispanics and blacks at
the same ratios as in the NHIS (Hispanics, 2.0:1; blacks 1.5:1). In
addition, the 1997 MEPS oversampled several policy relevant domains at
varying rates (see Table 1).
In order to select the MEPS sample, the NHIS households
are included in the frame based on the following definitions:
- Responding Household: NHIS household (HH)
with ACTION code = 4 (partial interview, no follow-up) or ACTION
code 10 (complete interview) and at least one person with HHSTAT
(household status) not equal 'D' (deleted).
- Eligible Household: All responding HHs
in the NHIS quarters and panels set aside for MEPS. (In most years,
the HHs eligible for MEPS are from calendar quarters 1−3 in 2 of 4
NHIS panels.)
Prior to sample selection, the NHIS occupied dwelling
units (DUs) within pre-specified sampling classes are hierarchically
sorted by the following measures:
- Calendar year quarter
- Interview week within each respective calendar quarter
- Census division
- State
- Metropolitan Statistical Area (MSA) classification
- NHIS primary sampling unit (PSU)
- NHIS segment within PSU
The household level sampling domain variables vary
slightly from year to year and include a hierarchical classification:
any Asian in household, any family in household with predicted poverty,
any Hispanic in household, any black in household, and all others (i.e.,
no Asian, predicted poverty, Hispanic, or black).
The sample of households for the 1998−2001 MEPS panels
ranged from approximately 5,200−11,000 households (Table 1, Column 7).
In 1999 and 2000, due to budgetary constraints, the number of MEPS PSUs
was reduced from 195 to 100. In 2002, some design enhancements were made
with an increased sample size and the number of PSUs also increased to
195 as in the earlier MEPS panels. The sample of households for the
2002−2006 MEPS from the 2001−2005 NHIS ranged from about 8,100 to 9,500
households. As in the earlier years, the oversample of Hispanics and
blacks in the NHIS carried over to MEPS. In addition, the NHIS
responding households eligible for MEPS that contained either Asian
Americans or families predicted (based on a statistical model) to have
low income (i.e., <200% Federal poverty level) were selected with
certainty. For the 2004−2006 MEPS panels, in addition to the certainty
selection of Asians and low income families, households containing
blacks and not among those households selected with certainty were
further oversampled. The sampling rates by subdomains, the number of
selected NHIS households, number of PSUs, and number of initial MEPS
Reporting Units by MEPS panel and year are shown in Table 1. (Note: The
sample sizes presented in Table 1 are confined to the new panel
introduced each year. The number of responding families and persons for
use in producing annual estimates are discussed later.)
Return to Table of Contents
Sampling Unit Definitions and Eligibility Criteria
The definition of dwelling units and group quarters in
the MEPS-HC are generally consistent with the definitions employed for
NHIS. The definitions used are:
- Dwelling unit (DU) is a house,
apartment, group of rooms, or single room occupied as separate
civilian non-institutional living quarters or vacant but intended
for occupancy as separate living quarters. This term is the NHIS
definition for households and is the unit sampled for the MEPS.
- Group quarters consist of a single
civilian non-institutional dwelling or structure in which nine or
more unrelated persons reside and where inhabitants are not
considered a part of any other dwelling unit.
A Reporting Unit (RU) is a person or group of persons in
the sampled DU who are related by blood, marriage, adoption, foster
care, or other family association. Each RU was interviewed as a single
entity for MEPS. Thus, the RU serves chiefly as a family-based "survey"
operations unit rather than an analytic unit. Regardless of the legal
status of their association, two persons living together as a "family"
unit are treated as a single RU if they chose to be so identified.
Unmarried college students (less than 24 years of age) who usually live
in a sampled household but who live away from home and go to school at
the time of the MEPS interview are treated as a student RU separate from
that of their parents for the purpose of data collection. Examples of
different types of RUs are:
- A married daughter and her husband living with her parents in
the same DU constitute a single RU;
- A husband and wife and their unmarried daughter, age 18, who
is living away from home while at college constitute two RUs; and
- Three unrelated persons living in the same DU would each
constitute a distinct RU, i.e., a total of three RUs.
Return to Table of Contents
MEPS Data Collection Eligibility
The only major difference in eligibility status for
housing units between NHIS and MEPS is that college dorms represent
ineligible dwelling units for MEPS. College aged students living away
from home during the school year were interviewed at their place of
residence for the NHIS, but, in contrast, are identified by and linked
to their parents’ household for MEPS. Once the MEPS sample is selected
from among the NHIS households characterized as NHIS respondents, 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, such students are identified through the sample
selection of their parents' RU. In MEPS, removing college students found
in college housing sampled for the NHIS eliminates the opportunity of
multiple chances of selection for MEPS for these students. Military
personnel not living in the same RU as civilians are ineligible for MEPS.
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. The initial number of RUs for each MEPS panel
is shown in Table 1, Column 8. Standard or primary RUs are the original
RUs from NHIS. However, a new RU will be created when members of the
household leave the primary RU and are followed according to the rules
of the survey.
Three key factors define a person’s interview status for
each round of data collection in the MEPS. These factors are: "in-scope"
status, Keyness status, and eligibility status.
In-scope
A person is considered as in-scope during a MEPS round
or a reference time period if he or she was a member of the U.S.
civilian, noninstitutionalized population at some time during the
specific data collection round or time period.
Return to Table of Contents
Keyness for MEPS—“Key and non-Key Persons”
The term "Keyness" is related to an individual’s chance
of being included in MEPS. A person is "Key" if a person is linked for
sampling purposes to the set of NHIS sampled households designated for
inclusion in MEPS. Specifically, a "Key" person was a member of an NHIS
household at the time of the NHIS interview or who becomes a member of
such a household after being out-of-scope at the time of the NHIS
(examples of the latter situation include newborns and persons returning
from military service, an institution, or living outside the United
States).
A "non-Key" person is one whose chance of selection for
the NHIS (and MEPS) was associated with a household eligible, but not
sampled for the NHIS, and who later becomes a member of a MEPS RU. MEPS
data are collected for the period of time a non-Key person is part of
the sampled unit to provide information for family-level analyses.
However, non-Key persons who leave a sample household unaccompanied by a
Key, in-scope member are not followed for subsequent interviews.
In summary, Keyness status is set at the time the person
enters MEPS, and a person’s Keyness status never changes. It should be
noted that a person might be Key even though not part of the civilian,
noninstitutionalized portion of the U.S. population. For example, a
person in the military may have been living with his or her civilian
spouse and children in a household sampled for NHIS. The person in the
military would be considered a "Key" person for MEPS; however, such a
person would not be eligible to receive a person-level sample weight if
he or she was never in-scope during a defined survey period.
Eligibility
The eligibility of a person for MEPS pertains to whether
or not data are to be collected for that person. All of the Key in-scope
persons of a sampled RU are eligible for data collection. The only
non-Key persons eligible for data collection are those who live in an RU
with at least one Key, in-scope person. Their eligibility continues only
for the time that they are living with at least one such person. The
only out-of-scope persons eligible for data collection are those living
with Key in-scope persons, again only for the time they live with such a
person. (Only persons in the military can meet this description e.g., a
person on full-time active military duty, living with a spouse who is
Key).
Return to Table of Contents
2007 MEPS (Panel 12) and 2006 NHIS Sample Redesign
As stated at the beginning of this report, the NHIS
traditionally is redesigned about every 10 years following the most
current decennial census to take account of demographic changes in the
population. Since its beginning, the MEPS-HC has been conducted in a set
of NHIS PSUs based on the sample design developed for the 1995−2004 NHIS.
The selection of PSUs for what is known as the 1995−2004 NHIS sample
design was based on 1990 Census data. It should be noted, however, that
the 1995 design was used for an additional NHIS year, that is, through
2005. NHIS implemented a new sample design in 2006, and it is
anticipated that the new sample design will be in place until the next
redesign around 2014. The new 2006 NHIS sample design is based on the
2000 Census. This section primarily focuses on changes that occurred in
the new NHIS design relative to the previous design and the associated
changes for the 2007 MEPS and forward.
As for the earlier NHIS sample designs, the 2006−2014
NHIS design is a complex multistage sample design of the civilian
noninstitutionalized U.S. population with stratification, clustering,
and oversampling of selected population subgroups. The new 2006 NHIS
sample of households was sampled independently from that which was
selected under the 1995−2004 design. However, the fundamental design
structure of the new 2006 NHIS sample design is very similar to the
previous sample design which was in place from 1995 to 2005. But, there
are some important differences which are highlighted in the following
discussion.
The target universe for the NHIS is all dwelling units
in the U.S. that contain members of the civilian noninstitutionalized
population. As in the previous design, the target universe was first
partitioned into primary sampling units (PSUs), each of which consisted
of a single county, a group of contiguous counties, or metropolitan
areas. Those PSUs defining the largest metropolitan statistical areas
were selected with certainty and were designated as self-representing
(SR) PSUs. The remaining PSUs in the universe were designated as non
self-representing (NSR) or non-certainty PSUs and a sample of these PSUs
was selected. The NSR PSUs were stratified by state and sampled into the
NHIS generally at the rate of two PSUs per state using a probability
proportional to population size according to the 2000 Census. Within a
few NSR strata with smaller population sizes, one PSU was selected. Most
of the self-representing PSUs of the previous design are still
self-representing in the new design. The differences in the location of
the PSUs between the two designs occur mostly in the NSR areas. The new
design has 428 PSUs. The previous design had 358 PSUs. The difference in
the number of PSUs is largely due to differences in how the PSUs are
defined in the new design compared to the previous design. In the new
design, NHIS is partitioning most of the SR PSUs into mini-PSUs, i.e.,
smaller geographic areas (one or more counties) known as SPSUs or
stratification PSUs. For example, most surveys will refer to a large SR
PSU like Boston as one big SR PSU. However, with the new NHIS design,
there are multiple distinct PSUs (SPSUs) for the Boston area.
Oversampling of the black and Hispanic populations was
retained in the 2006 NHIS design to facilitate estimation of health
related statistics for these two minority groups. The new sample design
also includes an oversample of the Asian population. As in the previous
design, at the second stage of sampling, the entire area within each
selected PSU was partitioned into a substrata consisting of single or
contiguous blocks or block equivalents. These substrata were assigned to
20 density strata defined by the concentration of blacks, Hispanics and
Asians from the 2000 Decennial Census. New construction housing (or
permit listings) within a PSU was included as its own substratum in
order to produce a current sample of households including new
construction. Thus, the number of substrata is 21, the same number as in
the previous design. The consideration of Asians in the definition of
the second-stage density strata was new under the 2006−2014 NHIS design;
the previous design had only taken into consideration the concentrations
of blacks and Hispanics. Also in the new design, the definitions that
constituted low, medium, and high concentrations of each minority group
were allowed to vary slightly from PSU to PSU. In the previous design,
the definitions were consistent in all PSUs. Finally, as with the
previous design, the area segments within each density stratum were
partitioned into supersegments or clusters of housing units. These
supersegments were subsequently sampled into the NHIS, and the housing
units within them assigned to each calendar year, quarter and week of
NHIS data collection. The sample adult selection process for the NHIS
was also revised under the new sample design in 2006 such that when
elderly black, Hispanic, or Asian persons 65 years or older were present
they have an increased chance of being selected as the sample adult. The
NCHS report describing the 1995−2005 design, Series 2, Number 130,
provides additional details that still apply to the new sample design.
This publication is available on-line at
http://www.cdc.gov/nchs/products/pubs/pubd/series/sr02/130-121/130-121.htm.
A new report providing more specific details of the 2006
NHIS sample design is being developed by NCHS.
Due to budgetary constraints, the total number of
sampled housing units under the 2006−2014 NHIS design was reduced
relative to the previous sample design
(
ftp://ftp.cdc.gov/pub/Health_Statistics/NCHS/Dataset_Documentation/NHIS/2006/srvydesc.pdf).
The reduction in households will come entirely from the nonminority
density strata. That is, the target number of households with black and
Hispanic respondents will be similar to that achieved under the old NHIS
design, while the number with Asian respondents will be greater under
the new design. The new NHIS sample design is anticipated to result in
approximately 87,500 persons residing in 35,000 households with
completed interviews.
Return to Table of Contents
2007−2015 MEPS Sample Design
The sample of households for MEPS Panels 1 through 11
(1996−2006) was generally selected from the first three calendar
quarters of households in NHIS Panels 1 and 3 as sampled under the
1995−2005 NHIS design. Starting in 2007, the households in MEPS Panels
12 through 20 (2007−2015) will be selected from NHIS Panels 1 and 4 as
designed for the 2006−2014 NHIS design. The change to a subsample of
responding households in NHIS Panels 1 and 4 rather than NHIS Panels 1
and 3 was made to maximize the number of overlapping PSUs utilized by
the MEPS across the two independent NHIS designs. That is, the PSUs in
NHIS Panels 1 and 4 as selected for the 2006−2014 NHIS design have the
most overlap with the PSUs in NHIS Panels 1 and 3 as selected in the
1995−2004/5 NHIS design. This will minimize field data collection costs
for the MEPS.
In general, each new MEPS Panel will continue to be
sampled from the first three calendar quarters of the prior year’s
responding NHIS annual sample among the two panels set aside for MEPS.
To reduce operational issues associated with fielding a new sample
design in the same year as implementation of a new windows-based
computer-assisted personal interview (CAPI) instrument, the 2007 MEPS
sample was limited to eligible responding housing units from the first
two calendar quarters, Panels 1 and 4 of the 2006 NHIS. The sample yield
and sampling rates for the MEPS Panel 12 fielded in January 2007 are
shown in Table 1. In particular, a total of 7,319 NHIS housing units,
representing 7,467 responding units, were selected for the MEPS Panel
12.
Return to Table of Contents
Sample Yields and Survey Response Rates
Sample Size
As described in the earlier "Summary of Sample Design"
section, to produce MEPS annual calendar year estimates, data are
combined across two overlapping panels. A preliminary data file is also
produced in MEPS which allows selected estimates to be produced for the
first half of the year at approximately six months following data
collection. These early estimates are also based on two overlapping
panels, namely data collected during Round 1 of a current year’s panel
and during Round 3 of the previously fielded panel. Tables 2a and 2b
provide a summary of the number of completed interviews (households
(dwelling units), families, and persons) by year and panel based on the
MEPS Point-in-Time Files and Annual Full Year Files, respectively.
Tables 3a and 3b provide the number of completed person-level interviews
by age group, gender, race-ethnicity group, region, and MSA status based
on the Point-in-Time Files and Annual Full Year Files, respectively.
While sample size is an often used indicator of the
reliability of estimates obtained from a survey, sample size is not the
only feature of the sample that affects the reliability of estimates for
a stratified multistage sample survey like the MEPS and its sample
frame, the NHIS. The NHIS sample is clustered with multiple stages of
selection and includes oversampling of selected minorities all of which
carry over to the MEPS. In addition, the MEPS sample requires additional
adjustments to account for differential selection probabilities along
with the post-survey adjustments for household and person-level
nonresponse. In this setting, the precision of the survey estimates can
be expected to be less (compared to a simple random sample) when the
clustering, multistage sample design factors, and unequal weighting are
accounted for in the estimates of variance. Users of the MEPS micro data
must be aware that standard statistical estimation software does not
properly account for these factors in the estimates of variability (such
as the standard error of sample estimates or corresponding confidence
intervals) for MEPS estimates. Several methodologies have been developed
for estimating standard errors for surveys with a complex sample design,
including the Taylor-series linearization method, balanced repeated
replication, and jackknife replication. Various software packages
provide analysts with the capability of implementing these
methodologies. The variables needed to calculate appropriate standard
errors based on the Taylor-series linearization method are included on
the MEPS public use files. Software packages that permit the use of the
Taylor-series linearization method include SUDAAN, Stata, SAS (version
8.2 and higher), and SPSS (version 12.0 and higher). Users of these
software packages should refer to the corresponding software user
documentation for complete information on the capabilities of each
package.
Return to Table of Contents
Response Rates
The general approach for calculation of MEPS response
rates along with an illustrative example is provided in this section. In
particular, response rates for annual 2002 calendar year data are
discussed. Because of the linkage of the NHIS and the MEPS, the response
rate for MEPS is a combination of the response rate for the NHIS and the
MEPS round specific response rates. Due to the overlapping panel design
for production of annual estimates, the calculation of the annual
response rates for MEPS likewise comprised the two overlapping panel
specific response rates. Further, the panel-specific response rates get
weighted by their respective sample sizes.
To understand the calculation of MEPS response rates,
some key features related to MEPS data collection are first discussed.
When an RU is visited for a round of data collection, any changes in RU
membership are identified. Such changes include RU members who have
moved to another location in the U.S., thus creating a new RU to be
interviewed for MEPS, as well as student RUs. Thus, the number of RUs
eligible for MEPS interviewing in a given round can only be determined
after data collection is fully completed. The ratio of the number of RUs
completing the MEPS interview in a given round to the number of RUs
characterized as eligible to complete the interview for that round
represents the "conditional" round specific response rate expressed as a
proportion. It is "conditional" in that it pertains to the set of RUs
characterized as eligible for MEPS for that round, and thus is
"conditioned" on prior participation rather than representing the
overall response rate through that round. For example, in Table 4, for
Panel 6, Round 2 the ratio of 9,222 (Row G) to 9,666 (Row F) multiplied
by 100 represents the percentage response rate for Round 2 (95.4 percent
when computed), conditioned on the set of RUs characterized as eligible
for MEPS for Round 2. Taking the product of the response rate of the
NHIS sample designated for use in MEPS (Row A) and the product of the
response rates (ratio of the number of completed interviews to the
number eligible) for each consecutive set of MEPS rounds beginning with
round one produces the overall response rate through the last MEPS round
specified.
An example response rate calculation for a full year
MEPS is as follows. To produce annual health care and expenditure
estimates for calendar year 2002, data from Panel 6 and Panel 7 are
combined. More specifically, data collected covering calendar year 2002
for Rounds 3 through 5 of Panel 6 are combined with corresponding 2002
data from the first three rounds of Panel 7 to produce calendar year
2002 estimates. The overall response rate for the combined sample in
Panels 6 and 7 for 2002 was obtained by computing the product of the
relative responding sample sizes and the corresponding overall panel
response rates and then summing the two products. Details of the
calculations as carried out for the 2002 MEPS annual response rate are
provided below.
Return to Table of Contents
Panel 7 Response Rate
For MEPS Panel 7, Round 1, a total of 8,083 households
was fielded in 2005 (Row C of Table 4), a nationally representative
subsample of the households who had responded to the 2001 NHIS. Table 4
further shows for each round the number of RUs eligible for interviewing
as well as the number of RUs completing the MEPS interview. Computing
the three individual round "conditional" response rates and then taking
the product of the three response rates and the factor 89.7 (the
percentage of the NHIS sampled households designated for use in MEPS,
i.e., those initially characterized as responding in NHIS) yielded an
overall response rate of 65.6 percent for Panel 7 through Round 3.
Panel 6 Response Rate
For MEPS Panel 6, a total of 10,651 households was
fielded in 2001 (as indicated in Row C of Table 4), a nationally
representative subsample of the households who had responded to the 2000
NHIS. Table 4 shows the number of RUs eligible for interviewing and the
number completing the household interview for all five rounds of Panel
6. The overall response rate for Panel 6 was computed in a similar
fashion to that of Panel 7, but covered all five rounds of MEPS
interviewing rather than just the first three rounds. The overall
response rate for Panel 6 through Round 5 was 64.0 percent.
2002 Combined Panel Response Rate
The combined response rate for the 2002 annual data was
obtained by taking a weighted average of the panel specific response
rates. The Panel 6 response rate was weighted by a factor of 0.55 and
the Panel 7 response rate was weighted by a factor of 0.45, reflecting
approximately the distribution of the overall sample between the two
panels. (Note: The panel specific weighting factors vary from year to
year.) The resulting overall annual response rate for the combined
panels was thus computed as (.55 x 64.0) plus (.45 x 65.6) for a final
combined 2002 annual response rate of 64.7 percent (as shown in the last
row of Table 4).
The individual panel-specific and the overall (combined
panel response rates) response rates for annual MEPS data analyses are
summarized in Table 5. These response rates reflect the response to both
the NHIS and the multiplicative MEPS round specific response rates. For
1996 thru 2000, the standardized response rates shown in Table 5 are
slightly different from those in the 1996 and 1997 Methodology Reports
and in the public use file documentations. This is due to a slight
modification in the methodology for computing the response rates. The
resulting changes are minor, but more importantly, the change provides
standardized response rates for tracking trends across time.
Return to Table of Contents
Development of Analysis Weights
Weights are developed for use in the derivation of
nationally representative population estimates to support analysis of
data collected in sample surveys. The analytical weights typically
account for any disproportionate probabilities of selection, unit
nonresponse, person-level survey attrition (e.g., in a longitudinal
survey), and an adjustment to make the weighted sample distributions
agree with known population estimates. This general approach is used in
the MEPS (Wun LM, Ezzati-Rice TM, et al, 2007), including (1) an
adjustment for dwelling unit nonresponse at Round 1 to account for
nonresponse among those households subsampled from NHIS for the MEPS,
(2) an adjustment for nonresponse at the person level to account for
survey attrition across the multiple rounds of data collection, and (3)
a final step of poststratification and raking to known population totals
for the civilian noninstitutionalized population of the United States.
More detailed information on the estimation strategies used in MEPS can
be found on the MEPS Web site: www.meps.ahrq.gov.
Summary
The MEPS, a comprehensive population-based health care
survey, is an important resource to inform health care policy and
practice. This report provides an update to the 1996 and 1997 MEPS
sample design descriptions with a comprehensive overview of the MEPS
Household Component sample design for 1998 through 2007. Since the set
of households selected for each new panel of the MEPS is a subsample of
those who participated in the previous year’s National Health Interview
Survey (NHIS), key statistical features of the NHIS sample design are
also included. In addition, this report includes a summary of the total
number of households sampled for the MEPS, the number of responding
dwelling units, families, and persons by year, and the overall annual
survey response rates.
Acknowledgment: The authors
gratefully acknowledge the expert technical review and helpful comments
provided by Steven B. Cohen, Director, Center for Financing, Access, and
Cost Trends (CFACT), AHRQ, and Steven R. Machlin, Division of
Statistical Research and Methods, CFACT, AHRQ.
Return to Table of Contents
References
Cohen SB. Sample Design
of the 1996 Medical Expenditure Panel Survey Household Component.
Rockville (MD): Agency for Health Care Policy and Research; 1997. MEPS
Methodology Report No. 2. AHCPR Pub. No. 97-0027.
Cohen SB. Sample Design
of the 1997 Medical Expenditure Panel Survey Household Component.
Rockville (MD): Agency for Health Care Research and Quality; 2000. MEPS
Methodology Report No. 11. AHRQ Pub. No. 01-0001.
Cohen SB. Design Strategies and Innovations in the
Medical Expenditure Panel Survey. Medical Care,
July 2003: 41(7) Supplement: III-5−III-12.
Cohen SB, Makuc DM, and Ezzati-Rice TM. Health insurance
coverage during a 24 month period: a comparison of estimates from two
national health surveys. Journal of Health Services and Outcomes
Research Methodology (2007) 7:125−144.
Cohen SB and Ezzati-Rice TM. (2006). Designing national
health care surveys to inform health policy. In Statistics:
A Guide to the Unknown, Fourth Edition, 89−101. Belmont (CA):
Duxberry of Thomson Brooks/Cole.
Wun LM, Ezzati-Rice TM, Diaz-Tena N, and Greenblatt J.
On modeling response propensity for dwelling unit (DU) level
non-response adjustment in the Medical Expenditure Panel Survey (MEPS).
Statistics in Medicine, 2007: 26:1875−1884.
Return to Table of Contents
Tables
Table 1. Key Sample Design Features
of the Medical Expenditure Panel Survey Household Component: 1996−2007
Panel |
Year |
Number of PSUs*
Source of Sample = NHIS*) |
Eligible NHIS* Quarters/ Panels |
Additional oversamplesa |
Sample rate |
Number of sampled households (Dwelling Units) |
Initial number of reporting units (RUs) |
1 |
1996 |
195 |
Q2,3 /P1,3 |
None |
1.000 |
10,597 |
10,799 |
2 |
1997 |
195 |
Q1,2,3/ P1,3 |
Adults (18+) with functional impairmentsb
Children (<18) with limitations
Persons 18−64 predicted to incur high medical expenditures
Persons with low incomec
Adults (18+) with other limitationsd
Persons 65 years+
Other |
0.428
1.000
1.000
1.000
0.600
0.300
0.300
0.300 |
6,300
478
601
596
1,238
194
647
2,546 |
6,461 |
3 |
1998 |
195 |
Q2 /P1,3 |
None |
1.000 |
5,166 |
5,410 |
4 |
1999 |
100 |
Q1,2,3 /P1 |
None |
0.945 |
6,900 |
7,103 |
5 |
2000 |
100 |
Q1,2,3 /P1 |
None |
0.741 |
5,380 |
5,533 |
6 |
2001 |
195 |
Q1,2,3 /P1,3 |
None
Q1,2
Q3 |
0.738
0.750
0.716 |
10,704
7,004
3,700 |
11,026 |
7 |
2002 |
195 |
Q1,2,3 /P1,3 |
Q1,2 Asian/<200% povertye
Q1,2 Other
Q3 Asian/<200% povertye
Q3 Other |
0.560
1.000
0.500
1.000
0.393 |
8,132
1,718
3,781
953
1,680 |
8,339 |
8 |
2003 |
195 |
Q1,2,3 /P1,3 |
Q1,2 Asian/<200% povertye
Q1,2 Other
Q3 Asian/<200% povertye
Q3 Other |
0.616
1.000
0.500
1.000
0.595 |
8,400
1,623
3,698
825
2,254 |
8,706 |
9 |
2004 |
195 |
Q1,2,3 /P1,3 |
Q1,2 Asian/<200% povertye
Q1,2 Black
Q1,2 Other
Q3 Asian/<200% povertye
Q3 Black
Q3 Other |
0.634
1.000
0.750
0.600
1.000
0.500
0.431 |
8,640
1,516
640
3,554
1,050
271
1,609 |
8,939 |
10 |
2005 |
195 |
Q1,2,3 /P1,3 |
Q1,2 Asian/<200% povertye
Q1,2 Black
Q1,2 Other
Q3 Asian/<200% povertye
Q3 Black
Q3 Other |
0.646
1.000
0.750
0.600
1.000
0.750
0.427 |
8,546
1,616
617
3,395
956
423
1,536 |
8,748 |
11 |
2006 |
195 |
Q1,2,3 /P1,3 |
Q1,2 Asian/<200% povertye
Q1,2 Black
Q1,2 Other
Q3 Asian/<200% povertye
Q3 Black
Q3 Other |
0.665
1.000
0.750
0.600
1.000
0.750
0.500 |
9,464
1,726
698
3,842
959
408
1,831 |
9,654 |
12 |
2007 |
183** |
Q1,2 /P1,4 |
Q1 Asian/<200% povertye
Q1 Hispanic
Q1 Black
Q1 Other
Q2 Asian/<200% povertye
Q2 Hispanic
Q2 Black
Q2 Other |
0.909
1.000
0.750
0.750
0.750
1.000
1.000
1.000
1.000 |
7,319
780
378
324
1,501
943
592
560
2,241 |
7,467 |
a
The oversampling of Hispanic persons and black persons carries over from
the NHIS for each MEPS panel. The oversampling of Asian persons carries
over from the NHIS starting with Panel 12 of MEPS.
b
Needs help in 1 or more activities of daily living (ADLs), such as bathing
and dressing.
c
Low incomes refers to incomes below 200% of the Federal poverty line.
d
Needs help in 10 or more instrumental activities of daily living (IADLs),
such as shopping or paying bills.
e
<200% poverty refers to incomes below 200% of the Federal poverty line.
*
NHIS is National Health Interview Survey; PSU is primary sampling unit.
**
The 2006 NHIS sample redesign included the partitioning of self-representing
PSUs into smaller geographic areas (one or more counties) known as SPSUs
or stratification PSUs. The number of PSUs shown for Panel 12 is comparable
to the numbers associated with the earlier MEPS panels.
Source: Center for Financing,
Access, and Cost Trends, Agency for Healthcare Research and Quality:
Medical Expenditure Panel Survey.
Return to Table of Contents
Table 2a. Number of responding
dwelling units, families, and persons, by year and panel.
MEPS Household Component Point-in-Time Files, 1996-2005
|
Dwelling Units |
Families |
Persons |
1996 Panel 01 |
8,793 |
9,388 |
23,612 |
1997 Combined
Panel 01
Panel 02 |
13,217
8,035
5,182 |
14,147
8,613
5,534 |
35,916
21,411
14,505 |
1998 Combined
Panel 02
Panel 03 |
8,816
4,625
4,191 |
9,597
5,097
4,500 |
24,454
12,908
11,546 |
1999 Combined
Panel 04 |
5,533
5,533 |
5,834
5,834 |
14,974
14,974 |
2000 Combined
Panel 04
Panel 05 |
9,323
4,990
4,333 |
9,927
5,362
4,565 |
25,094
13,546
11,548 |
2001 Combined
Panel 05
Panel 06 |
12,598
3,967
8,631 |
13,393
4,239
9,154 |
34,173
10,589
23,584 |
2002 Combined
Panel 06
Panel 07 |
14,459
7,952
6,507 |
15,482
8,625
6,857 |
39,571
21,620
17,951 |
2003 Combined
Panel 07
Panel 08 |
12,590
5,959
6,631 |
13,426
6,403
7,023 |
34,441
16,413
18,028 |
2004 Combined
Panel 08
Panel 09 |
12,782
6,109
6,673 |
13,664
6,600
7,064 |
34,797
16,530
18,267 |
2005 Combined
Panel 09
Panel 10 |
12,757
6,074
6,683 |
13,582
6,562
7,020 |
34,710
16,514
18,196 |
Source: Center for Financing,
Access, and Cost Trends, Agency for Healthcare Research and Quality, Medical
Expenditures Panel Survey
Return to Table of Contents
Table 2b. Number of
responding dwelling units, families, and persons, by year and panel.
MEPS Household Component Full-Year Files, 1996-2005
|
Dwelling Units |
Families* |
Persons |
1996 Panel 01 |
8,095 |
8,588 |
21,571 |
1997 Combined
Panel 01
Panel 02 |
12,043
7,366
4,677 |
12,986
7,925
5,061 |
32,636
19,622
13,014 |
1998 Combined
Panel 02
Panel 03 |
8,318
4,408
3,910 |
8,920
4,756
4,164 |
22,953
12,260
10,693 |
1999 Combined
Panel 03
Panel 04 |
8,671
3,639
5,032 |
9,278
3,925
5,353 |
23,565
9,979
13,586 |
2000 Combined
Panel 04
Panel 05 |
8,849
4,850
3,999 |
9,437
5,195
4,242 |
23,839
13,170
10,669 |
2001 Combined
Panel 05
Panel 06 |
11,864
3,836
8,028 |
12,732
4,114
8,618 |
32,122
10,298
21,824 |
2002 Combined
Panel 06
Panel 07 |
13,689
7,677
6,012 |
14,712
8,326
6,386 |
37,418
20,890
16,528 |
2003 Combined
Panel 07
Panel 08 |
11,929
5,771
6,158 |
12,742
6,147
6,595 |
32,681
16,000
16,681 |
2004 Combined
Panel 08
Panel 09 |
12,043
5,910
6,133 |
12,917
6,358
6,559 |
32,737
16,058
16,679 |
2005 Combined
Panel 09
Panel 10 |
11,918
5,832
6,086 |
12,680
6,278
6,402 |
32,320
15,904
16,416 |
*
Families defined by variable FAMID[yy] where FMRS1231 = 1.
Source: Center for Financing,
Access, and Cost Trends, Agency for Healthcare Research and Quality,
Medical Expenditures Panel Survey
Return to Table of Contents
Table 3a. Number
of completed person-level interviews by age, sex, race/ethnicity, region,
and MSA status (based on MEPS Point in Time files), MEPS 1996-2005
|
1996 |
1997 |
1998 |
1999
1 |
2000 |
2001 |
2002 |
2003 |
2004 |
2005 |
Total |
23,612 |
35,916 |
24,454 |
14,974 |
25,094 |
34,173 |
39,571 |
34,441 |
34,797 |
34,710 |
Age |
<1 |
360 |
566 |
367 |
200 |
376 |
514 |
558 |
518 |
511 |
519 |
1-17 |
6,532 |
10,298 |
7,162 |
4,219 |
6,987 |
9,459 |
11,453 |
10,188 |
10,082 |
10,118 |
18-24 |
2,133 |
3,194 |
2,204 |
1,308 |
2,230 |
3,125 |
3,610 |
3,179 |
3,268 |
3,243 |
25-44 |
7,230 |
10,530 |
7,003 |
4,449 |
7,228 |
9,773 |
11,151 |
9,560 |
9,650 |
9,432 |
45-64 |
4,707 |
7,224 |
4,925 |
3,149 |
5,400 |
7,440 |
8,524 |
7,286 |
7,501 |
7,676 |
65+ |
2,650 |
4,104 |
2,793 |
1,649 |
2,873 |
3,862 |
4,275 |
3,710 |
3,785 |
3,722 |
Sex |
Male |
11,257 |
17,033 |
11,567 |
7,241 |
12,056 |
16,339 |
18,832 |
16,258 |
16,405 |
16,406 |
Female |
12,355 |
18,883 |
12,887 |
7,733 |
13,038 |
17,834 |
20,739 |
18,183 |
18,392 |
18,304 |
Race/ethnicity |
Hispanic |
4,872 |
7,884 |
5,526 |
3,583 |
6,098 |
8,022 |
9,990 |
9,253 |
9,520 |
9,585 |
Black |
3,183 |
5,301 |
3,689 |
2,173 |
3,626 |
5,041 |
6,022 |
5,392 |
5,349 |
5,716 |
Asian |
690 |
1,004 |
1,006 |
437 |
726 |
1,175 |
1,643 |
1,516 |
1,461 |
1,370 |
Other |
14,867 |
21,727 |
14,233 |
8,781 |
14,644 |
19,935 |
21,916 |
18,280 |
18,467 |
18,039 |
Region |
Northeast |
4,754 |
6,990 |
4,503 |
2,599 |
4,084 |
5,570 |
6,370 |
5,315 |
5,450 |
5,205 |
Midwest |
5,101 |
7,578 |
4,913 |
3,030 |
5,203 |
7,093 |
7,729 |
6,628 |
6,642 |
6,610 |
South |
8,196 |
12,592 |
8,867 |
5,496 |
9,295 |
12,551 |
14,924 |
13,297 |
13,745 |
13,514 |
West |
5,561 |
8,756 |
6,171 |
3,849 |
6,512 |
8,959 |
10,548 |
9,201 |
8,960 |
9,381 |
MSA |
MSA |
18,694 |
28,147 |
19,294 |
11,757 |
19,676 |
27,265 |
31,625 |
27,393 |
27,710 |
28,723 |
Non MSA |
4,918 |
7,769 |
5,160 |
3,217 |
5,418 |
6,908 |
7,946 |
7,048 |
7,087 |
5,987 |
1
The 1999 PIT contains only one of the two panels.
Source: Center for Financing,
Access, and Cost Trends, Agency for Healthcare Research and Quality,
Medical Expenditures Panel Survey
Return to Table of Contents
Table 3b. Number of
completed person-level interviews by age, sex, race/ethnicity, region, and
MSA status (based on MEPS Full Year Files): MEPS-HC 1996-2005
|
1996 |
1997 |
1998 |
1999 |
2000 |
2001 |
2002 |
2003 |
2004 |
2005 |
Total |
21,571 |
32,636 |
22,953 |
23,565 |
23,839 |
32,122 |
37,418 |
32,681 |
32,737 |
32,320 |
Age |
<1 |
321 |
406 |
270 |
266 |
291 |
383 |
455 |
431 |
426 |
424 |
1-17 |
5,965 |
9,330 |
6,569 |
6,551 |
6,595 |
8,774 |
10,599 |
9,512 |
9,353 |
9,217 |
18-24 |
1,884 |
2,919 |
2,154 |
2,095 |
2,119 |
3,010 |
3,462 |
3,080 |
3,072 |
3,041 |
25-44 |
6,478 |
9,332 |
6,412 |
6,727 |
6,680 |
8,869 |
10,273 |
8,877 |
8,819 |
8,559 |
45-64 |
4,374 |
6,661 |
4,753 |
5,107 |
5,244 |
7,228 |
8,296 |
7,053 |
7,303 |
7,362 |
65+ |
2,549 |
3,988 |
2,795 |
2,819 |
2,910 |
3,858 |
4,333 |
3,728 |
3,764 |
3,717 |
Sex |
Male |
10,289 |
15,443 |
10,859 |
11,287 |
11,445 |
15,369 |
17,796 |
15,413 |
15,439 |
15,251 |
Female |
11,282 |
17,193 |
12,094 |
12,278 |
12,394 |
16,753 |
19,622 |
17,268 |
17,298 |
17,069 |
Race/ethnicity |
Hispanic |
4,638 |
7,542 |
5,585 |
5,852 |
5,936 |
7,637 |
9,427 |
8,866 |
9,022 |
8,990 |
NonHispanic |
16,933 |
25,094 |
17,368 |
17,713 |
17,903 |
24,485 |
27,991 |
23,815 |
23,715 |
23,330 |
Black |
2,907 |
4,815 |
3,430 |
3,239 |
3,471 |
4,699 |
5,570 |
5,094 |
4,991 |
5,260 |
Asian |
582 |
829 |
597 |
635 |
602 |
987 |
1,304 |
1,349 |
1,311 |
1,227 |
Other |
13,444 |
19,450 |
13,341 |
13,839 |
13,830 |
18,799 |
21,117 |
17,372 |
17,413 |
16,843 |
Region |
Northeast |
4,275 |
6,278 |
4,159 |
4,031 |
3,746 |
5,063 |
5,840 |
4,843 |
4,912 |
4,734 |
Midwest |
4,668 |
6,834 |
4,537 |
4,657 |
4,951 |
6,679 |
7,377 |
6,365 |
6,224 |
6,154 |
South |
7,494 |
11,446 |
8,340 |
8,764 |
8,901 |
12,003 |
14,212 |
12,704 |
13,130 |
12,656 |
West |
5,134 |
8,078 |
5,917 |
6,113 |
6,241 |
8,377 |
9,989 |
8,769 |
8,471 |
8,776 |
MSA Status |
MSA |
16,791 |
25,185 |
17,897 |
18,325 |
18,556 |
25,451 |
29,723 |
25,827 |
26,777 |
26,572 |
Non MSA |
4,592 |
7,127 |
4,817 |
5,038 |
5,283 |
6,671 |
7,695 |
6,854 |
5,960 |
5,748 |
Missing |
188 |
324 |
239 |
202 |
- |
- |
- |
- |
- |
- |
Source: Center for Financing,
Access, and Cost Trends, Agency for Healthcare Research and Quality,
Medical Expenditures Panel Survey
Return to Table of Contents
Table 4. Example sample
size and response rates - Full Year (Panel 7, Rounds 1-3 combined with Panel 6,
Rounds 3-5): MEPS 2002
|
Panel 6 |
Panel 7 |
A. Percentage of NHIS households designated for use in MEPS (those initially characterized as responding) |
89.9% |
89.7% |
B. Number of households sampled from the NHIS |
10,704 |
8,132 |
C. Number of Households sampled from the NHIS and eligible and fielded for MEPS |
10,651 |
8,083 |
D. Round 1 – Number of RUs *
eligible for interviewing |
11,556 |
8,710 |
E. Round 1 – Number of RUs with completed interviews |
9,377 |
7,008 |
F. Round 2 – Number of RUs eligible for interviewing |
9,666 |
7,197 |
G. Round 2 – Number of RUs with completed interviews |
9,222 |
6,802 |
H. Round 3 – Number of RUs eligible for interviewing |
9,380 |
6,937 |
I. Round 3 – Number of RUs with completed interviews |
9,001 |
6,673 |
J. Round 4 – Number of RUs eligible for interviewing |
9,117 |
NA |
K. Round 4 – Number of RUs with completed interviews |
8,843 |
NA |
L. Round 5 – Number of RUs eligible for interviewing |
8,892 |
NA |
M. Round 5 – Number of RUs with completed interviews |
8,781 |
NA |
P6: A x (E/D) x (G/F) x (I/H) x (K/J) x (M/L)
P7: A x (E/D) x (G/F) x (I/H) |
64.0% (Panel 6 through Round 5) |
65.6% (Panel 7 through Round 3) |
Overall combined response rate:
0.55 x P6 response rate +
0.45 x P7 response rate |
64.7% |
*
RU is Reporting Unit.
Source: Center for Financing,
Access, and Cost Trends, Agency for Healthcare Research and Quality:
Medical Expenditure Panel Survey.
Return to Table of Contents
Table 5. MEPS Individual
Panel and Combined Annual Response Rates
Response Rate (%) |
Calendar Year |
Year 2 Panel |
Year 1 Panel |
Adjustment Factors (Year 2, Year 1) |
Combined Overall |
1996 |
na |
70.66 |
na |
71.0 |
1997 |
64.06 |
69.49 |
na |
66.8 |
1998 |
65.72 |
69.02 |
na |
67.4 |
1999 |
65.88 |
66.14 |
na |
66.0 |
2000 |
63.68 |
68.27 |
(0.55, 0.45) |
65.7 |
2001 |
65.42 |
66.81 |
(0.33, 0.67) |
66.3 |
2002 |
63.99 |
65.60 |
(0.55, 0.45) |
64.7 |
2003 |
62.91 |
65.93 |
(0.49, 0.51) |
64.5 |
2004 |
62.66 |
63.50 |
(0.49, 0.51) |
63.1 |
Source: Center for Financing,
Access, and Cost Trends, Agency for Healthcare Research and Quality:
Medical Expenditure Panel Survey
− na is ‘not applicable’.
Return to Table of Contents
Suggested Citation:
Ezzati-Rice, TM, Rohde, F, Greenblatt, J,
Sample Design of the Medical Expenditure Panel Survey Household Component,
1998–2007. Methodology Report No. 22. March 2008. Agency
for Healthcare Research and Quality, Rockville, MD. http://www.meps.ahrq.gov/mepsweb/data_files/publications/mr22/mr22.shtml |