Return To Table Of Contents

• A list of employers or other insurance providers identified by MEPS HC respondents who report having private health insurance at the Round 1 interview.
   
• A Bureau of the Census list frame of private-sector business establishments.
   
• The Census of Governments from Bureau of the Census.

To provide an integrated picture of health insurance, data collected from the first sampling frame (employers and insurance providers identified by MEPS HC respondents) are linked back to data provided by those respondents. Data from the two Census Bureau sampling frames are used to produce annual national and state estimates of the supply and cost of private health insurance available to American workers and to evaluate policy issues pertaining to health insurance. National estimates of employer contributions to group insurance from the MEPS IC are used in the computation of Gross Domestic Product (GDP) by the Bureau of Economic Analysis.

The MEPS IC is an annual survey. Data are collected from the selected organizations through a prescreening telephone interview, a mailed questionnaire, and a telephone follow-up for nonrespondents.

MEPS data are collected under the authority of the Public Health Service Act. They are edited and published in accordance with the confidentiality provisions of this act and the Privacy Act. NCHS provides consultation and technical assistance.

As soon as data collection and editing are completed, the MEPS survey data are released to the public in staged releases of summary reports, microdata files and compendiums of tables. Data are released through MEPSnet, an online interactive tool developed to give users the ability to statistically analyze MEPS data in real time. Summary reports and compendiums of tables are released as printed documents and electronic files. Microdata files are released as electronic files.

Selected printed documents are available through the AHRQ Publications Clearinghouse. Write or call:

Be sure to specify the AHRQ number of the document you are requesting.

Additional information on MEPS is available from the MEPS project manager or the MEPS public use data manager at the Center for Financing, Access and Cost Trends, Agency for Healthcare Research and Quality, 540 Gaither Road, Rockville, MD 20850 (301-427-1406).

This documentation describes one in a series of public use event files from the 2004 Medical Expenditure Panel Survey (MEPS) Household Component (HC). Released as an ASCII data file (with related SAS and SPSS programming statements) and a SAS transport file, the 2004 Other Medical public use event file provides information on the purchases of and expenditures for visual aids, medical equipment, supplies, and other medical items for a nationally representative sample of the civilian noninstitutionalized population of the United States. Data from the Other Medical event file can be used to make estimates of the Other Medical event utilization and expenditures associated with medical items for calendar year 2004. The file contains 30 variables and has a logical record length of 214 with an additional 2-byte carriage return/line feed at the end of each record. As illustrated below, this file consists of MEPS survey data obtained in the 2004 portion of Round 3, and Rounds 4 and 5 for Panel 8, as well as Rounds 1, 2, and the 2004 portion of Round 3 for Panel 9 (i.e., the rounds for the MEPS panels covering calendar year 2004).

The Other Medical event file contains one record for each type of medical item reported as being purchased or otherwise obtained by the household respondent during the specified reference period. It should be noted that reference periods for reporting expenditures vary by type of medical item obtained. Expenditure data for visual aids, insulin, and diabetic supplies and equipment are collected during Rounds 3, 4, and 5 of Panel 8 and Rounds 1, 2 and 3 of Panel 9. Therefore, for these items, each round is a reference period. Expenditure data for other medical items, which include ambulance services, orthopedic items, hearing devices, prostheses, bathroom aides, medical equipment, disposable supplies, and home alterations are collected only in Rounds 5 (Panel 8) and 3 (Panel 9); for these items, the reference period is the entire year.

The purchase of medical equipment, supplies, and other medical items is based entirely on household reports. They were not included in the Medical Provider Component (MPC); therefore, all expenditure and payment data on the Other Medical event file are reported by the household.

Data from this event file can be merged with other 2004 MEPS HC data files for the purpose of appending person-level data, such as demographic characteristics or health insurance coverage, to each other medical record.

This file can also be used to construct summary variables of expenditures, source of payment, and related aspects of the purchase of medical items. Aggregate annual person-level information on expenditures for other medical equipment is provided on the MEPS 2004 Full Year Consolidated Data File where each record represents a MEPS sampled person. This aggregate information is provided for vision aids only and not other types of other medical equipment.

Data users/analysts should be aware of the limitations of the Other Medical event file. These limitations include the following:

The following documentation offers a brief overview of the types and levels of data provided, and the content and structure of the files and the codebook. It contains the following sections:

For more information on MEPS HC survey design, see S. Cohen, 1997; J. Cohen, 1997; and S. Cohen, 1996. A copy of the MEPS HC survey instrument used to collect the information on the dental file is available on the MEPS web site at the following address: <http://www.meps.ahrq.gov>.

The 2004 Other Medical Expenses public use data set consists of one event-level data file. The file contains characteristics associated with the Other Medical event and imputed expenditure data.

The 2004 Other Medical public use data set contains 9,580 other medical expenditure records; of these records, 9,316 are associated with persons having a positive person-level weight (PERWT04F). This file includes records for all household survey respondents who resided in eligible responding households and reported purchasing or otherwise obtaining at least one type of medical item, such as medical equipment, glasses, hearing devices, etc., during calendar year 2004. Some household respondents may have reported obtaining more than one type of medical item and, therefore, have several records on this file. Likewise, respondents who did not report obtaining a medical item in 2004 have no records on this file. These data were collected during the 2004 portion of Round 3, and Rounds 4 and 5 for Panel 8, as well as Rounds 1, 2, and the 2004 portion of Round 3 for Panel 9 of the MEPS HC. The persons represented on this file had to meet either (a) or (b) below:

Persons with no other medical events for 2004 are not included on this event-level OM file but are represented on the person-level 2004 Full Year Population Characteristics file.

Each record includes the following: type of medical item obtained; flat fee information; imputed sources of payment; total payment and total charge for the medical item; and a full-year person-level weight.

Data from this file can be merged with the MEPS 2004 Full Year Population Characteristics File using the unique person identifier, DUPERSID, to append person-level information, such as demographic or health insurance characteristics, to each record. Please see section 5.0 for details on how to merge MEPS data files.

MEPS began in 1996 and several annual data files have been released. As more years of data are produced, MEPS will become increasingly valuable for examining health care trends. However, it is important to consider a variety of factors when examining trends over time using MEPS. Statistical significance tests should be conducted to assess the likelihood that observed trends are attributable to sampling variation. MEPS expenditure estimates are especially sensitive to sampling variation due to the underlying skewed distribution of expenditures. For example, 1 percent of the population accounts for about one-quarter of all expenditures. The extent to which observations with extremely high expenditures are captured in the MEPS sample varies from year to year (especially for smaller population subgroups), which can produce substantial shifts in estimates of means or totals that are simply an artifact of the sample(s). The length of time being analyzed should also be considered. In particular, large shifts in survey estimates over short periods of time (e.g. from one year to the next) that are statistically significant should be interpreted with caution, unless they are attributable to known factors such as changes in public policy or MEPS survey methodology. Looking at changes over longer periods of time can provide a more complete picture of underlying trends. Analysts may wish to consider using techniques to smooth or stabilize trend analyses of MEPS data such as pooling time periods for comparison (e.g. 1996-97 versus 1998-99), working with moving averages, or using modeling techniques with several consecutive years of MEPS data to test the fit of specified patterns over time. Finally, researchers should be aware of the impact of multiple comparisons on Type I error because performing numerous statistical significance tests of trends increases the likelihood of inappropriately concluding a change is statistically significant.

The records on this file can be linked to all other 2004 MEPS-HC public use data sets by the sample person identifier (DUPERSID).

For each variable on the Other Medical event file, both weighted and unweighted frequencies are provided in the codebook (files H85CCB.PDFand H85CCB.ASP). The codebook and data file sequence list variables in the following order:

The following reserved code values are used:

Generally, values of -1, -7, -8, and -9 for non-expenditure variables have not been edited on this file. The values of -1 and -9 can be edited by the data users/analysts by following the skip patterns in the HC survey questionnaire (located on the MEPS web site: <http://www.meps.ahrq.gov/survey.htm>).

The codebook describes an ASCII data set (although the data are also being provided in a SAS transport file). The following codebook items are provided for each variable:

In general, variable names reflect the content of the variable, with an eight-character limitation. All imputed/edited variables end with an "X".

Variables were derived from the HC survey questionnaire or from the CAPI. The source of each variable is identified in Section D "Variable - Source Crosswalk" in one of four ways:

The names of the expenditure and source of payment variables follow a standard convention, are seven characters in length, and end in an "X" indicating edited/imputed. Please note that imputed means that a series of logical edits, as well as an imputation process to account for missing data, have been performed on the variable.

The total sum of payments and 12 source of payment are named in the following way:

The first two characters indicate the type of event:

In the case of the source of payment variables, the third and fourth characters indicate:

In addition, the total charge variable is indicated by TC in the variable name.

The fifth and sixth characters indicate the year (04). The seventh character, "X", indicates whether the variable is edited/imputed.

For example, OMSF04X is the edited/imputed amount paid by self or family for 2004 other medical equipment and expenditures.

The dwelling unit ID (DUID) is a five-digit random number assigned after the case was sampled for MEPS. The three-digit person number (PID) uniquely identifies each person within the dwelling unit. The eight-character variable DUPERSID uniquely identifies each person represented on the file and is the combination of the variables DUID and PID. For detailed information on dwelling units and families, please refer to the documentation for the 2004 Full Year Population Characteristics File.

EVNTIDX uniquely identifies each other medical expense event (i.e., each record on the OME file) and is the variable required to link other medical events to data files containing details on prescribed medicines (the MEPS 2004 Prescribed Medicines File). For details on linking, see Section 5.0, or the MEPS 2004 Appendix File, HC-085I.

FFEEIDX is a constructed variable that uniquely identifies a flat fee group, that is, all events that were part of a flat fee payment. FFEEIDX identifies a flat fee payment that was identified using information from the Household Component.

EVENTRN indicates the round in which the other medical event was reported. For most types of other medical expenditures on this file, data were collected only in Round 5 for Panel 8 and Round 3 for Panel 9; each record represents a summary of expenditures for items purchased or otherwise obtained for 2004. There are two exceptions:

Other medical expenditures (OMTYPE) include glasses or contact lenses, insulin, diabetic equipment/supplies, ambulance services, orthopedic items, hearing devices, prostheses, bathroom aids, medical equipment, disposable supplies, and alterations/modifications (to homes). When the interviewer did not know how to categorize types of medical item expenditures, these items were specified in the variable OMOTHOS (OMTYPE other specify). As a part of the editing process, other medical expenditures identified in OMOTHOS have been edited to appropriate OMTYPE categories. The edited (OMTYPEX, OMOTHOX) and unedited (OMTYPE, OMOTHOS) versions of both of these variables are included on this file.

A flat fee is the fixed dollar amount a person is charged for a package of services provided during a defined period of time. A flat fee group is the set of medical services that are covered under the same flat fee payment. The flat fee groups represented on the Other Medical file include flat fee groups where at least one of the other medical events, as reported by the HC respondent, occurred during 2004. By definition, a flat fee group can span multiple years. Furthermore, a single person can have multiple flat fee groups.

As noted earlier in Section 2.6.1.2 "Record Identifiers," the variable FFEEIDX uniquely identifies all events that are part of the same flat fee group for a person. On any 2004 MEPS event file, every event that is part of a specific flat fee group will have the same value for FFEEIDX. Note that prescribed medicine and home health events are never included in a flat fee group and none of the flat fee variables are on those event files.

FFOMTYPE indicates whether the 2004 other medical expenditure is the "stem" or "leaf" of a flat fee group. A stem (records with FFOMTYPE = 1) is the initial other medical service event, which is followed by other medical expense events that are covered under the same flat fee payment. The leaves of the flat fee group (records with FFOMTYPE = 2) are those other medical events that are tied back to the initial event (the stem) in the flat fee group. These "leaf" records have their expenditure variables set to zero. For the other medical events that are not part of a flat fee payment, the FFOMTYPE is set to -1, "INAPPLICABLE".

As described in Section 2.6.3.1, a flat fee payment covers multiple events and the multiple events could span multiple years. For situations where the medical item was obtained in 2004 as part of a group of events, and some of the events occurred before or after 2004, counts of the known events are provided on the other medical record. Variables that indicate events occurring before or after 2004 are the following:

If there are no 2003 events on the file, FFBEF04 will be omitted. Likewise, if there are no 2005 events on the file, FFTOT05 will be omitted. If there are no flat fee data related to the records in this file, FFEEIDX and FFOMTYPE will be omitted as well. Please note that the crosswalk in this document lists all possible flat fee variables.

Data users/analysts should note that flat fee payments are not common on the Other Medical file. There are only 13 records that are identified as being part of a flat fee payment group. In general, every flat fee group should have an initial visit (stem) and at least one subsequent visit (leaf). There are some situations where this is not true. For some of these flat fee groups, the initial visit reported occurred in 2004, but the remaining visits that were part of this flat fee group occurred in 2005. In this case, the 2004 flat fee group represented on this file would consist of one event (the stem). The 2005 "leaf events" that are part of this flat fee group are not represented on the file. Similarly, the household respondent may have reported a flat fee group where the initial visit began in 2003 but subsequent visits occurred during 2004. In this case, the initial visit would not be represented on the file. This 2004 flat fee group would then only consist of one or more leaf records and no stem.

Expenditures on this file refer to what is paid for the medical item. More specifically, expenditures in MEPS are defined as the sum of payments for each medical item that was obtained, including out-of-pocket payments and payments made by private insurance, Medicaid, Medicare and other sources. The definition of expenditures used in MEPS differs slightly from its predecessors: the 1987 NMES and 1977 NMCES surveys where "charges" rather than sum of payments were used to measure expenditures. This change was adopted because charges became a less appropriate proxy for medical expenditures during the 1990s due to the increasingly common practice of discounting. Although measuring expenditures as the sum of payments incorporates discounts in the MEPS expenditure estimates, these estimates do not incorporate any payment not directly tied to specific medical care events, such as bonuses or retrospective payment adjustments paid by third party payers. Another general change from the two prior surveys is that charges associated with uncollected liability, bad debt, and charitable care (unless provided by a public clinic or hospital) are not counted as expenditures because there are no payments associated with those classifications. While charge data are provided on this file, data users/analysts should use caution when working with this data because a charge does not typically represent actual dollars exchanged for services or the resource costs of those services, nor are they directly comparable to the expenditures defined in the 1987 NMES. For details on expenditure definitions, please refer to the following, "Informing American Health Care Policy" (Monheit et al., 2000). AHRQ has developed factors to apply to the 1987 NMES expenditure data to facilitate longitudinal analysis. These factors can be assessed via the CFACT data center. For more information see the Data Center section of the MEPS web site at <http://www.meps.ahrq.gov/datacenter.htm>. If examining trends in MEPS expenditures or performing longitudinal analysis on MEPS expenditures, please refer to section C, sub-section 2.1 for more information.

The general methodology used for editing and imputing expenditure data is described below. The MPC did not include either the dental events or other medical expenditures (such as glasses, contact lenses, and hearing devices). Therefore, although the general procedures remain the same for dental and other medical expenditures, editing and imputation methodologies were applied only to household-reported data. Please see below for details on the differences between these editing/imputation methodologies. Separate imputations were performed for flat fee and simple events, as well.

Logical edits were used to resolve internal inconsistencies and other problems in the HC survey-reported data. The edits were designed to preserve partial payment data from households and providers, and to identify actual and potential sources of payment for each household-reported event. In general, these edits accounted for outliers, copayments or charges reported as total payments, and reimbursed amounts that were reported as out-of-pocket payments. In addition, edits were implemented to correct for misclassifications between Medicare and Medicaid and between Medicare HMOs and private HMOs as payment sources. These edits produced a complete vector of expenditures for some events, and provided the starting point for imputing missing expenditures in the remaining events.

A weighted sequential hot-deck procedure was used to impute for missing expenditures, as well as total charge. This procedure uses survey data from respondents to replace missing data, while taking into account the respondents’ weighted distribution in the imputation process. Classification variables vary by event type in the hot-deck imputations, but total charge and insurance coverage are key variables in all of the imputations. Separate imputations were performed for nine categories of medical provider care: inpatient hospital stays, outpatient hospital department visits, emergency room visits, visits to physicians, visits to non-physician providers, dental services, home health care by certified providers, home health care by paid independents, and other medical expenses. Within each file, separate imputations were performed for flat fee and simple events. After the imputations were finished, visits to physician and non-physician providers were combined into a single medical provider file. The two categories of home care also were combined into a single home health file.

Expenditures on other medical equipment and services were developed in a sequence of logical edits and imputations. The household edits were used to correct obvious errors in the reporting of expenditures, and to identify actual and potential sources of payments. Some of the edits were global (i.e., applied to all events). Others were hierarchical and mutually exclusive. One of the more important edits separated flat fee events from simple events. This edit was necessary because groups of events covered by a flat fee (i.e., a flat fee bundle) were edited and imputed separately from individual events each covered by a single charge (i.e., simple events). Other medical services were imputed as flat fee events if the charges covered a package of health care services (e.g., optical), and all of the services were part of the same event type (i.e., a pure bundle). If a bundle contained any OM events with any other types of events, the services were treated as simple events in the imputations (See Section 2.6.3 for more detail on the definition and imputation of events in flat fee bundles.)

Logical edits were used to sort each event into a specific category for the imputations. Events with complete expenditures were flagged as potential donors for the hot-deck imputations, while events with missing expenditure data were assigned to various recipient categories. Each event with missing expenditure data was assigned to a recipient category based on the extent of its missing charge and expenditure data. For example, an event with a known total charge but no expenditure information was assigned to one category, while an event with a known total charge and partial expenditure information was assigned to a different category. Similarly, events without a known total charge and no or partial expenditure information were assigned to various recipient categories.

The logical edits produced nine recipient categories for events with missing data. Eight of the categories were for events with a common pattern of missing data and a primary payer other than Medicaid. Medicaid events were imputed separately because persons on Medicaid rarely know the provider’s charge for services or the amount paid by the state Medicaid program. As a result, the total charge for Medicaid-covered services was imputed and discounted to reflect the amount that a state program might pay for the care.

Separate hot-deck imputations were used to impute missing data in each of the other eight recipient categories. The donor pool included "free events" because in some instances, providers are not paid for their services. These events represent charity care, bad debt, provider failure to bill, and third party payer restrictions on reimbursement in certain circumstances. If free events were excluded from the donor pool, total expenditures would be over-counted because the distribution of free events among complete events (donors) is not represented among incomplete events (recipients).

IMPFLAG is a six-category variable that indicates if the event contains complete Household Component (HC) or Medical Provider Component (MPC) data, was fully or partially imputed, or was imputed in the capitated imputation process (for OP and MV events only). The following list identifies how the imputation flag is coded; the categories are mutually exclusive.

The approach used to count expenditures for flat fees was to place the expenditure on the first visit of the flat fee group. The remaining visits have zero payments. Thus, if the first visit in the flat fee group occurred prior to 2004, all of the events that occurred in 2004 will have zero payments. Conversely, if the first event in the flat fee group occurred at the end of 2004, the total expenditure for the entire flat fee group will be on that event, regardless of the number of events it covered after 2004. See section 2.6.3 for details on the flat fee variables.

Some respondents reported obtaining medical items where the payments were zero. This could occur for several reasons including (1) item or service was free, (2) bad debt was incurred, or (3) the item was covered under a flat fee arrangement beginning in an earlier year. If all of the medical events for a person fell into one of these categories, then the total annual expenditures for that person would be zero.

In addition to total expenditures, variables are provided which itemize expenditures according to major source of payment categories. These categories are:

Two additional source of payment variables were created to classify payments for events with apparent inconsistencies between insurance coverage and sources of payment based on data collected in the survey. These variables include:

Though relatively small in magnitude, data users/analysts should exercise caution when interpreting the expenditures associated with these two additional sources of payment. While these payments stem from apparent inconsistent responses to health insurance and source of payment questions in the survey, some of these inconsistencies may have logical explanations. For example, private insurance coverage in MEPS is defined as having a major medical plan covering hospital and physician services. If a MEPS sampled person did not have such coverage but had a single service type insurance plan (e.g., dental insurance) that paid for a particular episode of care, those payments may be classified as "other private." Some of the "other public" payments may stem from confusion between Medicaid and other state and local programs or may be from persons who were not enrolled in Medicaid, but were presumed eligible by a provider who ultimately received payments from the public payer.

Other medical expenditure data were obtained only through the Household Component Survey. For cases with missing expenditure data, other medical expenditures were imputed using the procedures described above. However, please note that expenditure data for insulin and diabetic supplies are not included on this file, but are included on the 2004 Prescribed Medicines File. Missing expenditure data associated with these records were not imputed. Charge and Payment variables in these cases carry a value of –1, "INAPPLICABLE".

OMSF04X - OMOT04X are the 12 sources of payment. OMTC04X is the total charge, and OMXP04X is the sum of the 12 sources of payment for the other medical expenditures. The 12 source of payment are: self/family (OMSF04X), Medicare (OMMR04X), Medicaid (OMMD04X), private insurance (OMPV04X), Veterans Administration (OMVA04X), TRICARE (OMTR04X), other Federal sources (OMOF04X), State and Local (non-federal) government sources (OMSL04X), Worker’s Compensation (OMWC04X), other private insurance (OMOR04X), other public insurance (OMOU04X), and other insurance (OMOT04X).

Expenditure variables on the 2004 other medical file have been rounded to the nearest penny. Person-level expenditure information released on the MEPS 2004 Person-Level Expenditure File will be rounded to the nearest dollar. It should be noted that using the MEPS event files to create person-level totals will yield slightly different totals than those found on the person-level expenditure file. These differences are due to rounding only. Moreover, in some instances, the number of persons having expenditures on the event files for a particular source of payment may differ from the number of persons with expenditures on the person-level expenditure file for that source of payment. This difference is also an artifact of rounding only. Please see the MEPS 2004 Appendix File, HC-085I, for details on rounding differences.

There is a single full year person-level weight (PERWT04F) assigned to each record for each key, in-scope person who responded to MEPS for the full period of time that he or she was in-scope during 2004. A key person either was a member of an NHIS household at the time of the NHIS interview, or became a member of a family associated with such a household after being out-of-scope at the time of the NHIS (the latter circumstance includes newborns as well as persons returning from military service, an institution, or living outside the United States). A person is in-scope whenever he or she is a member of the civilian noninstitutionalized portion of the U.S. population.

The person-level weight PERWT04F was developed in several stages. Person-level weights for Panels 8 and 9 were created separately. The weighting process for each panel included an adjustment for nonresponse over time and calibration to independent population figures. The calibration was initially accomplished separately for each panel by raking the corresponding sample weights to Current Population Survey (CPS) population estimates based on five variables. The five variables used in the establishment of the initial person-level control figures were: census region (Northeast, Midwest, South, West); MSA status (MSA, non-MSA); race/ethnicity (Hispanic, non-Hispanic with black as sole reported race, non-Hispanic with Asian as sole reported race, and other); sex; and age. A 2004 composite weight was then formed by multiplying each weight from Panel 8 by the factor .49 and each weight from Panel 9 by the factor .51. The choice of factors reflected the relative sample sizes of the two panels, helping to limit the variance of estimates obtained from pooling the two samples. The composite weight was again raked to the same set of CPS-based control totals. When poverty status information derived from income variables became available, a final raking was undertaken on the previously established weight variable. Control totals were established using poverty status (five categories: below poverty, from 100 to 125 percent of poverty, from 125 to 200 percent of poverty, from 200 to 400 percent of poverty, at least 400 percent of poverty) as well as the original five variables used in the previous calibrations.

The person-level weight for MEPS Panel 8 was developed using the 2003 full year weight for an individual as a "base" weight for survey participants present in 2003. For key, in-scope respondents who joined an RU some time in 2004 after being out-of-scope in 2003, the 2003 family weight associated with the family the person joined served as a "base" weight. The weighting process included an adjustment for nonresponse over Rounds 4 and 5 as well as raking to population control figures for December 2004. These control figures were derived by scaling back the population totals obtained from the March 2004 CPS to correspond to a national estimate for the civilian noninstitutionalized population provided by the Census Bureau for December 2004. Variables used in the establishment of person-level control figures included: census region (Northeast, Midwest, South, West); MSA status (MSA, non-MSA); race/ethnicity (Hispanic, black but non-Hispanic, Asian but non-Hispanic, and other); sex; and age. Overall, the weighted population estimate for the civilian noninstitutionalized population on December 31, 2004 is 289,659,890. Key, responding persons not in-scope on December 31, 2004 but in-scope earlier in the year retained, as their final Panel 8 weight, the weight after the nonresponse adjustment.

The person-level weight for MEPS Panel 9 was developed using the MEPS Round 1 person-level weight as a "base" weight. For key, in-scope respondents who joined an RU after Round 1, the Round 1 family weight served as a "base" weight. The weighting process included an adjustment for nonresponse over Round 2 and the 2004 portion of Round 3 as well as raking to the same population control figures for December 2004 used for the MEPS Panel 8 weights. The same five variables employed for Panel 8 raking (census region, MSA status, race/ethnicity, sex, and age) were used for Panel 9 raking. Similarly, for Panel 9, key, responding persons not in-scope on December 31, 2004 but in-scope earlier in the year retained, as their final Panel 9 weight, the weight after the nonresponse adjustment.

Note that the MEPS Round 1 weights (for both panels with one exception as noted below) incorporated the following components: the original household probability of selection for the NHIS; ratio-adjustment to NHIS-based national population estimates at the household (occupied dwelling unit) level; adjustment for nonresponse at the dwelling unit level for Round 1; and poststratification to figures at the family and person level obtained from the March 2004 CPS data base.

Variables used in the establishment of person-level control figures included: poverty status (below poverty, from 100 to 125 percent of poverty, from 125 to 200 percent of poverty, from 200 to 400 percent of poverty, at least 400 percent of poverty); census region (Northeast, Midwest, South, West); MSA status (MSA, non-MSA); race/ethnicity (Hispanic, non-Hispanic with black as sole reported race, non-Hispanic with Asian as sole reported race, and other); sex; and age. Overall, the weighted population estimate for the civilian noninstitutionalized population for December 31, 2004 is 289,659,890 (PERWT04F>0 and INSC1231=1). The weights of some persons out-of-scope on December 31, 2004 were also calibrated, this time using poststratification. Specifically, the weights of persons out-of-scope on December 31, 2004 who were in-scope some time during the year and also entered a nursing home during the year were poststratified to a corresponding control total obtained from the 1996 MEPS Nursing Home Component. The weights of persons who died while in-scope during 2004 were poststratified to corresponding estimates derived using data obtained from the Medicare Current Beneficiary Survey (MCBS) and Vital Statistics information provided by the National Center for Health Statistics (NCHS). Separate control totals were developed for the "65 and older" and "under 65" civilian noninstitutionalized populations. The sum of the person-level weights across all persons assigned a positive person level weight is 293,527,003.

The target population for MEPS in this file is the 2004 U.S. civilian noninstitutionalized population. However, the MEPS sampled households are a subsample of the NHIS households interviewed in 2002 (Panel 8) and 2003 (Panel 9). New households created after the NHIS interviews for the respective Panels and consisting exclusively of persons who entered the target population after 2002 (Panel 8) or after 2003 (Panel 9) are not covered by MEPS. Neither are previously out-of-scope persons who join an existing household but are unrelated to the current household residents. Persons not covered by a given MEPS panel thus include some members of the following groups: immigrants; persons leaving the military; U.S. citizens returning from residence in another country; and persons leaving institutions. The set of uncovered persons constitutes only a small segment of the MEPS target population.

This file is constructed for efficient estimation of utilization, expenditures, and sources of payment for other medical expenditures and to allow for estimates of number of persons who obtained medical items in 2004.

It is essential that the analyst examine all variables for the presence of negative values used to represent missing values. For continuous or discrete variables, where means or totals may be taken, it may be necessary to set minus values to values appropriate to the analytic needs. That is, the analyst should either impute a value or set the value to one that will be interpreted as missing by the computing language used. For categorical and dichotomous variables, the analyst may want to consider whether to recode or impute a value for cases with negative values or whether to exclude or include such cases in the numerator and/or denominator when calculating proportions.

Methodologies used for the editing/imputation of expenditure variables (e.g., sources of payment, flat fee, and zero expenditures) are described in Section 2.6.4.

While the examples described below illustrate the use of event-level data in constructing person-level total expenditures, these estimates can also be derived from the person-level expenditure file unless the characteristic of interest is event specific.

In order to produce national estimates related to other medical expense utilization, expenditures, and sources of payment, the value in each record contributing to the estimates must be multiplied by the weight (PERWT04F) contained on that record.

For example, the total number of other medical expense events for "GLASSES OR CONTACT LENSES" (OMTYPEX=1), for the civilian noninstitutionalized population of the U.S. in 2004 is estimated as the sum of the weight (PERWT04F) across all other medical expense event records with OMTYPEX=1. That is,

Subsetting to records based on characteristics of interest expands the scope of potential estimates. For example, the estimate for the mean out-of-pocket payment on "GLASSES OR CONTACT LENSES" (where the visit has a total expense greater than 0) should be calculated as the weighted mean of amount paid by self/family. That is,

This gives $149.09 as the estimated mean amount of out-of-pocket payment of expenditures associated with "GLASSES OR CONTACT LENSES" events and 53,316,048 as an estimate of the total number of such other medical expense events with expenditure. Both of these estimates are for the civilian noninstitutionalized population of the U.S. in 2004.

Another example would be to estimate the average proportion of total expenditures (where event expense is greater than 0) paid by private insurance per "GLASSES OR CONTACT LENSES" event. This should be calculated as the weighted mean of the proportion of the total "GLASSES OR CONTACT LENSES" expense paid by private insurance at the other medical expense event-level. That is,

This gives 0.1642 as the estimated mean proportion of total expenditures paid by private insurance for "GLASSES OR CONTACT LENSES" events with expenditure for the civilian noninstitutionalized population of the U.S. in 2004.

When calculating an estimate of the total number of persons with other medical expense events, users can use a person-level file or this event file. However, this event file must be used when the measure of interest is defined at the event-level. For example, to estimate the number of persons in the civilian noninstitutionalized population of the U.S. with a medical expense for ambulance service in 2004, this event file must be used. This would be estimated as

This file may be used to derive person-based ratio estimates. However, when calculating ratio estimates where the denominator is persons, care should be taken to properly define and estimate the unit of analysis up to person-level. For example, the mean expense for persons with other medical expense events is estimated as,

If the ratio relates to the entire population, this file cannot be used to calculate the denominator, as only those persons with at least one other medical expense event are represented on this data file. In this case, the Full Year Consolidated File, which has data for all sampled persons, must be used to estimate the total number of persons (i.e., those with events and those without events).

For example, to estimate the proportion of civilian noninstitutionalized population of the U.S. with at least one other medical event for ambulance services, the numerator would be derived from data on this event file, and the denominator would be derived from data on the person-level file. That is,

There have been several previous releases of MEPS Household Survey public use data. Unless a variable name common to several files is provided, the sampling weights contained on these data files are file-specific. The file-specific weights reflect minor adjustments to eligibility and response indicators due to birth, death, or institutionalization among respondents.

For estimates from a MEPS data file that do not require merging with variables from other MEPS data files, the sampling weight(s) provided on that data file are the appropriate weight(s). When merging a MEPS Household data file to another, the major analytical variable (i.e., the dependent variable) determines the correct sampling weight to use.

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. Various approaches can be used to develop such estimates of variance including use of the Taylor Series or various replication methodologies. Replicate weights have not been developed for the MEPS 2004 data. Variables needed to implement a Taylor Series estimation approach are provided in the file and are described in the paragraph below.

Using a Taylor Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The corresponding variables on the MEPS full year utilization database are VARSTR and VARPSU, respectively. Prior to 2003, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of the strata and PSU variable names denoted the year. However, beginning with the 2003 Point-in-Time PUF, the variance strata and PSUs have been developed to be compatible with all future PUFs. Thus, data from future years can be pooled and the variance strata and PSU variables provided can be used without modification for variance estimation purposes for estimates covering multiple years of data. There are 203 variance estimation strata, each stratum with either two or three variance estimation PSUs. Specifying a "with replacement" design in a computer software package such as SUDAAN (Shah, 1996) should provide 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 actual number available. For MEPS sample estimates for characteristics generally distributed throughout the country (and thus the sample PSUs), there are over 100 degrees of freedom associated with the corresponding estimates of variance. The following illustrates these concepts using two examples from Section 4.2.

Using a Taylor Series approach, specifying VARSTR and VARPSU as the variance estimation strata and PSUs (within these strata) respectively and specifying a "with replacement" design in a computer software package (i.e., SUDAAN) will yield standard error estimates of $2.92 and 0.0068 for the estimated mean of out-of-pocket payment and the estimated mean proportion of total expenditures paid by private insurance respectively.

Data from this file can be used alone or in conjunction with other files. This section provides instructions, or the details on where to find the instructions, for linking the 2004 other medical provider events with other 2004 MEPS public use files, including the 2004 prescribed medicines file and a 2004 person-level file.

Merging characteristics of interest from other MEPS files (e.g., 2004 Full Year Population Characteristics File or 2004 Prescribed Medicines) expands the scope of potential estimates. For example, to estimate the expenditures for medical equipment, visual aids, etc. for persons with specific demographic characteristics (such as age, race, and sex), population characteristics from a person-level file need to be merged onto the Other Medical event file. This procedure is shown below. The MEPS 2004 Appendix File, HC-085I, provides additional details on how to merge other MEPS data files.

Due to survey design issues, there are limitations/caveats that data users/analysts must keep in mind when linking the different files. These limitations/caveats are listed below. For detailed linking examples, including SAS code, data users/analysts should refer to the MEPS 2004 Appendix File, HC-085I.

The RXLK file provides a link from the MEPS event files to the prescribed medicine records on the 2004 Prescribed Medicine Event File. When using RXLK, data users/analysts should keep in mind that one other medical record can link to more than one prescribed medicine record. Conversely, a prescribed medicine event may link to more than one other medical record. When this occurs, it is up to the data user/analyst to determine how the prescribed medicine expenditures should be allocated among those other medical expenses.

Cohen, S.B. (1997). Sample Design of the 1996 Medical Expenditure Panel Survey Household Component. Rockville (MD): Agency for Health Care Policy and Research; 1997. MEPS Methodology Report, No. 2. AHCPR Pub. No. 97-0027.

Cohen, J.W. (1997). Design and Methods of the Medical Expenditure Panel Survey Household Component. Rockville (MD): Agency for Health Care Policy and Research; 1997. MEPS Methodology Report, No. 1. AHCPR Pub. No. 97-0026.

Cohen, S.B. (1996). 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.

Monheit, A.C., Wilson, R., and Arnett, III, R.H. (Editors) (1999). Informing American Health Care Policy. Jossey-Bass Inc., San Francisco.

Shah, B.V., Barnwell, B.G., Bieler, G.S., Boyle, K.E., Folsom, R.E., Lavange, L., Wheeless, S.C., and Williams, R. (1996). Technical Manual: Statistical Methods and Algorithms Used in SUDAAN Release 7.0, Research Triangle Park, NC: Research Triangle Institute.