MEPS HC-001: 1996 Panel Round 1 Population Characteristics
March 1997
Agency for Healthcare Research and Quality
Center for Financing, Access, and Cost Trends
540 Gaither Road
Rockville, MD 20850
(301) 427-1406
Table of Contents
A. Data Use Agreement
B. Background
1. Household Component
2. Medical Provider Component
3. Insurance Component
4. Nursing Home Component
5. Survey Management
C. Technical and Programming Information
1. General Information
2. Data File Information
2.1 Codebook Structure
2.2 Reserved Codes
2.3 Codebook Format
2.4 Variable Naming
2.5 File Contents
2.5.1 Survey Administration Variables
2.5.2 Demographic Variables
2.5.3 Employment Variables
2.5.4. Health Status Variables
2.5.5 Health Insurance Variables
3. Survey Sample Information
3.1 Sample Design and Response Rates
3.2 Sample Weights and Variance Estimation
3.2.1 Sample Weighting
3.2.2 Variance Estimation
4. Programming Information
D. Codebook
E. Crosswalk of Variables to Variable Source
F. Appendice
1. Household Survey Sample Design Repor
2. Household Survey Design and Methods Repor
A. Data Use Agreement
Individual identifiers have been removed from the micro-data contained in these data files.
Nevertheless, under sections 308 (d) and 903 (c) of the Public Health Service Act (42 U.S.C.
242m and 42 U.S.C. 299 a-1), data collected by the Agency for Health Care Policy and
Research (AHCPR) and /or the National Center for Health Statistics (NCHS) may not be used
for any purpose other than for the purpose for which it was supplied; any effort to determine
the identity of any reported persons or establishments, is prohibited by law.
Therefore in accordance with the above referenced Federal Statute, it is understood that:
1. No one is to use the data in this data set in any way except for statistical reporting and
analysis; and
2. If the identity of any person or establishment should be discovered inadvertently, then (a)
no use will be made of this knowledge, (b) The Director Office of Management AHCPR will
be advised of this incident, (c) the information that would identify any individual or
establishment will be safeguarded or destroyed, as requested by AHCPR, and (d) no one else
will be informed of the discovered identity.
3. No one will attempt to link this data set with individually identifiable records from any data
sets other than Medical Expenditure Panel Survey or the National Health Interview Survey.
By using this data you signify your agreement to comply with the above stated statutorily
based requirements with the knowledge that deliberately making a false statement in any matter
within the jurisdiction of any department or agency of the Federal Government violates 18
U.S.C. 1001 and is punishable by a fine of up to $10,000 or up to 5 years in prison.
The Agency for Health Care Policy and Research requests that users cite AHCPR and the
Medical Expenditure Panel Survey as the data source in any publications or research based
upon these data.
Return To Table Of Contents
B. Background
This documentation describes one in a series of public use files from the Medical Expenditure
Panel Survey (MEPS). The survey provides a new and extensive data set on the use of health
services and health care in the United States.
The MEPS is the third in a series of national probability surveys conducted by AHCPR on the
financing and utilization of medical care in the United States. The National Medical Care
Expenditure Survey (NMCES, also known as NMES-1) was conducted in 1977, the National
Medical Expenditure Survey (NMES-2) in 1987. Beginning in 1996, the MEPS continues this
series with design enhancements and efficiencies that provide a more current data resource to
capture the changing dynamics of the health care delivery and insurance system.
The design efficiencies incorporated into the MEPS are in accordance with the Department of
Health and Human Services (DHHS) Survey Integration Plan of June 1995, which focused on
consolidating the Department's surveys, achieving cost efficiencies, reducing respondent
burden, and enhancing analytical capacities. To accommodate these goals, new design features
in the current MEPS include linkage with the National Health Interview Survey (NHIS), from
which the sampling frame for the MEPS HC is drawn, and a change to continuous longitudinal
data collection for core survey components. The MEPS HC augments the NHIS by continuing
to collect data on a subset of NHIS respondents and links this information to data collected
from the respondents' medical providers, employers, and insurance providers.
The Medical Expenditure Panel Survey (MEPS) is conducted to provide nationally
representative estimates of health care utilization, expenditures, sources of payment, and
insurance coverage for the U.S. civilian noninstitutionalized population. The MEPS also
includes a nationally representative survey of nursing homes and their residents, as well as
medical providers and establishments. The MEPS is co-sponsored by the Agency for Health
Care Policy and Research (AHCPR) and the National Center for Health Statistics (NCHS).
The MEPS comprises four component surveys: the Household Component (HC), the Medical
Provider Component (MPC), the Insurance Component (IC), and the Nursing Home
Component (NHC). The HC serves as the core survey from which the MPC sample and part
of the IC sample are based. These are supplemented by the NHC. Together these surveys
yield comprehensive data that provide national estimates of the level and distribution of health
care use and expenditures, support health services research, and assess health care policy
implications.
Return To Table Of Contents
1. Household Component
The MEPS HC is a nationally representative survey of the U.S. civilian noninstitutionalized
population which collects medical expenditure data at both the person and household levels.
The focus of the MEPS HC is to collect detailed data on demographic characteristics, health
conditions, health status, use of medical care services, charges and payments, access to care,
satisfaction with care, health insurance coverage, income, and employment.
The MEPS HC collects data through an overlapping panel design. In this design, data are
collected through a preliminary contact followed by a series of six rounds of interviews over a
two-and-a-half year period. Two calendar years of medical expenditures and utilization are
collected from each household and are captured using computer-assisted personal interviewing
(CAPI) technology. This series of data collection rounds is launched again each subsequent
year on a new sample of households to provide overlapping panels of survey data, which when
combined with other ongoing panels will provide continuous and current estimates of health
care expenditures.
The sampling frame for the MEPS HC is drawn from respondents to the NHIS, conducted by
the NCHS. The NHIS provides a nationally representative sample of the U.S. civilian noninstitutionalized population and reflects an oversampling of Hispanics and blacks. A
subsample of 10,500 households was drawn from the NHIS sampling frame for the initial 1996
MEPS HC panel. Every five years the MEPS HC sample size is increased and targets
oversampling of policy-relevant population subgroups, beginning with the 1997 panel. Initially
these subgroups will include: 1) adults with functional impairments; 2) children with
limitations; 3) individuals between the ages of 18-64 predicted to have high levels of medical
expenditures; and 4) individuals with family incomes less than 200 percent of the poverty
level.
Return To Table Of Contents
2. Medical Provider Component
To supplement and validate information on medical care events reported in the MEPS HC, the
MPC of the MEPS contacts those medical providers identified by MEPS household
respondents. The MEPS MPC sample includes all reported hospitals, hospital physicians,
home health agencies, and pharmacies. Also included in the MPC are all office-based
physicians providing care for HC respondents receiving Medicaid, office-based physicians
associated with a 75 percent sample of households receiving care through an HMO or managed
care plan, and a 25 percent sample of remaining households.
The 1996 sample is projected to provide data from approximately 2,700 hospitals, 12,400
office-based physicians, 7,000 separately billing doctors, and 500 home health providers. Data
are collected in the MPC on medical and financial characteristics of medical events reported by
HC respondents, including diagnoses (ICD-9s and DSM-IVs), physician procedure codes
(CPT-4s), inpatient stay codes (DRGs), charges, payments, and the reasons for any difference
between charges and payments. The MPC is conducted through telephone interviews and
mailed survey materials.
Return To Table Of Contents
3. Insurance Component
The MEPS IC collects data on health insurance plans obtained through employers, unions, or
other private health insurance sources. Data obtained in the MEPS IC include the number and
types of private insurance plans offered, benefits associated with these plans, premiums,
contributions by employer vs. employee, eligibility requirements, and employer
characteristics.
Establishments participating in the MEPS IC are selected through four sampling frames: 1) a
list of employers or other insurance providers identified by respondents in the MEPS HC who
report having private health insurance at the Round 1 interview; 2) a Census Bureau list frame
of private sector business establishments; 3) the Census Bureau's Census of Governments; and
4) an Internal Revenue Service list of the self-employed.
To provide an integrated picture of health insurance,
data collected from the first sampling frame (i.e., employers and insurance
providers) are linked back to data provided by the HC
respondents. Data from the other three sampling frames are collected to provide
annual national and state estimates on the supply of private health insurance
available to American
workers and to evaluate policy issues pertaining to health insurance.
Designed as an annual panel survey, each year the MEPS IC sample includes approximately
7,000 establishments identified through the MEPS HC, 27,000 identified through the business
establishments list frame, 1,900 governments from the Census of Governments, and 1,000 self-employed persons. Data are collected from the selected organizations through a prescreening
telephone interview, a mailed questionnaire, and a nonresponse telephone follow up.
Return To Table Of Contents
4. Nursing Home Component
The 1996 NHC of the MEPS is a survey of nursing homes and persons resident in or admitted
to nursing homes at any time during calendar year 1996. The MEPS NHC gathers information
on the demographic characteristics, residence history, health and functional status, use of
services, use of prescription medications, and health care expenditures of nursing home
residents. Nursing home administrators and designated staff also provide information on
facility size, ownership, certification status, services provided, revenues and expenses, and
other facility characteristics. A community questionnaire obtains data from next of kin or
other knowledgeable persons in the community on income, assets, family relationships, and
care-giving information for the sampled nursing home resident. Under the DHHS Survey
Integration Plan, the MEPS NHC is designed to be conducted every five years.
The 1996 NHC sample was selected using a two-stage stratified probability design. The first
stage was used to select facilities; the second stage sampled facility residents, selecting from both
persons in residence on January 1, 1996, and those admitted between January 1 and December
31, 1996. The sample frame for facilities was derived from the National Health Provider
Inventory, which is updated periodically by NCHS. MEPS NHC data were collected in person in
three rounds of data collection using the CAPI system over a year-and-a-half period. Community
data were collected by telephone using computer-assisted survey interviewing (CASI)
technology. At the end of data collection, the sample will consist of approximately 800
responding facilities, 3,100 January 1 residents, and approximately 2,200 eligible admissions.
Return To Table Of Contents
5. Survey Management
MEPS data are collected under the authority of the Public Health Service Act and are being
edited and published in accordance with the confidentiality provisions within this act and those
of the Privacy Act. Consultation and technical assistance are received from the NCHS.
Data collection is conducted under contract by Westat, Inc., Rockville, MD; the National
Opinion Research Center at the University of Chicago; and through an interagency agreement
with Bureau of the Census. Technical consultation is provided by Medstat, Inc., Boston, MA.
Data processing support is provided under contract by Social & Scientific Systems, Inc.,
Bethesda, MD.
As soon as data collection and editing are completed, the MEPS survey data are released to the
public in staged releases of summary reports and micro data files. Summary reports are made
available as hard copy documents and as electronic files. Micro data files are released on CD-ROM and/or electronic files. Hard copy documents and CD-ROMs will be available free of
charge through the AHCPR Publications Clearinghouse at 1-(800) 358-9295, or can be purchased
from the National Technical Information Services (NTIS) at (703) 487-4650. The mailing
address is 5285 Port Royal Road, Springfield, VA, 22161. If calling from outside the U.S., dial
(410) 381-3150. Selected electronic files will be available on the Internet in the MEPS section of
the AHCPR home page: http://www.ahcpr.gov
Additional information on MEPS is available from the MEPS project manager or the MEPS
public use data manager at the Center for Cost and Financing Studies, Agency for Health Care
Policy and Research, 540 Gaither Road, Rockville, MD 20850 (301/427-1406).
Return To Table Of Contents
C. Technical and Programming Information
1. General Information
This documentation describes the first public use data file to be released from the 1996 Medical
Expenditure Panel Survey Household Component (MEPS HC). Released as an ASCII file (with
related SAS programming statements), this public use file provides information collected on a
nationally representative sample of the civilian non-institutionalized population of the United
States during the first part of 1996. This file consists of data obtained in Round 1 of the survey
and contains variables pertaining to survey administration, demographics, employment, health
status, and health insurance. The data are being released prior to final data cleaning and editing
in order to provide the research and policy community prompt access to MEPS data. Analysts
should consider this data as preliminary as they have not been subject to the same level of quality
control procedures which are usually performed on products of this type. Future plans include a
capability of linking this data file to the 1995 National Health Interview Survey (NHIS).
Information on merging to NHIS data files will be provided in future data releases.
The following documentation offers a brief overview of the types and levels of data provided, the
content and structure of the files and the codebook, and programming information. It contains
the following sections:
Data File Information
Survey Sample Information
Programming information
Codebook
Variable/Questionnaire Crosswalk
Detailed information on sample design and data collection methods can be found in Appendices
1 and 2, Sample Design of the 1996 Medical Expenditure Panel Survey Household Component
and Design and Methods of the Medical Expenditure Panel Survey Household Component. The
individual questionnaires used in Round 1 to collect the information on this file are also included
(see README2.TXT file).
Return To Table Of Contents
2. Data File Information
This public use dataset contains variable and frequency distributions for a total of 24,676 persons
from the MEPS Household Component of the 1996 Panel of the Medical Panel Expenditure
Survey. This count includes all household survey respondents who resided in eligible responding
households. Of these persons 23,612 were assigned a positive person level weight. For each
variable both weighted and unweighted frequencies are provided. In conjunction with the weight
variable (WGTSP1) provided on this file, data for these persons can be used to make estimates
for the civilian noninstitutionalized U. S. population as of the first half of 1996.
The records on this file can be linked to all MEPS public use data sets by the sample person
identifier (DUPERSID).
Return To Table Of Contents
2.1 Codebook Structure
The codebook and data file sequence lists variables in the following order:
Unique person identifiers
Demographic variables
Employment variables
Health Status variables
Health Insurance variables
Weight and variance estimation variables
Return To Table Of Contents
2.2 Reserved Codes
The following reserve code values are used:
VALUE DEFINITION
-1
INAPPLICABLE Question was not asked due to skip pattern
-6
INAPPLICABLE Question was not asked due to person being under age 5
-7
REFUSED Question was asked and respondent refused to answer question.
-8
DK Question was asked and respondent did not know answer
-9
NOT ASCERTAINED Interviewer did not record the data
Return To Table Of Contents
2.3 Codebook Format
This codebook describes an ASCII data set and provides the following programing identifiers for
each variable:
IDENTIFIER |
DESCRIPTION |
Name |
Variable name (maximum of 8
characters) |
Description |
Variable descriptor (maximum 40
characters) |
Format |
Number of bytes |
Type |
Type of data: numeric (indicated by
NUM) or character (indicated by CHAR) |
Start |
Beginning column position of
variable in record |
End |
Ending column position of variable
in record |
Return To Table Of Contents
2.4 Variable Naming
In general, variable names reflect the content of the variable, with an 8 character limitation.
Edited variables end in an X, and are so noted in the variable label. Variables contained in this
delivery were derived either from the questionnaire itself or from the CAPI. For variables
coming directly from one question, the question number will appear in the question number field.
For
variables which were constructed from multiple questions, the question numbers can be found in
Section E - Variable/Question Crosswalk. Variables which were derived from CAPI have no
corresponding question.
Return To Table Of Contents
2.5 File Contents
2.5.1 Survey Administration Variables
The survey administration variables contain information related to conducting the interview,
household and family composition, and person-level status codes. Data for the survey
administration variables were derived from the sampling process, the CAPI programs, or were
computed based on information provided by the respondent in the re-enumeration section of the
questionnaire.
Detailed information on the sampling process, including the link to the National Health Interview
Survey, is found in Section 3 and Appendix 1.
Dwelling Units, Reporting Units, Families, and Persons
The definitions of Dwelling Units (DUs) and Group Quarters in the MEPS Household Survey are
generally consistent with the definitions employed for the National Health Interview Survey. The
dwelling unit ID (DUID) is a five-digit random ID number assigned after the case was sampled
for MEPS. The person number (PID) uniquely identifies all persons within the dwelling unit.
The variable DUPERSID is the combination of the variables DUID and PID.
A Reporting Unit (RU) is a person or group of persons in the sampled dwelling unit who are
related by blood, marriage, adoption or other family association, and who are to be interviewed
as a group in MEPS. Thus, the RU serves chiefly as a family-based "survey operations" unit
rather than an analytic unit. Members of each RU within the DU are identified by the variable
RULETTER. Regardless of the legal status of their association, two persons living together as a
"family" unit were treated as a single reporting unit if they chose to be so identified. Examples of
different types of reporting units are:
1. A married daughter and her husband living with her parents in the same dwelling constitute a single reporting unit.
2. A husband and wife and their unmarried daughter, age 18, who is living away
from
home at college constitute two reporting units.
3. Three unrelated persons living in the same dwelling unit would each constitute a
distinct reporting unit, three reporting units in all.
Unmarried college students under 24 years of age who usually live in the sampled household, but
were living away from home and going to school at the time of the Round 1 MEPS interview,
were treated as a Reporting Unit separate from that of their parents for the purpose of data
collection. The variable RUSIZE indicates the number of persons in each RU, treating students
as RUs separate from their parents. However, for many analytic objectives, the student reporting
units would be combined with their parents' reporting unit, treating the combined entity as a
single family.
The variable FAMID1 identifies a family (i.e. persons related to one another by blood, marriage,
adoption or foster care). FAMID1 differs from RU in that it combines student reporting units
with their parent reporting unit. The variable UNITSIZE indicates the RU size when students are
combined back into their associated parent RUs for analytical purposes. Family-level analyses
should use the UNITSIZE variable.
Note: the variables RUSIZE and UNITSIZE exclude ineligible (ELIGRND1=2) persons;
analysts should exclude such persons from all family-level analyses.
There are several other variables that characterize the reporting unit. RUCLASS indicates
whether the RU was fielded as a standard, new or student RU. The variables REGION and MSA
indicate the geographic location of the dwelling unit. REGION is coded according to the Census
regions, and MSA reflects the June 30, 1996 definition of metropolitan statistical areas.
Administration Dates
A number of variables containing date information are included on each record. These variables
were derived in CAPI at the family level and then moved to the person-level. The day on which
the interview was conducted is provided in month, day and year format (RUENDMM,
RUENDDD, and RUENDYY). All Round 1 interviews were conducted between March and
August of 1996.
Reference Period Dates
The beginning and ending reference period dates are included for each person (BEGREFMM,
BEGREFDD, BEGREFYY, ENDREFMM, ENDREFDD, and ENDREFYY). The reference
period is the period of time for which data were collected in Round 1 and it was determined
during the interview for each person by the CAPI program. The reference period for most
persons identified at NHIS began on January 1, 1996 and ended on the date of the Round 1
interview. In some instances the beginning of the reference period was set by CAPI to be later
than January 1 based on the date the person entered the RU. The end date of the reference period
is prior to the date of the interview if the person was deceased during the round, left the RU, or
was institutionalized prior to the Round 1 interview.
Return To Table Of Contents
Respondent Identifiers
The respondent for the Round 1 interview is identified in RND1RESP. Only one respondent is
identified for each RU. In instances where the interview was completed in more than one
session, only the first respondent is indicated. The variable PROXY indicates whether the
respondent for the RU was a proxy or another RU member.
Person Status
The variables INSCOPE, KEYNESS, ELIGRND1, and PSTATUS describe the overall status of
each person for the Round 1 data collection. These variables are set based on sampling
information and responses provided in the re-enumeration section.
Through the re-enumeration section of the Round 1 questionnaire, each member of a reporting
unit was classified as "key" or "non-key", "in-scope" or "out-of-scope", and "eligible" or
"ineligible" for MEPS data collection. To be included in the set of persons used in the derivation
of MEPS person level estimates, a person had to be a member of the civilian non-institutionalized population for some period of time in the Round 1 reference period. Because a
person's eligibility for the survey might have changed since the NHIS interview, a sampling re-enumeration of household membership was conducted at the start of the MEPS Round 1
interview (and at the start of each subsequent interview). Only persons who were "in-scope",
"key", and "eligible for data collection" were assigned person level weights and thus are to be
used in the derivation of person level national estimates from the MEPS.
In-Scope
A person was classified as in-scope (INSCOPE) if he or she was a member of the U.S. civilian,
non-institutionalized population at some time during the Round 1 interview.
Return To Table Of Contents
Keyness
The term "keyness" is related to an individual's chance of being included in MEPS. A person is
key if that person is appropriately linked to the set of 1995 NHIS sampled households designated
for inclusion in MEPS. Specifically, a key person either was a member of an NHIS household at
the time of the NHIS interview, or became a member of such a household after being out-of-scope prior to joining that household (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, who happened to have become a
member of a MEPS reporting unit by the time of the MEPS round 1 interview. MEPS data, (e.g.,
utilization and income) were collected for the period of time a non-key person was part of the
sampled unit to permit family level analyses. However, non-key persons who leave a sample
household would not be recontacted for subsequent interviews. Non-key individuals are not part
of the target sample used to obtain person level national estimates.
It should be pointed out that a person may be key even though not part of the civilian, non-institutionalized portion of the U.S. population. For example, a person in the military may be
living with his or her civilian spouse and children in a household sampled for the 1995 NHIS.
The person in the military would be considered a key person for MEPS. However, such a person
would not receive a person-level sample weight so long as he or she was in the military. All key
persons who participated in the first round of the 1996 MEPS received a person level sample
weight except those who were in the military. The variable indicating "keyness" is KEYNESS.
Eligibility
The eligibility of a person for MEPS pertains to whether or not data were to be collected for that
person. All key, in-scope persons of a sampled RU were eligible for data collection. The only
non-key persons eligible for data collection were those who happened to be living in the same
RU as one or more key persons, and their eligibility continued only for the time that they were
living with a key person. The only out-of-scope persons eligible for data collection were those
who were living with key in-scope persons, again only for the time they were living with a key
person. Only military persons meet this description. A person was considered eligible if they
were eligible at any time during Round 1. The variable indicating "eligibility" is ELIGRND1,
where 1 is coded for persons eligible for data collection for at least a portion of the Round 1
reference period, and 2 is coded for persons not eligible for data collection at any time during the
first round reference period.
Person Disposition Status
Person disposition status (PSTATUS) is a variable which indicates a person's status with respect
to response and eligibility at the time of the interview. It describes the reasons for either going
forward with data collection or terminating data collection for each person in the MEPS. Using
PSTATUS, one could identify persons who moved during the reference period, died, were born,
institutionalized or who were in the military.
The following codes specify the value labels for the PSTATUS codes:
0 Incorrectly listed in RU at NHIS
11 Person in household, not full-time active-duty military
12 Person in household, full-time active-duty military, out-of-scope
13 Moved, full-time student living away from home
14 Person in original household, full-time active duty, in-scope for part of Round
31 Deceased in reference period
32 Institutionalized in health care facility in reference period
33 Institutionalized in non-health care facility in reference period
34 Moved outside U.S. in reference period
35 Moved, full-time active duty in reference period
36 Institutionalized (type unknown) in reference period
41 Moved within U.S. in reference period
42 Person who joined RU in reference period
51 Newborn in reference period
61 Deceased prior to reference period
62 Institutionalized prior to reference period
63 Move prior to reference period
64 Full-time military at military facility, moved prior to reference period
81 Moved, full-time student, no response, data imputed
Return To Table Of Contents
Implications of these Characterizations of Sampled Persons
Person level weights were created for responding persons who were key, eligible, and in-scope.
The only "non-respondents" to receive person level weights were college aged students living
away from home during the school year who were key, eligible, and in-scope and whose parents
or guardians were MEPS Round 1 respondents. In general, data were imputed for these college
students.
Persons who were non-key and/or out-of-scope but who were eligible for data collection and
were part of a responding family (reporting unit) are considered part of the family for analysis
purposes and have an associated family level weight. In a small number of these cases, there are
missing data. Future data releases will contain imputed data for these cases.
Thus, persons in the military who are members of an RU will have received a family level weight
but not a person level weight. Persons who join an RU who were previously in-scope and thus
had their chance for participation in MEPS associated with a household not selected for the 1995
NHIS also will have received a family level weight but not a person level weight. Persons who
joined an RU as civilians and who were out-of-scope prior to joining the RU and thus did not
have a chance for selection for the 1995 NHIS will have received both a family level weight and
a person level weight.
Return To Table Of Contents
2.5.2 Demographic Variables
These variables provide information about the demographic characteristics of each person. As
noted below, some variables have edited and imputed values.
Sex
The sex of each RU member (SEX), as collected during the NHIS interview, was verified and, if
necessary, corrected during the Round 1 MEPS interview. The sex of each new RU member
(persons who were not members of the RU at the time of the NHIS interview) was also obtained
during the Round 1 interview. When sex was not available from the NHIS interview and was not
ascertained during the Round 1 interview, it was assigned in the following way. If available in
the edited/imputed NHIS data, sex was assigned from that source. If not available from that
source, the first name of the person was used to assign sex, if obvious. If the sex of the
individual was still not determined, sex was randomly assigned. Assignment was made in this
way for nine persons: for six nonresponding students, two siblings, and one newborn.
Age
The date of birth and age of each RU member were asked during the Round 1 MEPS interview
(DOBMM, DOBYY, AGE1X). When age was not reported, an age value was obtained through
the following steps. If available, age was obtained from the originally collected 1995 NHIS data;
if not available from the data as collected, age as edited/imputed for NHIS was used. If the NHIS
age data were used, one year was added to the NHIS age to obtain the MEPS age, accounting for
the time elapsed between the two surveys. If not available from that source, the mid-point of the
age interval of the RU member estimated by the MEPS Round 1 respondent was used (see
question RE57B). If not available there, age was imputed as the mid-point of the interval
estimated by the Round 1 MEPS interviewer (see question RE57C). If not available from that
source, demographic information and relationships of all members of the RU were reviewed to
determine a reasonable value to assign for age. For purposes of confidentiality, the variable
AGE1X was top coded at 90 years.
Race, Race/Ethnicity, Hispanic Ethnicity, and Hispanic Ethnicity Group
Race (RACEX) and Hispanic ethnicity (HISPANX) were asked for each RU member during the
Round 1 MEPS interview. When race and/or ethnicity was not reported in the Round 1
interview, values for these variables were obtained in the following sequence. When available,
they were obtained from the originally collected NHIS data. If not available from this source,
edited/imputed NHIS data were used. If not ascertained from the edited/imputed NHIS data, the
race, ethnicity, and relationships of all members of the RU were used to impute race or ethnicity.
This approach was used to resolve a residual group of 20 cases, 14 of which were missing both
race and ethnicity and six of which were missing only race. The variable RACETHNX
indicating both race and ethnicity (e.g., with categories such as "Hispanic" and "black but not
Hispanic") reflects the imputations done for RACEX and HISPANX. The specific Hispanic
ethnicity group is given in the unedited variable HISPCAT.
Return To Table Of Contents
Student Status and Educational Attainment
The variable FTSTUD1 indicates whether the person is currently a full-time student. This
variable has valid values only for persons between the ages of 17 - 23 who are not living away at
school and not in grades 1-12. Completed years of education is given in the variable EDUCYR1.
Information was obtained from questions RE 103-105. Children who are 5 years of age or older
and who never attended school were coded as 0; children under the age of 5 years were coded as
-6 "under age 5-inapplicable" regardless of whether they attended school or not.
The variable indicating highest degree (HIGHDEG1) was obtained from two questions: high
school diploma (RE 104) and highest degree (RE 105). Persons under 16 years of age were
coded as 8 "inapplicable". In cases where the response to the highest degree question was "no
degree" and highest grade was 13 through 17, the variable was coded as 3 "high school diploma".
If highest grade completed for those with a "no degree" response was "refused" or "don't know",
the variable was coded as 1 "no degree".
Marital Status and Spouse ID
Marital status was collected in RE13 and RE97 and reported as MARRY1X. Persons under the
age of 16 were coded as 6 "under 16 - inapplicable". Those whose marital status changed during
the round were identified through edits where inconsistencies were detected between the reported
marital status of partners. In instances where there were discrepancies, other person-level
variables were reviewed to determine the edited marital status. For example, when one partner
was reported as married and the other partner reported as widowed, the data were reviewed to
determine if one partner should be coded as 8 "widowed in Round 1".
The spouse's ID is reported in SPOUSID1. This is the PID of the person identified as the spouse.
If no spouse was identified in the household, the value was coded as 995 "no spouse in
household". Those with unknown marital status are coded as 996. Persons under the age of 16
are coded as 997 "inapplicable".
The SPOUSIN1 variable indicates whether a person's spouse is present in the RU. If the person
has no spouse in the household, the value was coded as 2. For persons under the age of 16 the
value was coded as 3.
Return To Table Of Contents
Military Service and Service Era
The variable DIDSERVE indicates if the person ever served in the Armed Forces. Persons under
the age of 16 were coded as 3 "under 16-inapplicable". Persons currently on full time active duty
status are identified in the variable ACTDUTY1. Those under 16 years of age were coded as 3
"under 16-inapplicable" and those over the age of 59 were coded as 4 "over 59-inapplicable".
Service in specific eras was coded according to the responses to RE 96 in the variables VETWW
(either World War I or World War II), VETKOR (Korean War era), VETVIET (Vietnam War
era), VETPVIET (Post-Vietnam War era), VETOTH (other service era). Those under the age of
16 were coded as 3 "under 16-inapplicable" and those who never served in the military were
coded as 4 "never in military".
Relationship to the Reference Person within Reporting Units
The variable REFREL1X indicates the relationship of each individual to the reference person of
the reporting unit (RU). For the reference person, this variable is coded to the value "self"; for all
other persons in the RU, relationship to the reference person is indicated by codes representing
"husband/spouse", "wife/spouse", "son", "daughter", "female partner", "male partner" etc. If the
relationship of an individual to the reference person was not ascertained during the Round 1
interview, relationships between other RU members were used where possible to assign a
relationship to the reference person. If Round 1 data were not sufficient to identify the
relationship of an individual to the reference person, relationship variables from the 1995 NHIS
data were used to assign a relationship. A code of 91, meaning "other related", was used to
indicate infrequently observed relationship descriptions such as "mother of partner". In the event
that a meaningful value could not be determined, the relationship variable was assigned a missing
value code. For 340 cases, the relationship codes were changed to agree with reported marital
status.
Return To Table Of Contents
2.5.3 Employment Variables
Round 1 employment edits focused on the construction of a person-level hourly wage variable
and on logical editing of the employment section health insurance variables. Preliminary logical
editing of several other employment variables was also carried out. The primary focus of these
edits was to assure the correct assignment of "missing" and "Inapplicable" values to specific
observations. Very broadly, missing values could occur when an individual had been asked a
question during the interview, but gave a "don't know" answer or refused to respond to the
question. "Inapplicable" codes were assigned when an individual was never asked a particular
question because of the skip patterns in the survey (e.g., individuals who indicated that they were
not employed were not asked questions about their wages or hours worked).
Employment Status and Self-Employment
Before a logically-edited hourly wage variable could be constructed for persons holding a current
main job who were not self-employed, it was necessary to construct logically-edited variables for
employment status (EMPST1) and self-employment (SELFCM). All persons age 16 and over
were asked if they were currently employed for pay. Subjects could respond that they were
currently employed for pay, that they were not currently employed but had a job to return to, that
they were not currently employed but had a job during the reference period, or that they were not
currently employed and did not have a job during the reference period. The flow of the
employment questionnaire ensured that these are mutually exclusive categories. Thus, the
inapplicable group for the employment status variable included only those who were less than 16
years of age. Persons responding that they were currently employed were asked to identify their
current main job, and current main job holders were asked whether they were self-employed or
worked for someone else. Thus, the inapplicable category for the current main job self-employment variable (SELFCM) included persons under age 16 and all persons age 16 or older
who were not currently employed.
Current main job holders were asked a variety of additional
ions about health insurance,
other benefits, and other job characteristics such as number of persons employed at the
establishment. Certain questions, namely those regarding benefits, were not asked of the self-employed. These current main job variables are discussed later in this section. The major focus
of the Round 1 editing, however, was the construction of an edited hourly wage variable for non
self-employed current main job holders. Note that any individuals in the "not ascertained"
category for current main job holder status are also categorized as "not ascertained" for other
variables based on the current main job.
Finally, two variables that relate to jobs other than the current main job and one variable that
relates to all jobs were also edited for release in the Round 1 file. The first two variables contain
information on whether the employee held (HELDNM1X) or was offered (OFERNM1X) health
insurance through a job other than the current main job. The variable that relates to all jobs
indicates whether an individual ever retired from any job (RETIRED). More detailed
descriptions of these variables follow.
Return To Table Of Contents
Hourly Wage
The Round 1 hourly wage variable (HRWAGEX) was constructed for persons age 16 or older
who were currently employed, but not self-employed, at their current main job. The
construction of an hourly wage for the current main job was complicated by the large number
of potential source variables that fed into it. Individuals may be paid by the hour, they may be
salaried with their salary based on different time periods (e.g., monthly, bimonthly, weekly),
or they may be paid on the basis of piecework, commission, or bonuses. In addition, some
individuals initially refused to provide a wage or salary estimate. Sample members who
initially refused to answer the hourly wage question were subsequently asked to indicate the
range in which their wage fell. The hourly wages for individuals who responded to the
bracketed wage questions were imputed using the median wage for hourly wage respondents
having wages in each of the respective wage brackets, disaggregated by sex.
Subjects who indicated that they were paid by salary, but refused to provide a salary amount,
were asked to provide an hourly wage. If they also refused to provide an hourly wage, they were
asked the series of bracketed wage questions.
Hourly wage for the current main job was constructed for 9,447 cases. Values for hourly wage
varied from approximately 5 cents per hour to $2,125 per hour. These minimum and maximum
values, however, were extreme outliers; the 1st percentile was $2.1/hour and the 99th percentile
was $43.75/hour. The median wage was $11.00/hour. For confidentiality reasons,
HRWAGEX has been rounded to the nearest 5 cents.
How the Hourly Wage was Edited
Given the complexity of the algorithms used to construct the hourly wage, a variable was
created which summarized the basic pathways used to estimate hourly wages for the current
main job. This variable (HRWAY) is included on the file.
In the simplest case, hourly wage was reported directly by the respondent. For other
individuals, construction of the hourly wage was based upon their salary, the time period on
which the salary was based, and the number of hours worked per time period. The number of
hours worked per time period was not always available; when this occurred it was assumed
that individuals worked the equivalent of 40 hours per week during the relevant time period.
(These individuals are separately identified by the variable.) Note that this inflates the number
of hours worked per week for those who, in fact, worked less than 40 hours per week.Analysts
should also be aware that two measures of hours were collected in the Employment Section of
the interview: (1) the hours on which the employee's wages or salary was based, and (2) the hours
usually worked. Measure (1) was used in calculating hourly wage whenever it was available
from the respondent. When measure (1) was not available, measure (2) was used.
Individuals who initially refused to provide a wage (or salary amount that enabled hourly wage
to be calculated) were asked if their wages fell within certain brackets (less than $4.25, $4.25
or higher but less than $10.00, $10.00 or higher but less than $15.00, $15.00 or higher).
Respondents in each of these wage brackets were then imputed the median wage for the
corresponding brackets from the initial respondents by sex. Finally, some individuals indicated
that they were paid by piecework, commission, or bonus. If this was the only manner in which
they were paid, income received from these sources was used to construct hourly wage. If
they received income directly from wages or salary, however, income from these other sources
was not used in constructing hourly wage.
Since HRWAY is essentially a variable that indicates the manner in which HRWAGEX was
constructed, the inapplicable and undetermined categories of HRWAY match those of
HRWAGEX.
Return To Table Of Contents
Hours Worked Per Week
All individuals who indicated that they were employed were asked to indicate the number of
hours they worked per week whether they were wage earners or self-employed. As mentioned
above, MEPS contains two measures of hours per week: usual hours worked per week, and
contractual hours per week. For most cases, hours worked per week (HRPRWK1X) is based
on usual hours per week. The CAPI programming was such that usual hours worked per week
was not available for all employed persons; when it was not available the contractual hours
variable was used. This occurred when the value for the variable HRWAY was 3 or 8.The
inapplicable category for HRPRWK1X includes all persons under age 16 and all persons age 16
and over who are not currently employed. The "not determined" category for HRPRWK1X
includes persons age 16 and over who are "not determined" on employment status, as well as
employed persons who refused to indicate how many hours per week they worked or said that
they "didn't know" how many hours per week they worked.
Employment Related Health Insurance Variables
Five health insurance variables were edited for the Round 1 release:
1. Health insurance held from current main job (HELDCM1X);
2. Health insurance offered from the current main job (OFERCM1X);
3. Health insurance held from non-current main job (HELDNM1X);
4. Health insurance offered from non-current main job (OFERNM1X); and5. Health insurance disavowed (DISVOW1).
The most straightforward health insurance variables were health insurance actually held from the
current main job (HELDCM1X) and health insurance offered from the current main job
(OFERCM1X). The first editing task for these variables was properly distinguishing between
missing and inapplicable code values. Inapplicable values were appropriate for persons under
age 16, or persons age 16 or older who did not hold a current main job, or were self-employed
with no employees. Individuals who report obtaining health insurance through a job that is self-employed with no employees can be identified with the variable HPRIVS1 (see section 2.5.5).
As with the health insurance variables for current main jobs, the editing of the health insurance
variables for non current main jobs focused on distinguishing nonresponse from inapplicable.
However, the definition of the inapplicable group for the non current main jobs was more
complex than for the current main jobs. In addition to the inapplicable group defined for the
current main job health insurance variables, the inapplicable group for the non current main
job health insurance variables included those with only a current main job.
More than 700 individuals reported in the Employment Section of the interview that they
received health insurance coverage from a job, but indicated later in the Health Insurance Section
that they did not receive health insurance from that job. Nearly 80 percent of these cases
occurred in reference to health insurance coverage from non-current main jobs. There are a
variety of reasons why this might have happened, including differences in the flow of the Health
Insurance and Employment Sections of the instrument or reporting errors in the Employment
Section which were revealed by responses in the Health Insurance Section (e.g., respondents
could have been offered health insurance on a non current main job, but actually obtained it
through their spouse's job and only realized this when asked questions in the health insurance
part of the interview).
To allow analysts to identify these individuals easily, a variable (DISVOW1) was constructed
that flags those individuals with differences in health insurance coverage between the
Employment and Health Insurance sections.
Responses for health insurance held in the employment section (HELDCM1X and HELDNM1X)
were recoded to be consistent with the variables indicating hospital/physician coverage in the
Health Insurance Section of the survey. Note, however, that the health insurance offered
variables (OFERCM1X and OFERNM1X) were set to "not ascertained" for these persons. This
is because the questions about whether health insurance was offered by an employer were
skipped in the employment section for those who had already indicated that they held insurance
from that employer.
Return To Table Of Contents
Employment Benefit Variables
Four employee benefit variables were edited for the initial Round 1 release:
1. Paid sick leave for doctor visits at current main job (PAYDRVS),
2. Paid sick leave at current main job (SICPAY),
3. Paid vacation at current main job (PAYVAC), and
4. Pension plan at current main job (RETPLN).
The inapplicable groups for all of these variables are those under age 16, those age 16 or older
without a current main job, and those 16 or older who are self-employed at their current main
job.
Other Employment Variables
Several other employment variables were edited for the Round 1 release. With the exception
of a retirement variable, all of these variables pertained to the current main job:
1. Organization type at current main job (JOBORG),
2. More than one location of business at current main job (MORELOC),
3. Labor union member at current main job (UNION),
4. Self-employed at current main job (SELFCM), and
5. Number of employees at location of current main job (NUMEMP).
With the exception of JOBORG and NUMEMP, these variables are largely self-explanatory.
The main point to note about JOBORG is that the private and public categories refer to wage
earners, whereas sole proprietorship, partnership, and incorporated pertain to self-employed
individuals.
NUMEMP has been top-coded at 500 for confidentiality reasons. Consequently, all
individuals working for an establishment with 500 or more employees have a NUMEMP value
of 500. Except for MORELOC, which includes the self-employed in the inapplicable category,
these current main job variables include self-employed workers in the universe. Thus, the
inapplicable category is somewhat narrower than the non self-employed current main job
variables discussed earlier.
Finally, Round 1 includes a retirement status variable (RETIRED). RETIRED is coded as 1 if
the individual retired from any job. The universe for this variable is all persons age 55 or
older who had a job; consequently, the inapplicable group consists of respondents less than age
55 and those over 55 who never held a job.
Return To Table Of Contents
2.5.4. Health Status Variables
The Round 1 Health Status edits involved the construction of person-level variables based on
information collected in the Condition Enumeration and Health Status sections of the
questionnaire. The majority of Health Status questions were initially asked at the family level to
ascertain if anyone in the household had a particular problem or limitation. These were followed
up with questions to determine which household member had each problem or limitation. In
addition, logical edits were performed in constructing the person-level variables to assure that
family-level and person-level values were consistent. Particular attention was given to cases
where missing values were reported at the family level to ensure that appropriate information was
carried to the person level. Inapplicable cases were those where a question was never asked
because of a skip patterns in the survey (e.g., individuals who were 13 years of age or older were
not asked some follow-up verification questions). Fifty individuals who either died during 1996
prior to the Round 1 interview or who were non-responding students were coded as inapplicable
on the health status variables.
Perceived Health Status and Mental Health Status
Perceived health status (RTEHLTH1) and mental health status (MNTHLTH1) were collected in
the Condition Enumeration section. These questions (CE 01 and CE 02) asked the respondent to
rate each person in the family according to the following categories: excellent, very good, good,
fair, and poor. No editing was done to these variables.
IADL and ADL Help/Supervision
The Instrumental Activities of Daily Living (IADL) Help or Supervision variable (IADLHLP1)
was constructed from a series of three questions. The initial question (HE01) determined if
anyone in the family received help or supervision with IADLs such as using the telephone,
paying bills, taking medications, preparing light meals, doing laundry, or going shopping. If the
response was "yes," a follow-up question (HE02) was asked to determine which household
member received this help or supervision. For persons under age 13, a final verification question
(HE03) was asked to confirm that the IADL help or supervision was the result of an impairment
or physical or mental health problem. If the response to the final verification question was "no,"
IADLHLP1 was coded as "no" for persons under the age of 13.
If no one in the family was identified as receiving help or supervision with IADLs, all members
of the family were coded as receiving no IADL help or supervision. In cases where the response
to the family-level question was "don't know," "refused," or otherwise missing, all persons were
coded according to the family-level response. In cases where the response to the family-level
question (HE01) was "yes" but no specific individuals were identified in the follow-up question
as having IADL difficulties, all persons were coded as "don't know."
The Activities of Daily Living (ADL) Help or Supervision variable (ADLHELP1) was
constructed in the same manner as IADLHLP1, but using questions HE04-HE06.
Use of Assistive Technology, Functional Difficulties, and Social/Recreational Limitations
The variables indicating use of assistive technology (AIDHELP1, from question HE07),
functional difficulties (WALKLIM1, from question HE09), and social/recreational limitations
(SOCLIMT1, from question HE22) were collected initially at the family level. If there was a
"yes" response to the family-level question, a second question identified which individual had the
difficulty. Each individual identified as having the difficulty was coded "yes" on the appropriate
variable; all remaining family members were coded "no." If the family-level response was "don't
know," "refused," or otherwise missing, all persons were coded with the family-level response.
In cases where the family-level response was "yes" but no specific individual was identified as
having difficulty, all family members were coded as "don't know."
Return To Table Of Contents
Work, Housework, and School Limitations
The variable indicating limitations in work, housework, or school (ACTLIMT1) was constructed
in the same manner as AIDHELP1, using questions HE19-HE20. If ACTLIMT1 was "yes" and
the person was 5 years of age or older, a follow-up question (HE20A) was asked to identify the
specific limitation or limitations for each person. These included working at a job
(WRKLIMT1), doing housework (HSELIMT1), or going to school (SCHLIMT1). WRKLIMT1,
HSELIMT1, and SCHLIMT1 have values of "yes" or "no" only if ACTLIMT1 was "yes;" they
were all coded as inapplicable if ACTLIMT1 was "no," "refused," or otherwise missing. When
ACTLIMT1 was "don't know," these variables were all coded as "don't know."
Cognitive Limitations
The variable (COGLIMT1) was collected at the family level as a three-part question (HE24-01 to
HE24-03) indicating if any of the adults in the family (1) experience confusion or memory loss,
(2) have problems making decisions, or (3) require supervision for their own safety. If a "yes"
response was obtained to any item, the persons affected were identified in HE25 and COGLIMT1
was coded as "yes." Remaining family members not identified were coded as "no" for
COGLIMT1.
If responses to HE24-01 though HE24-03 were all "no," or if two of three were "no" and the
remaining was "don't know," "refused," or otherwise missing, all family members were coded as
"no." If responses to the three questions were combinations of "don't know," "refused," and
missing, all persons were coded as "don't know." If the response to any of the three questions
was "yes" but no individual was identified in HE25, all persons were coded as "don't know."
Return To Table Of Contents
2.5.5 Health Insurance Variables
Constructed variables are provided for each source of health insurance coverage collected during
the MEPS Round 1 interview. With the exception of Medicaid and Medicare, the constructed
variables represent sources of insurance as reported by respondents. Minimal editing was
performed on the Medicare and Medicaid variables to assign persons to coverage from these
sources. All other coverage types are unedited and unimputed. Observations that contain edits
can be identified by comparing the edited and unedited versions of the Medicaid and Medicare
variables.
Public sources include Medicare, CHAMPUS/CHAMPVA (CHMPNOW1), Medicaid and other
public hospital/physician coverage. State-specific program participation (STATPROG1) was
also identified but is not considered health insurance for the purpose of this survey. The
Medicare and CHAMPUS/CHAMPVA variables indicate coverage at the time of the Round 1
interview date. All other public sources of insurance (including STATPROG1) indicate coverage
at any time during Round 1.
Medicare
Medicare (MCARNOW1) coverage was edited (MCARNW1X) for persons age 65 or over.
Within this age group, individuals were assigned Medicare coverage if:
1. They answered yes to a follow-up question on whether or not they received Social Security benefits; or
2. They were covered by Medicaid, other public hospital/physician coverage or Medigap coverage; or
3. Their spouse was covered by Medicare.
Medicaid and Other Public Hospital/Physician Coverage
Questions about other public hospital/physician coverage were asked in an attempt to identify
Medicaid recipients who may not have recognized their coverage as Medicaid. These questions
were asked only if a respondent did not report Medicaid directly. Respondents reporting other
public hospital/physician coverage were asked follow-up questions to determine if their
coverage was through a specific Medicaid HMO or if it included some other managed care
characteristics. Respondents who identified managed care from either path were asked if they
paid anything for the coverage and/or if a government source paid for the coverage. In an
attempt to identify coverage through Medicaid expansion programs, persons with private
insurance that was not employment-related were also asked if any government agency
contributed toward their premium.
The Medicaid variable (MCAID1) has been edited (MCAID1X) to include persons who paid
nothing for their other public hospital/physician insurance when such coverage was through a
Medicaid HMO or reported to include some other managed care characteristics. In addition, a
small number of persons reporting AFDC or SSI coverage (questions included in the Round 1
interview for this purpose) were assigned Medicaid coverage.
To assist users in further editing sources of insurance, this tape contains variables constructed
from the other public hospital/physician series that measure whether:
1. The respondent reported some type of managed care and paid something for the
coverage (OTPUBA1);
2. The respondent did not report any managed care (OTPUBB1);
3. A federal, state or local government paid anything toward the cost of the managed
care coverage (OTGOVPY1).
Variables are also included from the private insurance series that identify whether the federal
government (PRGVPYA1) or a state/local government (PRGVPYB1) contributed toward the
private non-employment related insurance. The variables OTPUBA1, OTPUBB1, OTGOVPY1,
PRGVPYA1, PRGVPYB1 are provided only to assist in editing and should not be used to make
insurance estimates.
Return To Table Of Contents
Private Insurance
Variables identifying private insurance in general (PRIV1) and specific private insurance sources
[employer group (PRIVEG1); union group (PRIVU1); non-group (PRIVNG1); and other group
(PRIVOG1)] were constructed. Private insurance sources identify coverage in effect at any time
during Round 1. Separate variables identify covered persons and policyholders (policyholder
variables begin with the letter "H"). These variables indicate coverage or policyholder status
within a source and do not distinguish between persons who are covered or are policyholders on
one or more than one policy within a given source. In some cases, the respondent was unable to
characterize the source of insurance (PRIVDK1). Covered persons (but not policyholders) are
identified when the policyholder is living outside the RU (PRIVOUT1). A source was
considered to provide insurance if, at a minimum, coverage was provided for hospital and
physician services. Sources of insurance with missing information regarding the type of
coverage were assumed to contain hospital/physician coverage. Persons without private
hospital/physician insurance were not counted as privately insured.
Health insurance through a job (PRIVEG1, PRIVU1, PRIVS1) was initially asked about in the
Employment Section of the interview and later confirmed in the Health Insurance Section.
Respondents also had an opportunity to report employer group (PRIVEG1) and union sponsored
(PRIVU1) insurance for the first time in the Health Insurance Section, but this insurance was not
linked to a specific job. Insurance that was initially reported in the Employment Section
contained information on whether the insurance was sponsored through the employer, a union or
both. If both the employer and union or the employer alone sponsored the insurance through a
particular job that insurance is classified as employer group insurance only. If the insurance was
sponsored solely through a union it was classified as union sponsored insurance.
All insurance reported to be through a job classified as self-employed, with firm size of 1
(PRIVS1) was initially reported in the Employment Section and verified in the Health Insurance
Section. Unlike the other employment-related variables (PRIVEG1 and PRIVU1), self-employed-firm size 1 (PRIVS1) health insurance could not be reported in the Health Insurance
section for the first time. The variable PRIVS1 has been constructed to allow users to determine
if the insurance should be considered employment-related.
Insurance initially reported in the Employment Section but not confirmed in the Health Insurance
Section can be identified using the variable DISVOW1 (see Section 2.2.3). A value of 1 or 2 for
this variable identifies disavowal of a specific insurance source only and does not necessarily
imply that a person is uninsured.
Private insurance that was not employment-related (PRIVOG1, PRIVNG1, PRIVDK1, and
PRIVOUT1) was reported in the Health Insurance Section only.
Public Insurance Coverage and Any Insurance in Round 1
Persons identified as insured in Round 1 are those reporting coverage under Medicare, Medicaid,
CHAMPUS/CHAMPVA, other public hospital/physician or private hospital/physician insurance
(including Medigap plans). A person is considered uninsured if not covered by one of these
insurance sources.
Persons covered only by state-specific programs (STATPROG1), for example, Maryland Kidney
Disease Program or Colorado Child Health Plan and those without hospital/physician benefits
(for example, private insurance for dental or vision care only, accidents or specific diseases) were
not considered to be insured when constructing the variables INSURED1 and PUBLIC1.
Return To Table Of Contents
3. Survey Sample Information
3.1 Sample Design and Response Rates
The MEPS is designed to produce estimates at the national and regional level over time for the
civilian, non-institutionalized population of the United States and some subpopulations of
interest. The data in the Round 1 file pertain to approximately the first third of calendar year
1996. Two more rounds of data collection will cover the remainder of 1996 while a fourth round
will follow the same sample into 1997.
The 1996 MEPS sample consisted of a subsample of households (occupied dwelling units) that
responded to the 1995 National Health Interview Survey (NHIS) in the two panels reserved for
the MEPS. Analysis can be undertaken using both the individual and the family as units of
analysis.
For detailed information on the sample design, see Appendix 1.
MEPS--Linked to the National Health Interview Survey
The sample of 10,639 households (occupied dwelling units) for the 1996 Panel of MEPS
consisted of a nationally representative subsample of the households responding to the 1995
National Health Interview Survey (NHIS). The NHIS sample design has three stages of sample
selection: an area sample of PSUs; a sample of segments (single or groups of blocks or block
equivalents) within sampled PSUs; and a sample of housing units within segments. Among
initially sampled households, those containing Hispanics and blacks were oversampled at rates
of approximately 2 and 1.5, respectively, times the rate of remaining households. These same
rates of oversampling are reflected in the MEPS sample of households. The only major
difference in the definition of a household between NHIS and MEPS is that college aged students
living away from home during the school year were interviewed at their place of residence for the
NHIS but were identified by and linked to their parents' household for MEPS.
Response Rates
Since the 1996 MEPS Household Survey sample was selected from a nationally representative
sub-sample of households who were part of the 1995 NHIS, the overall response was derived
from three component response rates. The NHIS response rate achieved for the households
eligible for the MEPS was 93.9 percent. Of the 10,639 responding NHIS dwelling units eligible
for the MEPS, 99.6 percent were identified with sufficient information to permit MEPS data
collection. Within these dwelling units, there were 11,429 eligible reporting units targeted for
interviews in Round 1, of which 83.1 percent responded to the first core MEPS interview. Two
percent of the reporting units fielded in Round 1 could not be located; 15 percent were located
and declined to participate in the MEPS interview, accounting for the 17 percent reporting unit
nonresponse. Overall, the joint NHIS - Round 1 response rate for the 1996 MEPS household
survey was 77.7 percent (.939 x .996 x .831).
Return To Table Of Contents
3.2 Sample Weights and Variance Estimation
3.2.1 Sample Weighting
The use of sample weights permits the derivation of estimates for the U.S. civilian, non-institutionalized population and subgroups of this population based on the sample data. Two
weights are provided on the public use file: a person level weight and a family level weight.
The Person Level Weight
The name of the person level weight variable is WGTSP1. A person level weight was assigned
to all key, eligible members of the U.S. civilian, non-institutionalized population for whom data
were collected in the first Round of data collection for the 1996 MEPS. This weight reflects the
original household probability of selection for the NHIS, ratio-adjustment to NHIS national
population estimates at the household level, adjustment for non-participation in MEPS at the
dwelling unit level, and poststratification to figures obtained from March, 1996 Current
Population Survey data at the family and person level. The person level poststratification
reflected population distributions across census region; race/ethnicity (Hispanic, black/non-Hispanic, other); sex; and age. Overall, the weighted population estimate is 263,515,813 for the
civilian noninstitutionalized population.
The Family Level Weight
The name of the family level weight is WGTRU1. A family was defined to be two or more
persons living together who are related by blood, marriage, adoption, or foster care. The
members of a "family" unit can vary over time due to births, deaths, and migration in and out of
family units. For the purposes of assigning a family level weight for the first round of MEPS, a
family unit was defined as the set of related people living together during the reference period for
whom data were collected (from January 1, 1996 to the date of the Round 1 interview). Persons
who died during the reference period were considered to be Round 1 family members, as were
people for whom data could be collected for a portion of the round if a person left the civilian,
non-institutionalized population later in the round (i.e., if a person was institutionalized, left the
country, or joined the military). College age students living away from home during the school
year also were considered family members. A family member need not be key nor in-scope.
Such persons are family members for the "snapshot" of the family represented by the Round 1
period of time and may have made important contributions to such items as a family's income or
health care coverage.
All responding family units with at least one key, eligible, in-scope person as well as reporting
units consisting of a single key, in-scope, eligible respondent received a family level weight. At
the family level, poststratification to March, 1996 CPS figures was undertaken reflecting factors
such as family type (reference person married, spouse present, male reference person, no spouse
present, female reference person, no spouse present), size of family, age of reference person,
location of family (census region and MSA status), and race/ethnicity of reference person. The
weighted estimate of the number of family units (including single person units) containing at
least one member of the U.S. civilian non-institutionalized population is 110,206,950.
NOTE: Analysts should exclude ineligible persons (ELIGRND1=2) from all family level
analyses.
Return To Table Of Contents
3.2.2 Variance Estimation
To obtain estimates of variability (such as the standard error of sample estimates or
corresponding confidence intervals) for estimates based on MEPS survey data, one needs to take
into account the complex sample design of MEPS for both person and family level analyses.
Various approaches can be used to develop such estimates of variance, using a Taylor series
method for variance estimation or alternative replication methodologies. Replicate weights have
not been developed for the Round 1 MEPS data; the focus here is to identify the variables needed
to implement a Taylor series estimation approach.
Using such an approach, variance estimation strata and the variance estimation PSUs within
these strata must be specified. The corresponding variables on the MEPS Round 1 data base are
VARSTRT1 and VARPSU1, respectively. Specifying a "with replacement" design in a
computer software package such as SUDAAN should provide estimated standard errors
appropriate for assessing the variability of MEPS survey estimates. It should be noted that the
number of degrees of freedom associated with estimates of variability indicated by such a
package may not appropriately reflect the number available. For MEPS sample estimates for
characteristics generally distributed throughout the country (and thus the sample PSUs), a
reasonable rule of thumb is that there are roughly 170 degrees of freedom associated with the
corresponding estimates of variance.
Return To Table Of Contents
4. Programming Information
Description: MEPS 1996 Panel Round 1 Population Characteristics
File Name: HC001.DAT
Number of Observations: 24,676
Number of Variables: 111
Record Length: 264
Record Format: fixed
Record Identifier and Sort Key: DUPERSID
Return To Table Of Contents
D. Codebook
Return To Table Of Contents
E. Crosswalk of Variables to Variable Source
SURVEY ADMINISTRATION VARIABLES - PUBLIC USE
VARIABLE |
LABEL |
SOURCE |
DUID |
Dwelling Unit ID |
Assigned in sampling |
PID
|
Person Number (PN) |
Assigned in sampling or
by CAPI |
FAMID1 |
Family Identifier |
Assigned by CAPI |
DUPERSID |
Sample Person ID (DU + PN) for Public Use |
Assigned in sampling |
RUENDMM |
Date of Interview (Date Started: Month) |
CAPI Derived |
RUENDDD |
Date of Interview (Date Started: Day) |
CAPI Derived |
RUENDYY |
Date of Interview (Date Started: Year) |
CAPI Derived |
RULETTER |
RU Letter |
CAPI Derived |
RUSIZE |
RU Size |
CAPI Derived |
RUCLASS |
RU Fielded As: Standard, New, or Student |
CAPI Derived |
UNITSIZE |
RU Size-Includes Students |
CAPI Derived |
REGION |
Census Region |
Assigned in sampling |
MSA |
MSA |
Assigned in sampling |
RND1REF |
Reference Person at Round 1 |
RE 42-45 |
RND1RESP |
1st Respondent Indicator for Round1 |
RE 6, 8 |
BEGREFDD |
Reference Period Begin Date: Day |
CAPI Derived |
BEGREFMM |
Reference Period Begin Date: Month |
CAPI Derived |
BEGREFYY |
Reference Period Begin Date: Year |
CAPI Derived |
ENDREFDD |
Reference Period End Date: Day |
CAPI Derived |
ENDREFMM |
Reference Period End Date: Month |
CAPI Derived |
ENDREFYY |
Reference Period End Date: Year |
CAPI Derived |
KEYNESS |
Person Key Status |
RE Section |
ELIGRND1 |
Eligibility: Eligible / Not Eligible |
RE Section |
INSCOPE |
Inscope |
RE Section |
PSTATUS |
Person Disposition Status |
RE Section |
Return To Table Of Contents
DEMOGRAPHIC VARIABLES - PUBLIC USE
VARIABLE |
LABEL |
SOURCE |
SEX |
Sex |
RE 12, 57, 61 |
AGE1X |
Age (Edited/Imputed) |
RE 12, 57-66 |
DEMOGRAPHIC VARIABLES - PUBLIC USE (CONTINUED)
VARIABLE |
LABEL |
SOURCE |
DOBMM |
Date of Birth: Month |
RE 12, 57-66 |
DOBYY |
Date of Birth: Year |
RE 12, 57-66 |
RACEX |
Race (Edited/Imputed) |
RE 101, 102 |
RACETHNX |
Race/Ethnicity (Edited/Imputed) |
RE 98-102 |
HISPANX |
Hispanic Ethnicity (Edited/ Imputed) |
RE 98-100 |
HISPCAT |
Specific Hispanic Ethnicity Group |
RE 98-100 |
FTSTUD1 |
Currently a Full-Time Student |
RE 11A, 108 |
EDUCYR1 |
Completed Years of Education |
RE 103-105 |
HIGHDEG1 |
Highest Degree |
RE 104-105 |
MARRY1X |
Marital Status (Edited) |
RE 13, 97 |
SPOUSID1 |
Spouse ID |
RE 13, 97 |
SPOUSIN1 |
Marital Status W/Presence of Spouse |
RE 13, 97 |
DIDSERVE |
Ever Served in Armed Forces |
RE 18, 95 |
ACTDUTY1 |
Is Person Military Full Time Active Duty |
RE 14, 96A |
REFREL1X |
Relation to Reference Person (Edited/Imputed) |
RE 76-77 |
Return To Table Of Contents
EMPLOYMENT VARIABLES - PUBLIC USE
VARIABLE |
LABEL |
SOURCE |
EMPST1 |
Person Employment Status (Edited) |
EM 1, 2, and 3 |
HRWAGEX |
Person Hourly Wage (Edited) |
EW 1-24 |
HRWAY |
How Wage was Edited |
EW 1-24 |
HRPRWK1X |
Hours per Week (Edited) |
EM 104, 105; EW 117 |
HELDCM1X |
HI Held from Current Main Job (Edited) |
EM 113; HX 2, 48 |
HELDNM1X |
HI Held from Non-Current Main Job (Edited) |
EM 17, 26, 39, 52, 69, 81
HX 2, 48 |
OFERCM1X |
HI Offered from Current Main Job (Edited) |
EM 113, 114; HX 2, 48 |
OFERNM1X |
HI Offered from Non-Current Main Job
(Edited) |
EM 17, 26, 39, 52, 69, 81
HX 2, 48 |
DISVOW1 |
HI Disavowed in HI Section |
HELDCM1X, HELDNM1X
HX 2 |
JOBORG |
Organization Type at Current Main Job |
EM 94-96 |
NUMEMP |
Number of Employees at Location of CMJ |
EM 91, 92 |
RETIRED |
Retired from Any Job |
EM 77- 80 |
SELFCM |
Self-Employed at Current Main Job |
EM 1, 4, 5, 11, 12 |
Return To Table Of Contents
HEALTH STATUS VARIABLES - PUBLIC USE
VARIABLE |
LABEL |
SOURCE |
IADLHLP1 |
LTC Flag: IADL Help/Supervision |
HE 1-3 |
ADLHELP1 |
LTC Flag: ADL Help/Supervision |
HE 4-6 |
AIDHELP1 |
LTC Flag: Use Aids/Special Equipment |
HE 7-8 |
WALKLIM1 |
LTC Flag: Walk/Bend/Stoop Difficulties |
HE 9-10 |
ACTLIMT1 |
LTC Flag: Work/Housework/School Limit |
HE 19, 20 |
WRKLIMT1 |
Limited Ability to Work |
HE 19-20, 20A |
HSELIMT1 |
Limited Ability Doing Housework |
HE 19-20, 20A |
SCHLIMT1 |
Limited Ability Going to School |
HE 19-20, 20A |
SOCLIMT1 |
LTC Flag: Social Limitations |
HE 22-23 |
COGLIMT1 |
LTC Flag: Cognitive Limitations |
HE 24-25 |
Return To Table Of Contents
HEALTH INSURANCE VARIABLES - PUBLIC USE
VARIABLE |
LABEL |
SOURCE |
CHMPNOW1 |
PID Cov by CHAMPUS/VA - RD 1 Int Date |
RE 14, 96A; HX 13; HQ 1-2;
HX 12 (RU level) |
MCAID1 |
PID Covered by MEDICAID - RD 1 |
HX 11, 10 (RU level) |
MCAID1X |
PID Covered by MEDICAID - RD 1 (Edited) |
HX 11, 15, 18-19, 41-43, 45
HX 10, 14, 16 (RU level) |
OTPUBA1 |
PID Covered by/Pays Other Gov MCAID
HMO - RD 1 |
HX 15, 41-43, 45, 14
(RU level) |
OTPUBB1 |
PID Covered by Other Public Not MCAID
HMO - RD 1 |
HX 15, 41-43, 14 (RU level) |
STATPRG1 |
PID Covered from State-Specific Prog - RD 1 |
HX 17, 19, 16 (RU level) |
PUBLIC1 |
PID Covered from Public Ins Plan - RD 1 |
CHMPNOW1; MCAID1X;
OTPUBA1; OTPUBB1;
MCARNW1X |
MCARNOW1 |
PID Covered by MEDICARE - RD 1 Int Date |
HX 6-7 |
MCARNW1X |
PID Covered By MEDICARE - RD 1 Int Date
(Edited) |
HX 7, 9, 11, 15 (PRIV1 and
HX 48);
HX 6, 10, 14 (RU level) |
Return To Table Of Contents
HEALTH INSURANCE VARIABLES - PUBLIC USE (CONTINUED)
VARIABLE |
LABEL |
SOURCE |
PRIV1 |
PID has Private Health Ins Coverage - RD 1 |
PRIVEG1; PRIVU1;
PRIVS1; PRIVOG1;
PRIVNG1; PRIVDK1;
PRIVOUT1 |
PRIVEG1 |
PID has Cov from Emp Group Plan - RD 1 |
HX 2, 23, 48; HP 9, 11, 15, 16
HELDCM1X, HELDNM1X,
EM 117 |
PRIVU1 |
PID has Cov from Union Group Plan - RD 1 |
HX 2, 3, 23, 48; HP 9, 11, 15,
16; HELDCM1X,
HELDNM1X, EM 117 |
PRIVS1 |
PID has Cov from Self-Emp-1 Plan - RD 1 |
HX 3, 48; EM 17, 18, 26, 27,
39, 40, 52, 53, 69, 70, 81, 82,
91, 92; HP 9, 15-16 |
PRIVOG1 |
PID has Cov from Other Group Plan - RD 1 |
HX 23, 48; HP 1, 2, 11, 15-16 |
PRIVNG1 |
PID has Cov from Nongroup Plan - RD 1 |
HX 23, 48; HP 11, 15-16 |
PRIVDK1 |
PID has Cov from Private DK Plan - RD 1 |
HX 23, 48; HP 11, 15-16 |
PRIVOUT1 |
PID has Cov from Holder Outside RU - RD 1 |
HX 23, 48; HP 15-16 |
HPRIV1 |
PID is Holder of Private Ins Plan - RD 1 |
HPRIVEG1; HPRIVOG1;
HPRIVS1; HPRIVU1;
HPRIVNG1; HPRIVDK1 |
HPRIVEG1 |
PID is Holder of Emp Group Plan - RD 1 |
PRIVEG1; HP 9, 11 |
HPRIVU1 |
PID is Holder of Union Group Plan - RD 1 |
PRIVU1; HP 9, 11 |
HPRIVS1 |
PID is Holder of Self-Emp-1 Ins - RD 1 |
PRIVS1; HP 9 |
HPRIVOG1 |
PID is Holder of Other Group Plan - RD 1 |
PRIVOG1; HP 11 |
HPRIVNG1 |
PID is Holder of Nongroup Plan - RD 1 |
PRIVNG1; HP 11 |
HPRIVDK1 |
PID is Holder of Private DK Plan - RD 1 |
PRIVDK1; HP 11 |
INSURED1 |
PID is Insured - RD 1 |
PUBLIC1; PRIV1 |
Return To Table Of Contents
WEIGHTS - PUBLIC USE
VARIABLE |
LABEL |
SOURCE |
WGTSP1 |
Person Round 1 Weight |
Constructed |
WGTRU1 |
Family Level Round 1 Weight |
Constructed |
VARSTRT1 |
Variance Estimation Stratum |
Constructed |
VARPSU1 |
Variance Estimation PSU |
Constructed |
Return To Table Of Contents
F. Appendices
Appendix 1
Sample Design of the 1996 Medical Expenditure Panel Survey
Household Component
March 1997
Steven B. Cohen
Agency for Health Care Policy and Research
Center for Cost and Financing Studies
2101 E. Jefferson St., Suite 501
Rockville, Maryland 20852
(301) 594-1406
1.0 Introduction
The Household Component of the 1996 Medical Expenditure Panel Survey (MEPS) was
designed to produce national and regional estimates of the health care utilization, expenditures,
sources of payment, and insurance coverage of the U.S. civilian noninstitutionalized population.
The MEPS includes surveys of medical providers, employers, and other health insurance
providers to supplement the data provided by household respondents. The design of the MEPS
permits both person-based and family-level estimates. The scope and depth of this data
collection effort reflects the needs of government agencies, legislative bodies, and health
professionals for the comprehensive national estimates needed in the formulation and analysis of
national health policies.
The MEPS collects data on the specific health services that Americans use, how
frequently they use them, the cost of these services and how they are paid, as well as data on the
cost, scope, and breadth of private health insurance held by and available to the U.S. population.
The MEPS is unparalleled for the degree of detail in its data, as well as its ability to link health
service medical expenditures and health insurance data to the demographic, employment,
economic, health status, utilization of health services, and other characteristics of survey
respondents. Moreover, the MEPS is the only national survey that provides a foundation for
estimating the impact of changes in sources of payment and insurance coverage on different
economic groups or special populations of interest, such as the poor, elderly families, veterans,
the uninsured, and racial and ethnic minorities.
In this paper, the sample design of the MEPS, initially referred to as the National Medical
Expenditure Survey (NMES-3), is described. The 1996 MEPS used the 1995 National Health
Interview Survey (NHIS) as the sample frame for the survey. The redesigned MEPS reflects the
first stage of implementation of the Department of Health and Human Services (DHHS) Survey
Integration Plan, which provides directives targeted to improve the analytic capacity of programs,
fill major data gaps, and establish a framework in which DHHS data activities are streamlined
and rationalized. Through this effort, specifically through a linkage to the NHIS, the MEPS has
achieved a number of significant design improvements and analytic enhancements.
Attention is given to the resultant design efficiencies and enhancements in analytical
capacity that have been and will be realized through the MEPS sample design integration with
the NHIS. The report includes a summary of sample size specifications and precision targets for
national population estimates and health care expenditure estimates for policy-relevant
population subgroups. A discussion is also provided regarding the modification of the MEPS
from a periodic annual survey to an ongoing continuous data collection effort with each
expenditure panel of households followed for two years.
Return To Table Of Contents
2.0 Redesign of the Medical Expenditure Survey as a Component of the DHHS Survey
Integration Plan
As part of the Reinventing Government Part II (REGO II) activities, the DHHS targeted
the improvement of the analytical capacity of DHHS programs, the filling of major data gaps,
and the establishment of a survey consolidation framework in which DHHS data activities are
streamlined and rationalized. A Survey Consolidation Working Group was charged with
developing a consensus plan for meeting these objectives (Hunter, Arnett, Mathiowetz, et al.,
1995; Arnett, Hunter, and Cohen, et al., 1996). A major concentration of the Survey Integration
Plan was the redesign of the health care expenditure and insurance studies conducted by the
DHHS which include the National Medical Expenditure Survey (NMES), the Medicare Current
Beneficiary Survey (MCBS), the National Employer Health Insurance Survey (NEHIS), and the
NHIS. The proposed integrated survey design was specified to achieve significant cost
efficiencies by eliminating duplicative efforts and reducing overall respondent burden.
Furthermore, by virtue of integrating the design features of the component surveys, their
respective analytical capacities are enhanced. A number of survey design enhancements were
also proposed to improve upon current survey design capabilities. These included considering an
ongoing longitudinal survey effort and allowing for a future capacity to derive state-specific
health care estimates. Consideration was also given to the inclusion of a periodic institutional
component in the survey to provide national use and expenditure estimates for the population
residing in nursing homes (Hunter, Arnett, Mathiowetz, et al., 1995).
2.1 Design Enhancements and Efficiencies to be Achieved Through Survey Integration
One of the attractions of the DHHS Survey Integration Plan was the enhanced analytical
capacity that would be achieved by the distinct surveys that would be linked through design
integration. This could be realized by sample size expansions that would occur through survey
mergers such as the planned integration between the MEPS and the MCBS and the consolidation
of employer surveys conducted by the DHHS. Also, use of the NHIS as a sample frame for the
MEPS would increase the analytical content of the resultant linked surveys. Through design
integration of the respective surveys sponsored by the DHHS, inefficiencies associated with
duplicative survey efforts would be significantly reduced. Another goal was to achieve reductions
in survey design costs by the implementation of a uniform framework for DHHS-sponsored
surveys with overlapping analytical focus with respect to questionnaire content, data editing,
imputation, estimation, database structure, and development of analytic files. Additional
efficiencies in survey operations are anticipated in future years as a consequence of conducting
an annual medical expenditure survey rather than one every decade.
By moving to this integrated, annual household data collection effort, the DHHS expands and
enhances its analytic capabilities as described below:
Retains the design of the core NHIS household interview. This core will provide cross-sectional population statistics on health status and health care utilization with sufficient
sample size to allow for analyses based on detailed breakdowns of age, race, sex,
income, and other sociodemographic characteristics. The core will also allow the use of
data on a broad range of topics currently provided by the NHIS.
Retains the analytical capacity to obtain both annual and quarterly population estimates of
health care utilization and the prevalence of health conditions, for the nation and for
policy-relevant population subgroups.
Provides the ability to model individual (and family-level) health status, access to care
and use, expenditure, and insurance behavior over the year and examine the distribution
of these measures across individuals. The longitudinal feature of the MEPS to collect data
over multiple years further enhances the capacity to model behavior over time.
Provides the ability to relate data from a detailed sample (e.g., MEPS) to a larger sample
(e.g., NHIS) to enhance the utility of the MEPS for national health account estimation and
microsimulation modeling, including disaggregation by age group or geographic area.
Provides the potential to expand to state-level estimates for marginal costs using the
enhanced 358 PSU sample design of the NHIS.
The longitudinal (over several years) aspect of the MEPS integrated data collection effort
provides the following:
- An increase in statistical power to examine change or make comparisons over
time;
- The capacity to examine changes over time as well as changes in the relationship
among measures of health status, access to care, health care use, expenditures,
health insurance coverage, employment, functional limitations and disabilities,
and demographic characteristics.
Return To Table Of Contents
2.2 MEPS Household Component
The original NMES-3 sample design called for an independent screening interview to
identify a nationally representative sample and facilitate oversampling of policy-relevant
population subgroups. Data collection and training costs associated with this independent
screening interview were projected to exceed $8 million. As part of the DHHS Survey
Integration Plan, the separate screening interview to identify the expenditure survey sample was
eliminated. As an alternative, the NHIS was specified as the sampling frame for the medical
expenditure survey, MEPS. The NHIS is an ongoing annual household survey of approximately
42,000 households (109,000 individuals) conducted by the National Center for Health Statistics
(NCHS) to obtain national estimates for the U.S. civilian noninstitutionalized population on
health care utilization, health conditions, health status, insurance coverage, and access. In
addition to the cost savings achieved by the substitution of the NHIS as the MEPS sample frame,
the design modification will result in an enhancement in analytical capacity of the resultant
survey data. Use of the 1995 NHIS data in concert with the data collected for the 1996 MEPS
provides an additional capacity for longitudinal analyses not available in the original (NMES-3)
design. Furthermore, the greater number and dispersion of the sample PSUs that comprise the
MEPS national sample should result in improvements in precision over the original design
specifications.
To fill major data gaps identified by the DHHS, the MEPS is specified as a continuous
survey with sample peaks at five year intervals. The initial sample of 10,597 NHIS dwelling
units selected for the 1996 MEPS is reduced from the original 1996 plan to also permit estimates
for calendar year 1997. An overlapping panel design will be adopted for the MEPS, where the
1996 panel will be followed for data collection through 1997. A new nationally representative
sample of 6,300 dwelling units will be selected from the 1996 NHIS to supplement the 1996
MEPS panel in order to meet the original precision specifications for the specified policy-relevant population subgroups for calendar year 1997, with the exception of the elderly. These
policy-relevant population subgroups consist of:
Adults (18+) with functional impairments.
Children with limitations of activity.
Individuals predicted to incur high medical expenditures.
Individuals predicted to have incomes less than 200% of the poverty level.
(Cohen, 1996).
A preliminary contact with the NHIS responding households selected for the MEPS study was
made prior to the start of the MEPS, to announce the survey and introduce record-keeping
activities. The revised study design of the MEPS includes several components: the Household
Component (HC) consisting of an overlapping panel design in which any given sample panel is
interviewed a total of six times over three consecutive years to yield annual data for two calendar
years; the Medical Provider Component (MPC) with a sample of medical providers that treated
HC persons; and the Insurance Component (IC) with a sample of employers and other sources of
health insurance of HC persons. The survey is co-sponsored by the Agency for Health Care
Policy and Research (AHCPR) and the NCHS. Westat and the National Opinion Research Center
(NORC) are the data collection organizations for the 1996 MEPS Household Component.
Return To Table Of Contents
2.3 MEPS Household Component Sample Design
The 1996 MEPS Household Component sample was selected from households that
responded to the 1995 NHIS. The NHIS has been designed to permit nationally representative
subsamples to be selected by restricting the sample to one of four distinct panels. Any
combination of one to four panels will provide a nationally representative sample of households.
Furthermore, each NHIS panel subsample for a given quarter of a calendar year is nationally
representative. The 1996 MEPS household sample linked to the 1995 NHIS was selected from
two of the four 1995 NHIS panels and encompassed half of the households in the NHIS sample
during the second and third quarters of 1995.
The complete 1995 NHIS sample (panels 1-4) consists of 358 Primary Sampling Units
(PSUs, e.g., counties or groups of contiguous counties) with a targeted sample of approximately
42,000 responding households. The sample PSUs selected for the NHIS were stratified by
geographic (Census region and state), metropolitan status, and sociodemographic measures
(Judkins, Marker, and Waksberg, 1994). Within sample PSUs, a sample of blocks (segments)
were selected after being stratified by measures of minority population density which allowed for
an oversample of areas with high population concentrations of blacks and Hispanics. A
nationally representative sample of approximately 71,000 addresses within sampled blocks was
selected and targeted for further screening to facilitate an oversample of blacks and Hispanics as
part of the 1995 NHIS interview.
The 1995 NHIS subsample selected for the 1996 MEPS consists of 195 PSUs. In the two
targeted quarters of 1995 these PSUs include approximately 1,675 sample segments (second
stage sampling units) and 10,597 responding NHIS households. This NHIS sample reflects an
oversample of households with Hispanics and blacks at the following approximate ratios of
representation relative to the remaining households (Hispanics 2.0:1, blacks 1.5:1). This 1996
MEPS sample will constitute a panel that will be surveyed to collect annual data for two
consecutive years.
A new 1997 MEPS panel sample will be selected as a nationally representative subsample
from households that respond to the 1996 NHIS. More specifically, the 1997 MEPS sample
linked to the 1996 NHIS will be selected from two of the four NHIS panels and will reflect
additional disproportionate sampling in order to satisfy the precision requirements specified for
the 1997 MEPS household survey, which generally coincide with the original plan for the 1996
survey (Cohen, 1996). As in 1995, the complete 1996 NHIS sample will consist of 358 PSUs
with a targeted sample of approximately 42,000 responding households. The nationally
representative 1996 NHIS subsample reserved for the 1997 MEPS prior to additional
subsampling will be obtained from the same 195 PSUs selected for the 1996 MEPS household
sample and include approximately 21,000 responding NHIS households as eligible for sample
selection. Once again, this NHIS sample reflects an oversample of Hispanics and blacks at the
following approximate ratios of representation relative to the remaining households (Hispanics
2.0:1, Blacks 1.5:1). A nationally representative subsample of approximately 6,300 NHIS
responding households (6,480 reporting units) will be selected for the new 1997 MEPS panel.
This sample will consist of an oversample of the following policy-relevant subgroups:
Adults (18+) with functional impairments.
Children with limitations of activity.
Individuals predicted to incur high medical expenditures.
Individuals predicted to have incomes less than 200% of the poverty level.
An oversample of non-functionally impaired elderly individuals was not planned for the 1997
survey, given the availability of the 1997 MCBS and the planned future consolidation of the
MCBS and the MEPS. The MCBS is an annual person-based survey to obtain the same types of
estimates derivable from the MEPS household survey on the health care utilization, expenditures,
sources of payment, and health insurance coverage for Medicare beneficiaries. The new 1997
MEPS panel will be fielded to collect annual data for two consecutive years.
As part of the redesign, the 1997 MEPS Household Component sample will consist of the
new nationally representative 1997 MEPS panel in combination with the second year of the 1996
MEPS sample. Overall, the 1997 MEPS household sample will consist of approximately 13,700
reporting units (total adjusted for MEPS Round 1 "split-offs," though not reflecting new split-offs in Rounds 2 and 3) completing the full series of MEPS interviews to obtain calendar year
use and expenditure data for calendar year 1997. Sample selection procedures for the 1997
MEPS sample will be implemented in-house by AHCPR staff, based on data keyed from the
1996 NHIS interviews.
In 1998, a new MEPS sample of approximately 5,200 households (5,350 reporting units)
will be selected as a nationally representative subsample of households that responded to the
1997 NHIS. In addition, the entire 1997 panel of 5,397 reporting units will be continued to
obtain calendar year 1998 data on health care use and expenditures (with a targeted round-specific response rate of 97 percent). Consequently, the MEPS sample for 1998 will consist of
approximately 9,500 reporting units (adjusted for split-offs in Round 1) completing three core
rounds of data collection to obtain calendar year data (4,457 households from the new sample,
5,078 from the 1997 MEPS sample). In 1998, the 1996 MEPS panel will be retired.
For years 1998-2001, the survey will scale back to an overall sample of approximately
9,500 reporting units completing three core rounds of data collection to obtain calendar year data
on health care utilization and expenditures, with approximately 5,000 continuing from the
previous year for each of the years. In 2002, the survey would begin the five year cycle again
with an increase to 13,700 reporting units (adjusted only for Round 1 split-offs) completing three
core rounds of data collection to obtain calendar year data on health care utilization and
expenditures. Coupled with data from the MCBS, this would provide the DHHS with the analytic
capabilities first proposed for the 1996 NMES-3 with respect to sample size.
Return To Table Of Contents
2.4 Dwelling Units, Reporting Units, and Other Definitions
The definitions for Dwelling Units and Group Quarters in the MEPS Household
Component are generally consistent with the definitions employed for the NHIS. A Reporting
Unit is a person or group of persons in the sampled dwelling unit that are related by blood,
marriage, adoption or other family associations, who are to be interviewed at the same time in
MEPS. Examples of discrete reporting units are:
(1) a married daughter and her husband living with her parents in the same dwelling
are considered one reporting unit.
(2) a husband and wife and their unmarried daughter, age 18, who is living away from
home at college constitute one family, but two reporting units.
(3) three unrelated persons living in the same dwelling unit would be three reporting
units.
College students under 24 years of age who usually live in the sampled household, but are
currently living away from home and going to school, will be treated as separate reporting units
for the purpose of data collection.
The 1996 MEPS sample consisted of households (dwelling units) that responded to the
1995 NHIS in the two panels reserved for the MEPS, with the basic analysis unit defined as the
person. Analysis is planned with both the individual and the family as units. Through the
reenumeration section of the Round 1 questionnaire, the status of each individual sampled at the
time of the NHIS interview is classified as "key" or "non-key," "in-scope" or "out-of-scope," and
"eligible" or "ineligible" for MEPS data collection. For an individual to be in-scope and eligible
for person-level estimates derived from the MEPS household survey, the person needs to be a
member of the civilian noninstitutionalized population for some period of time in the calendar
year of analytical interest. Because a person's eligibility for the survey may have changed since
the NHIS interview, sampling reenumeration takes place in each subsequent reinterview for
persons in all households selected into the core survey. The keyness, in-scope, and eligibility
indicators, together, define the target sample to be used for person-level national estimates. Only
persons who are key, in-scope, and eligible for data collection will be considered in the
derivation of person-level national estimates from the MEPS.
Key Persons: Key survey participants are defined as all civilian non-institutionalized
individuals who resided in households that responded to the nationally representative NHIS
subsample reserved for the MEPS (e.g. approximately 10,600 households from the 1995 NHIS),
with the exception of college students interviewed at dormitories. Members of the armed forces
that are on full time active duty and reside in responding NHIS households which include other
family members who are civilian non-institutionalized individuals are also to be defined as key
persons, but will be considered out of scope for person-level estimates derived for the survey.
All other individuals who join the NHIS reporting units that define the 1996 MEPS
household sample (in Round 1 or later MEPS rounds) and did not have an opportunity for
selection during the time of the NHIS interview will also be considered key persons. These
include newborn babies, individuals who were in an institution or outside the country moving to
the United States, and military personnel previously residing on military bases who join MEPS
reporting units to live in the community.
College students under 24 years of age interviewed at dormitories in the 1995 NHIS will
be considered ineligible for the 1996 MEPS sample and not included in that sample.
Furthermore, any unmarried college students under 24 years of age that responded to the 1995
NHIS interview while living away at school (not in a dormitory) will be excluded from the
sample if it is determined in the MEPS Round 1 interview that the person is unmarried, under 24
years of age, and a student with parents living elsewhere who resides at his/her current housing
only during the school year. If, on the other hand, the person's status at the time of the MEPS
Round 1 interview is no longer that of an unmarried student under 24 years of age living away
from home, then the person will be retained in the 1996 MEPS sample as a key person.
Alternatively, at the time of the MEPS Round 1 interview with NHIS sample
respondents, a determination will be made if there are any related college students under 24 years
of age who usually live in the sampled household, but are currently living away from home and
going to school. These college students are considered key persons and will be identified and
interviewed at their college address, but linked to the sampled household for family analyses.
Some of these college students living away from home at the time of the Round 1 interview will
have been identified as living in sampled household at the time of the 1995 NHIS interview. The
remainder will be identified at the time of the MEPS Round 1 interview with the NHIS sampled
households.
Non-key Persons: Persons who were not living in the original sampled dwelling unit at
the time of the 1995 NHIS interview and who had a non-zero probability of selection for that
survey will be considered non-key. If such persons happen to be living in sampled households
(in Round 1 or later rounds), MEPS data (e.g., utilization and income) will be collected for the
period of time they are part of the sampled unit to permit family analyses. Non-key persons who
leave any sampled household will not be recontacted for subsequent interviews. Non-key
individuals are not part of the target sample used to obtain person-level national estimates.
In situations where key persons from the NHIS sampled household selected for MEPS
move out (in Round 1 or later rounds) and join or create another family, data on all members of
this new household who are related by blood, marriage, adoption or foster care to the persons
from the NHIS sampled household will be obtained from the point in time that the NHIS sampled
person joined that new household. Similarly, data will be collected (in Round 1 and later rounds)
on all related persons who join NHIS sampled households selected into the MEPS.
Persons in NHIS sampled households selected in the MEPS who subsequently enter an
institution and leave the civilian, noninstitutionalized population of the United States will require
data collection during their stay in institutions that are nursing homes. Alternatively, persons in
NHIS sampled households selected in the MEPS who subsequently enter institutions that are not
nursing homes and leave the civilian, noninstitutionalized population of the United States do not
require any data collected in these institutions that are not nursing homes (this also applies for
military service or moving out of the U.S.), but their whereabouts must be monitored during the
field period. Upon their return to the U.S. civilian noninstitutional population, these persons shall
once again be subject to HC data collection.
MEPS Data Collection Eligibility: In order for a MEPS reporting unit to be eligible for
data collection, the unit must include at least one individual who is key and in-scope for some
period of time during the reference period for a given round of data collection. If this condition
holds, the persons who are key and in-scope and all other individuals who are members of the
reporting unit (living together and related by blood, marriage, adoption or other family
associations) are eligible for data collection in a given round of the MEPS.
Return To Table Of Contents
2.5 Sample Size Targets and Precision Requirements
The 1996 MEPS sample size targets require approximately 9,000 reporting units yielding
the complete series of core interviews (i.e., Rounds 1-3) to obtain use and expenditure data for
calendar year 1996. The expected yield at each of the stages of data collection for each new
MEPS sample linked to the NHIS is: (1) a NHIS response rate of 94 percent at the household
level; (2) a response rate of 85 percent (83 percent achieved for the 1996 MEPS) among
reporting units at Round 1 (conditioned on a completed NHIS interview); a round-specific
response rate of 95 percent among reporting units at Rounds 2; a round-specific response rate of
97.5 percent among reporting units at Round 3; a round-specific response rate of 97 percent
among reporting units at Rounds 4 and 5; and a round-specific response rate among reporting
units of 98 percent at Round 6 (See Table 1). Consequently, the targeted response rate for
obtaining calendar year 1996 data on health care utilization and expenditures from the 1996
MEPS sample is 77 percent, conditioned on response to the NHIS (interviews for Rounds 1-3),
or 72 percent overall.
The response rate target for the core MEPS household survey for obtaining calendar year
1997 data on health care utilization and expenditures from the new 1997 MEPS sample is 79
percent conditioned on response to the NHIS (interviews for Rounds 1-3), or 74 percent overall
(See Table 1). Furthermore, the minimum acceptable response rate target for the core MEPS
household survey within a PSU is 65 percent for calendar year 1997 data from the new MEPS
panel, conditioned on NHIS response (interviews for Rounds 1-3), and is 60 percent for calendar
years 1996 and 1997 for the 1996 MEPS panel (interviews for Rounds 1-5, conditioned on
response to the NHIS).
It should be noted that the 1995 NHIS response rate achieved for the households eligible
for the MEPS was 94 percent. Of 10,639 responding NHIS dwelling units eligible for the MEPS,
99.6 percent were identified with the necessary information to facilitate MEPS data collection.
Of the 11,424 eligible reporting units targeted for interviews in Round One, 9,488 responded to
the first core MEPS interview (83.1 percent). Overall, the joint NHIS - Round One response rate
for the 1996 MEPS household survey was 77.7 percent (.939 x .996 x .831).
Table 1. Expected number of responding reporting units and associated response rate for each
round of data collection of the 1996 and the 1997 MEPS Household Component.
|
1995 NHIS
Linked
Sample |
Calendar Year
1996 |
Calendar Year
1997 |
Calendar Year
1998 |
1996 MEPS
Panel |
|
Round
1A |
Round
2A |
Round
3A |
Round
4A |
Round
5A |
Round
6A |
Responding
Reporting Units
(by Round)
(Response rate by Round) |
111,424
210,800
(94%) |
19,488
(83%)
|
39,018
(95%) |
38,792
(97.5%) |
38,528
(97%) |
38,272
(97%) |
38,106
(98%) |
|
1996 NHIS
Linked
Sample |
|
Calendar Year
1997 |
Calendar Year
1998 |
1997 MEPS
Panel |
|
|
|
Round
1B |
Round
2B |
Round
3B |
Round
4B |
Responding Reporting Units
(by Round)
(Response rate by Round) |
16,857
26,480
(94%) |
|
|
15,828
(85%) |
35,536
(95%) |
35,397
(97.5%) |
35,235
(97%) |
1 Includes Round 1 "splits-offs" (family member(s) that move apart from the originally sampled household) in Round 1 of the
1996 and 1997 MEPS panels.
2 Original sample of Reporting Units.
3 Does not include new split-offs after Round 1 in counts.
The estimates of response rates in Table 1 are for the original sample of NHIS responding reporting units, with the inclusion of
split-offs in Round 1. The rates specified in the table are also expected to apply to split-offs in subsequent rounds, i.e.,
households that will be created in the course of the survey field period as a result of key persons moving away from originally
sampled NHIS households.
Source: Agency for Health Care and Policy Research. 1996 Medical Expenditure Panel Survey--Household Component.
The sample size specifications have been set to meet precision requirements developed
for the MEPS. Given the major changes in the design of the survey that were required as a
consequence of the DHHS Survey Integration Plan, the sample size constraints placed on the
MEPS as a consequence of restricting the sample to the 195 PSU NHIS subsample, and use of
the first quarter of the 1995 NHIS sample for inclusion in a Disability Survey sponsored by the
Assistant Secretary of Planning and Evaluation, DHHS, the precision requirements for the first
year of the MEPS were relaxed relative to the original design specifications of the NMES-3
(Cohen, 1996; DiGaetano, 1994).
For the 1996 MEPS sample, the relative standard error for a population estimate of 20
percent for the overall population at the family level was specified to be no more than 2.7
percent; and the relative standard error for a population estimate of 20 percent for the overall
population at the person level was specified to be no more than 1.7 percent. For example, if it
was determined that the national population estimate of the percentage of the population ever
uninsured in 1996 was 20 percent, the standard error of the estimate should not exceed 0.34
percent. That would translate to a 95 percent confidence interval of (19.33%, 20.67%) for the
insurance coverage estimate that characterized the nation at the person level. Preliminary
design work suggested that a 1996 MEPS sample selected from a nationally representative
1995 NHIS subsample characterized by 195 PSUs, 1,675 segments, and approximately 9,000
responding households at the end of Round 3, with disproportionate sampling rates that ranged
from 1.0 to 0.5, should yield average design effects for MEPS survey estimates of annual use
and expenditure measures in the 1.5-1.6 range.
The 1996 MEPS sample linked to the NHIS was designed to produce unbiased
estimates for the four Census regions. This NHIS linked sample reflects an oversample of
Hispanics and blacks at the following ratios of representation relative to the remaining
households (Hispanics 2.0:1, blacks 1.5:1). The overall expected sample yield after three
rounds of data collection at the person level is approximately 22,000 overall, with 3,400
black/non-Hispanic individuals and 4,200 Hispanic individuals. The average design effect
target for survey estimates for the 1996 MEPS is 1.6. The sample design should satisfy the
following precision requirements for mean estimates of the following measures of health care
utilization and expenditures at the person level: (total health expenditures; utilization and
expenditure estimates for inpatient hospital stays; physician visits; dental visits and prescribed
medicines).
Demographic group |
Persons at the end
of Round 3 |
Average relative
standard error |
1. Black/Non-Hispanics |
3,400 |
.065 |
2. Hispanics |
4,200 |
.055 |
3. Overall Population |
22,000 |
.025 |
Return To Table Of Contents
2.6 Procedures for Data Collection
For a description of the preliminary contact with households responding to the NHIS
and subsampled as part of a MEPS panel, see "Design and Methods of the Medical
Expenditure Panel Survey, Household Component" by Joel Cohen.
HC Main Rounds 1-5
Five interviews will be conducted with each NHIS panel selected for the MEPS at
three- to four-month intervals over an approximately 24-month field period. The first three of
these rounds (Rounds 1A-3A) define the 1996 MEPS Household Component, and will collect
the main body of annual utilization and expenditure data for calendar year 1996. Rounds 3A-5A of the 1996 MEPS panel will be combined with Rounds 1B-3B of the 1997 MEPS panel to
yield the sample base for the 1997 MEPS Household Component and the source of annual
estimates for that calendar year. All interviews will be conducted in person through a
Computer-assisted Personal Interview (CAPI) as the principal data collection mode. Round 1
will ask about the period since January 1 of the MEPS year to the date of that interview;
Round 2 will ask about the time since the Round 1 interview through the date of the Round 2
interview; and Round 3 will collect data since the date of the Round 2 interview through the
date of the Round 3 interview in 1997.
Questionnaires for these field rounds will parallel those used in 1987 NMES with some
modifications implemented for the 1992 Feasibility Study and with further changes indicated
by the latter experience and the FAMES (NMES-3) pretest. The instruments contain items that
are asked once in the life of the study, items that are asked repeatedly in each round, and items
that are updated in later rounds. Questions asked only once include basic sociodemographic
characteristics. Core questions asked repeatedly include health status, health insurance
coverage, employment status, days of restricted activity due to health problems, medical
utilization, hospital admissions, and purchase of medicines. For each health encounter
identified, data will be obtained on the nature of health conditions, the characteristics of the
provider, the services provided, the associated charges, and sources and amounts of payment.
Permission forms for medical providers and for sources of employment and private
health insurance coverage will be collected in the field. Under this design, anyone who reports
being employed but not covered by private health insurance will be asked to sign a permission
form that will allow contact with the employer. A sample of medical providers identified by
MEPS respondents will be contacted in the survey of medical providers, MPC, to verify and
supplement information provided by the family respondent in the household interview;
employers and other health insurance providers will be contacted in the survey of health
insurance providers, IC, to verify analogous insurance information and to collect other
information on insurance characteristics that household respondents would not typically know.
As a consequence of a successful test in the Feasibility Study, copies of policies
providing private insurance coverage to sampled persons will be collected from household
respondents. These requests will be initiated in Round 1 and will be followed up in Round 2
for eligible individuals who have not provided copies of their policies at the time of the first
request. Sampled persons will be asked to provide the policies directly or to obtain them from
their health insurance providers. A description of the type of documents to be collected, a list
of the policies identified by the respondent, and request forms to be given to providers will be
given to interviewing staff to assist in this effort.
HC Main Round 6
Round 6 is concerned with obtaining valuable ancillary information before a MEPS
panel is retired. For the 1997 MEPS, it will take place after April 15, 1998 and ask for tax
filing information details. Comparable information will be collected for the 1996 panel in
Round 4 after April 15, 1997. Administration of the majority of Round 6 interviews will be by
telephone from the interviewers' homes; in-person interviews will be conducted for those
respondents without access to a suitable telephone or for those for whom telephone
administration is not feasible, e.g., respondents with hearing or comprehension problems.
Return To Table Of Contents
3.0 Summary
The benefits of the redesigned MEPS include significant cost savings, enhanced
analytical capacities, increased opportunities for longitudinal analyses, reduction of major data
gaps, and major improvements in providing timely data access to the research community at
large. The MEPS will provide information to help understand how the dramatic growth of
managed care, changes in private health insurance, and other dynamics of today's market-driven health care delivery system have affected, and are likely to affect, the kinds, amounts,
and costs of health care that Americans use. The survey will also provide necessary data for
projecting who benefits from and who bears the cost of changes to existing health policy and
the creation of new policies.
The MEPS data will serve as the primary source to inform research efforts which
examine how health care use and expenditures vary among different sectors of the population,
such as the elderly, veterans, children, disabled persons, minorities, the poor, and the
uninsured; and how the health insurance of households varies by demographic characteristics,
employment status and characteristics, geographic locale, and other factors. The MEPS data
will provide answers to questions about private health insurance costs and coverage, such as
how employers' costs vary by region, and help evaluate the growing impact of managed care
and of enrollment in different types of managed care plans.
The first MEPS data will be available on public use data tapes starting as early as
spring 1997. MEPS data also will be used in a series of studies to be published by AHCPR,
and by Agency and other researchers publishing in the scientific literature. As a consequence
of the shift to a continuous ongoing annual survey, additional efficiencies in survey data
collection, data editing and imputation tasks will be realized, as well as further improvements
in the timely release of MEPS data products to the research community.
Return To Table Of Contents
4.0 References
Arnett RA, Hunter E, Cohen S, Madans J, Feldman J. The Department of Health and Human
Services' Survey Integration Plan. Proceedings of the American Statistical Association,
section on Government Statistics; 1996 Aug; Chicago (IL).
Cohen SB. The redesign of the Medical Expenditure Panel Survey: A component of the
DHHS Survey Integration Plan. Proceedings of the COPAFS Seminar on Statistical
Methodology in the Public Service; 1996 Nov; Bethesda (MD).
DiGaetano R. Sample design of the Household Component of the National Medical
Expenditure Survey (NMES-3). Draft report. Rockville (MD): Westat, Inc.; 1994.
Hunter E, Arnett R, Mathiowetz N, Cohen S, Madans J, Feldman J. HHS Survey Integration
Plan: Background materials, 1995.
Judkins D, Marker D, Waksberg J. National Health Interview Survey: Research for the 1995
redesign. Draft report prepared for the National Center for Health Statistics. Westat, Inc.;
1994.
Return To Table Of Contents
Household Survey Sample Design Report
Appendix 2
Household Survey Design and Methods Report
Design and Methods of The Medical Expenditure Panel Survey
Household Component
March 1997
Joel Cohen
Agency for Health Care Policy and Research
Center for Cost and Financing Studies
2101 E. Jefferson Street, Suite 501
Rockville, Maryland 20852
(301) 594-1406
The Medical Expenditure Panel Survey (MEPS), is the third in a series of nationally
representative surveys of medical care use and expenditures sponsored by the Agency for Health
Care Policy and Research (formerly the National Center for Health Services Research). The first
of these surveys, called the National Medical Care Expenditure Survey (NMCES) was conducted
in 1977, and the second, called the National Medical Expenditure Survey (NMES), in 1987. The
1996 MEPS, which is co-sponsored by the National Center for Health Statistics (NCHS), will
update the 1987 data to reflect the dramatic changes that have occurred in the U.S. health care
system over the last decade.
Major changes have taken place in the health care delivery system of the nation since the
last NMES survey was conducted almost ten years ago. The most notable is the rapid expansion
of managed care arrangements such as HMOs, PPOs, and other provider networks that seek to
minimize the increases in health care costs, as well as the appearance of new hybrid forms of
health insurance coverage. Changes such as these have affected both the private and public
sectors. The new MEPS is needed to provide information about the current state of the health
care system in the U.S., and the changes that have taken place since the last national survey of
medical expenditures was conducted in 1987. The information collected by the MEPS will also
provide valuable baseline data for use in evaluating future changes in the system.
The revised MEPS study design enhances the
capabilities to study change over time and the effects of new health policies.
These are important objectives in view of the various health
reform initiatives that are being implemented by states and the Federal government.
The revised design allows for the production of annual estimates for two
calendar years, and also permits the
tracking of changes in employment, income, health status, and medical care use
and expenditures over the two consecutive years during which households in
the 1996 panel will be interviewed.
In addition, the National Health Interview Survey (NHIS) baseline data are available
for persons in the 1996 and 1997 MEPS panels, thereby adding another data
point for comparisons of change
over time.
The MEPS extends the NMES series of studies on medical expenditures and health
insurance, and provides for the first time, data suitable for detailed analysis of trends and changes
in these areas. The survey is a unique resource for a number of reasons, including:
(1) Scope. MEPS provides information on a broad spectrum of the population, as the
survey sample base represents the civilian noninstitutionalized population and, in a
separate component survey, the population institutionalized in nursing homes. The
MEPS also provides information on all types of health care services, expenditures, and
sources of payment for both individuals and families.
(2) Population Basis. The fact that MEPS is a survey of persons allows population
groups that are or may become of special policy concern to be identified and analyzed.
This is especially important for analyzing the effect of particular eligibility requirements
on the enrollment and budgets of public programs and on those who are not eligible for
such programs.
(3) Cost-effectiveness. MEPS will collect data needed by groups that might otherwise
sponsor separate or overlapping surveys, or do without crucial information needed for
important decisions. Experience has demonstrated that broad-based data on use,
expenses, and financing of health care collected from a nationally representative sample
can meet the data needs of a wide variety of users in a cost-effective manner.
The original sample design of the NMES household surveys has been revised for the
MEPS. Instead of defining the MEPS sample through an initial screening round, the sample in
the new design is selected as a nationally representative subsample from households that
participated in the NHIS. The 1996 MEPS sample (based on the 1995 NHIS) will be carried
forward into 1997 and combined with a new subsample of households responding to the 1996
NHIS. These two panel samples (the 1996 MEPS sample and the new MEPS selections from the
1996 NHIS) will jointly define the sample base for the 1997 MEPS Household Component.
Exhibit 1 is a diagram of the study design for the 1996 and 1997 MEPS Household Components.
Exhibit 2 summarizes various features of the study design for the Household Component.
In 1996, the MEPS sample linked to the 1995
NHIS was selected from a nationally representative NHIS subsample that included
195 PSU's and approximately 1,700 segments,
yielding approximately 10,500 responding NHIS households that MEPS recontacted.
This NHIS subsample reflects an oversample of Hispanics and blacks. Other
groups with high public policy
relevance in the areas of health care use and financing are targeted for oversample
as part of the MEPS 1997 panel to improve the precision of the estimates
for those groups.
Households selected for participation in the 1996 or the 1997 MEPS household surveys
are interviewed in person five times (Rounds 1-5), and a last time during a brief telephone
interview (Round 6). The rounds of data collection are spaced approximately 4 months apart.
The interviews take place with a family respondent who reports for him/herself and for other
family members.
Preliminary Contact. Mail and telephone contacts take place prior to the first MEPS
interview (Round 1) with the NHIS participating households selected for each MEPS panel. The
purpose of the Preliminary Contact is to enlist the household respondent into the MEPS study
and plan for the delivery of study record-keeping materials prior to the start of the study
observation period on January 1st of the survey year. An advance letter announcing the MEPS
survey is mailed in December to the family respondent at the address where the NHIS interview
was conducted. That letter is followed up with an interviewer telephone call to confirm the
arrival of the letter, verify the identity of the household, identify the MEPS family respondent (if
different from the NHIS respondent), and announce the future mailing of a study calendar and
record file. These materials are sent accompanied by $5 to compensate respondents for the time
and effort devoted to keeping records in preparation for the Round 1 interview. A second
telephone call confirms the arrival of these materials and arranges for the most convenient time
to conduct the Round 1 interview.
Households without telephones or those that can not be reached using the telephone number from
NHIS, are contacted by mail and asked to return a postcard identifying a telephone number where
the study can contact them (e.g., number at work, neighbor's house, etc.).
Core rounds. Data collection
for the MEPS Household Component takes place using the Computer-assisted
Personal Interview (CAPI) system. The study instrumentation is organized
as
a core instrument that is administered in each of the first 5 rounds of data
collection, with periodic supplements added in selected rounds to deal with
specific topics in greater depth.
Dependent interviewing methods, in which respondents are asked to confirm or
revise data provided in earlier interviews will be used to update information
in several of the core
questionnaires, such as employment and health insurance, after the initial interview.
Core Instrument: The core instrument will collect data about all persons in sampled
households. The core instrument includes questionnaires on: demographics, health status
and conditions, utilization, charges and payments, prescribed and over-the-counter
medicines purchased, employment, and health insurance.
Periodic Supplements: Supplements scheduled for inclusion in the survey include
questionnaires on: access to care and satisfaction, income and assets, long-term care, and
alternative care.
Self-Administered Questionnaire (SAQ): All adults in sample households are asked to
complete an SAQ in Round 2. This questionnaire collects information about health
behaviors and opinions that would be difficult if not impossible to collect on a proxy
basis from the family respondent. Similar information is collected for children as part of
the regular interview with the household survey respondent, usually the mother.
Medical Provider Permission Forms: Requests for signed permission forms take place in
Round 1 of the survey, much earlier than in past NMES studies, in order to expedite the
timetable for the later Medical Provider Component (MPC) of the survey, which collects
data about specific medical events directly from providers. Because results from a
previous methodological study suggested that early requests for signed permission forms
involving office-based physicians have a modest negative effect on survey cooperation
rates in later rounds, the requests for signed permission forms in Round 1 will be limited
to events taking place in hospitals. In Round 2 and subsequent rounds, requests for
signed permission forms will apply to all types of MPS-eligible medical providers
(hospitals, physicians, and home health agencies), including those associated with
utilization reported in Round 1.
Health Insurance Permission Forms: Signed permission forms are needed to contact
sources of employment and private health insurance coverage in the Insurance
Component of the survey, which collects data directly from individuals' sources of health
insurance (typically their employers). These requests will be initiated in Round 2, and
apply to the insurance sources associated with plans held at the time of the Round 1
interview.
Health Insurance Policy Booklet Requests: Following procedures tested successfully in a
previous methodological study, MEPS interviewers will attempt to secure, directly from
respondents, health insurance booklets or other summary materials that describe the
characteristics of private plans held by family members at the time of the Round 1
interview. The requests for policy information will include all sources of private
insurance coverage, not just employment-related coverage. Respondents are reimbursed
$15 for the time and effort involved in procuring policy booklets.
Provider Directories: To expedite the identification of medical providers and assist with
the preparation of an unduplicated list of medical providers for the fielding of MPS,
interviewers use a computerized database (directory) of health providers that has been
loaded into the CAPI laptop. Search software also loaded into the laptops enables
interviewers to query the database of providers in the course of the MEPS interview. If a
match is found in the database for the provider nominated by the household respondent,
the matched directory record is associated with the household member. Directory records
include the following information for each provider: a unique provider ID; the provider's
name, address and telephone number; and the provider's specialty (for individual office-based physicians).
At the most basic level, the objective of
the MEPS Household Component is the collection of data that can be used to
produce annual estimates for a variety of measures related
to the characteristics of individuals, their health insurance coverage, and their
health care use, expenditures, and sources of payment for care. The data
can also be used to support behavioral
analyses that inform researchers and policymakers about how the characteristics
of individuals and families, including their health insurance, affect medical
care use and spending.
Data obtained in this study will be used to produce, for example, the following national
estimates for calendar years 1996 and 1997:
annual estimates of health care use and expenditures for persons and families.
annual estimates of sources of payment for health care expenses, including amounts paid
by public programs, such as Medicare and Medicaid, and by private insurance, as well as
out-of-pocket payments.
annual estimates of health care use, expenditures and sources of payment for persons and
families by type of service, including: inpatient hospital stays, ambulatory care, home
health care, dental care, and purchases of prescribed and over-the-counter medicines.
the number and characteristics of the population eligible for each of the public programs,
including the use of services and expenditures of the population eligible for benefits
under Medicare, Medicaid, CHAMPUS/VA and the Veterans Administration.
the number, characteristics, use of services, expenditures and benefits of persons and
families with individual or group coverage, commercial and nonprofit coverage, and
coverage through HMOs or other managed care arrangements.
In addition to national estimates, data collected in this longitudinal study will be used to
study the determinants of the use of services and expenditures, and the effects of individual
characteristics and policy changes on medical care use and expenses. These behavioral analyses
will include studies of:
social and demographic factors such as employment and income.
methods of financing health care and health insurance.
the health habits, life styles and behavioral patterns of individuals and families.
the health needs of specific subpopulation groups of current or potential policy interest,
such as the elderly and members of racial or ethnic minorities.
Finally, data collected in this survey in conjunction with data from the 1977 NMCES and
the 1987 NMES will be used to study trends in the nature and distribution of national health
expenditures, sources of care, and amounts and types of services consumed by the U.S.
noninstitutionalized population.
Return To Table Of Contents
Exhibit 1. Panel Design for the MEPS Household Component, 1996 and 1997
|
Calendar Year
1996 |
Calendar Year
1997 |
Calendar Year
1998 |
1996 Panel (from 1995 NHIS) |
Round
1 |
Round
2 |
Round
3 |
Round
4 |
Round
5 |
Round
6 |
Field period |
3/96 -
7/96 |
8/96 -
11/96 |
2/97 -
5/97 |
8/97-
11/97 |
2/98 -
5/98 |
6/98 -
7/98 |
Responding households |
9,500 |
9,000 |
8,800 |
8,500 |
8,300 |
8,100 |
1997 Panel (from 1996 NHIS) |
|
|
Round
1 |
Round
2 |
Round
3 |
Round
4 |
Field period |
|
|
3/97 -
7/97 |
8/97 -
11/97 |
2/98 -
5/98 |
8/98-
11/98 |
Responding households |
|
|
5,800 |
5,500 |
5,400 |
5,200 |
Total Responding Households |
9,500 |
9,000 |
14,600 |
14,000 |
13,700 |
13,300 |
Return To Table Of Contents
Exhibit 2. Design Features of the MEPS Household Component, 1996 Panel
Feature |
1995 |
1996 |
1997 |
1998 |
Data
collection |
Preliminary
contact |
Round
1 |
Round
2 |
Round
3 |
Round
4 |
Round
5 |
Round
6 |
Reference period |
- |
1/1/96 to date of
Round 1
interview |
Date of Round 1
interview to date
of Round 2
interview |
Date of Round 2
interview to date
of Round 3
interview |
Date of Round 3
interview to date
of Round 4
interview |
Date of Round 4
interview to
12/31/97 |
- |
Field period |
12/95 -
1/96 |
3/96 -
7/96 |
8/96 -
11/96 |
2/97 -
5/97 |
8/97 -
11/97 |
2/98 -
5/98 |
6/98 -
7/98 |
Interview
mode |
Mail /
Telephone |
In-person/
CAPI |
In-person/
CAPI |
In-person/
CAPI |
In-person/
CAPI |
In-person/
CAPI |
Telephone |
Return To Table Of Contents
Return to the
MEPS Homepage
|