MCBS Chartbook

Charts to explore the Medicare population

Download this Chart

Download this Chart

Download this Chart

Download this Chart

About the Chartbook


Medicare is the nation's health insurance program for persons aged 65 years and over and for persons younger than 65 years who have a qualifying disability. People under age 65 who receive Social Security Disability Insurance (SSDI) payments generally become eligible for Medicare after a two-year waiting period, while those diagnosed with end-stage renal disease (ESRD) and amyotrophic lateral sclerosis (ALS) become eligible for Medicare with no waiting period. Medicaid is a federal-state health insurance program for low-income Americans. Dually eligible beneficiaries are persons who are eligible for both Medicare and Medicaid.

The Medicare Current Beneficiary Survey (MCBS) was implemented in 1991 to serve as a source of information for administering the Medicare program, estimating health care expenditures for beneficiaries, and providing a better understanding of the health and well-being of the Medicare population. A leading source of information on Medicare and its impact on beneficiaries, the MCBS provides important data on beneficiaries that are not available in CMS administrative data and plays an essential role in monitoring and evaluating beneficiary health status and health care policy. The MCBS collects comprehensive data on beneficiary health insurance coverage, health care utilization and costs, access to care, and satisfaction with care, as well as special interest topics including drug coverage, knowledge about the Medicare program, and housing characteristics. Data from the MCBS are used to inform many government programs and analyses, including fiscal projections produced by the Congressional Budget Office and the Medicare Payment Advisory Commission, and are published in a wide array of peer-reviewed journals.

The MCBS is a continuous, multi-purpose longitudinal survey, representing the population of Medicare beneficiaries aged 65 and over and beneficiaries aged below 65 with certain disabling conditions living in the United States. The MCBS is sponsored by the Office of Enterprise Data and Analytics (OEDA) of the Centers for Medicare & Medicaid Services (CMS). The MCBS is authorized by section 1875 (42 USC 139511) of the Social Security Act and is conducted by NORC at the University of Chicago for the U.S. Department of Health and Human Services. The OMB Number for this survey is 0938-0568. In its rotating panel design, each beneficiary is statistically sampled as part of a panel and is interviewed up to three times per year over a four-year period. The MCBS has conducted continuous data collection since 1991.

About the MCBS Chartbook

The MCBS Chartbook provides the public with a collection of interactive charts and corresponding tables presenting estimates from both the MCBS Survey File and Cost Supplement File. The Survey File contains data collected directly from respondents and supplemented by administrative items plus facility (non-cost) information and Medicare Fee-for-Service claims. The Cost Supplement File contains both individual event and summary files and can be linked to the Survey File to conduct analyses on health care cost and utilization.

The MCBS Chartbook is organized into the following four domains and includes MCBS data from 2015 through 2021:

  • Medicare Population Overview - these charts show demographic and socioeconomic characteristics of Medicare beneficiaries, including insurance coverage and residence status.
  • Beneficiary Health & Well-being - these charts show health and disability status, reported chronic and other common health conditions, health behavior risk factors, and self-reported preventive behaviors of Medicare beneficiaries.
  • Health Care Access & Satisfaction - these charts show usual source of care reported by Medicare beneficiaries, as well as their access to care and satisfaction with care.
  • Health Care Use & Expenditures - these charts show health care service utilization and expenditures by Medicare beneficiaries.
Beginning with the release of 2021 MCBS data, this online MCBS Chartbook replaces the PDF version of the MCBS Chartbook that was updated and released annually on the CMS MCBS website to disseminate current estimates on the Medicare population (see the MCBS Resources section). MCBS estimates from 2015 through 2020 can be found in both this online version of the MCBS Chartbook and the previous MCBS Chartbook PDFs.

How to Use the Chart Tools

The Chart Tools allow users to create charts (and the corresponding tables) for measures of interest using a series of drop-down menus and/or filter lists.

  • For the Medicare Population Overview, Beneficiary Health & Well-being, and Health Care Access & Satisfaction Domains, use the drop-down menus to select the Topic Area (e.g., Preventive Care), Measure (e.g., Flu Shot), and Year (e.g., 2021).
  • For the Health Care Use & Expenditures Domain, use the drop-down menus to select the Service Type (e.g., Inpatient Hospital Services), Universe (Medicare Beneficiaries Living Only in the Community), and Year (e.g., 2021). Then use the filter list to select the metric (e.g., User Rates).
  • The drop-down menus and filter lists dynamically update based on data availability for selected measures or years.

Once the applicable drop-down menus and filter lists have been selected, the overall estimate and all available socio-demographic breakdowns (including the overall rate) for the selected measure/service type will populate in the chart. Hover over a particular bar in the chart to view additional details about that response. Use the four buttons below the chart to view corresponding documentation for each chart, including the:

  • Source statement and footnotes
  • Table showing the point estimates and standard errors
    • Select Download this Table to export a Excel file (.xlsx) of the displayed table, including the title, source statement, and footnotes.
  • Definitions of each measure
  • Measure construction details of each measure, including logic, numerator, denominator, and variable and segment information

There are also two options available for sharing the displayed chart:

  • Select Download this Chart to export a PNG (.png) of the displayed chart, including the title, source statement, and footnotes.
  • Select Share this Chart to obtain a shareable URL for the displayed chart. This URL will take users directly to the specific chart you have selected using the drop-down menus and filter lists.
Note, estimates that are suppressed do not appear in the chart, and are represented by asterisks in the table. See the Methodology below for more information about suppression criteria.

MCBS Resources

To access additional resources on the MCBS, check out the following links:

MCBS Resources Links
CMS MCBS Website
MCBS Limited Data Sets (LDS)
MCBS Public Use Files (PUFs)
MCBS Data User’s Guides, Methodology Reports, Codebooks, and LDS Variable Crosswalks
MCBS Chartbook PDFs, Financial Well-being PUF, and PUF Table Packages
MCBS Early Look, Data Briefs, Infographics, and Tutorials
MCBS Bibliography
MCBS Questionnaires and Questionnaire User Documentation
MCBS Interactives ‐ COVID-19 Data Tool, Survey File PUF Data Tool, and Financial Well-being Data Tool


This Methodology provides information about the production of the estimates and standard errors presented in the Chartbook. For details about the MCBS sample design, survey operations, and data files, please see the MCBS Methodology Report and MCBS Data User's Guide.

Data Sources

The MCBS data are made available to users via two annual Limited Data Set (LDS) files.

  • Survey File – demographic information, health insurance coverage, health status and conditions, and responses regarding access to care and satisfaction with care.
  • Cost Supplement File – comprehensive accounting of health care use and expenditures.
Each of the two LDS releases contain multiple files, called segments, which are easily linkable through a common beneficiary key ID.

The first step in preparing the data for the Chartbook analysis is to extract all of the relevant variables from the LDS files, including survey weights. The MCBS files contain data suitable for analysis, but the Chartbook estimates occasionally require performing additional data-related tasks to create the desired presentation. For example, the Chartbook combines categories for demographic variables such as education and race/ethnicity. In addition, some of the measures presented in the Chartbook draw on multiple variables, such as chronic condition measures that are coded based on whether at least one of many possible variable conditions is true. In both of these cases, new variables are generated, referred to as analytic variables, rather than overwrite the original LDS file variables. This allows flexibility to develop various chart presentations and allows for tracking all chart and table estimates back to the original source variables.


Once all the analytic variables are constructed, a new master dataset is created by including all the analytic variables for each beneficiary. Chartbook estimates are generally produced using the Survey File or Cost Supplement File ever-enrolled weights, as applicable. The Survey File ever-enrolled weights are used to produce estimates that represent the population that was ever enrolled in Medicare for any period during the calendar year. The Cost Supplement File ever-enrolled weights represent the same population as the Survey File ever-enrolled weights but are available for a smaller subset of beneficiaries who have met a minimum threshold of utilization data collection in the calendar year. This threshold is described in the MCBS Methodology Reports (see the MCBS Resources section). Estimates generated using data from Topical segments, which were fielded in the winter and summer rounds following the data year, use the special non-response adjustment weights that are specific to each Topical segment. For more information on Topical segments and weights, please consult the MCBS Survey File Data User's Guides (see the MCBS Resources section).

The MCBS interviews a sample of Medicare beneficiaries. Therefore, estimated standard errors are reported for all estimates in the Chartbook. The standard error is a measure of the statistical precision of an estimate. The standard errors reported in the tables in this Chartbook reflect the complex sample design of the MCBS. The MCBS is based on a stratified, three-stage, probability-proportional-to-size (PPS) sample design (with post-stratified weights). To take into account the complex sample design features of the MCBS, important adjustments to the variance estimates are required (not doing so in the case of the MCBS would tend to underestimate the standard errors and the resulting confidence intervals would tend to be too narrow). Most commercial software packages today (including R, STATA, SUDAAN, and complex survey procedures in SAS®) include techniques to accommodate the complex design of the MCBS.

Variance Estimation

This Chartbook uses the balanced repeated replication (BRR) method to generate standard errors for estimates, implementing SAS survey procedures. To explain how this process works in practice, an example is presented here using the normal approximation method of computing a confidence interval. The Chartbook indicates that 17.0 percent of all Medicare beneficiaries are veterans in 2020. The standard error of this estimate (0.5 percent) can be used to assess its statistical precision by constructing a 95 percent confidence interval, which can be calculated by using the formula

P ± 1.96se(P),

Where P is the estimated (weighted) sample percentage and se(P) is an estimate of the standard error of P. Based on this formula, the approximate 95 percent confidence interval for the estimated proportion of Medicare beneficiaries who are veterans in 2020 is 17.0 percent plus or minus 1.96*0.5=1.0 percent, for a 95 percent confidence interval of (16.0, 18.0). Theoretically, if the same survey was repeated on 100 samples of the same size, the true population proportion to fall within this confidence interval 95 times out of 100. For additional information on variance estimation for the MCBS, refer to the MCBS Advanced Tutorial on Weighting and Variance Estimation (see the MCBS Resources section).

Missingness and Suppression

Due to low levels of item non-response and to maintain trends, all estimates in the Chartbook exclude beneficiaries for whom LDS data are missing for a given measure. Imputations were not performed on the LDS file variables used in the production of the Chartbook, as the LDS variables have already undergone thorough editing, quality control checks, and imputation prior to release. For more detailed information regarding data editing and imputation procedures conducted, please consult the MCBS Methodology Report (see the MCBS Resources section).

There are occasions in which certain categories of variables are excluded from a chart by design. When estimates are presented in charts or tables for these measures, beneficiaries in the excluded categories are not shown in the chart but are still included in the denominator for the estimate, meaning that totals across the categories in the chart may not add up to 100 percent.

Suppression is used in order to protect the confidentiality of Medicare beneficiaries by avoiding the release of information that can be used to identify individual beneficiaries. Estimates with a denominator of less than 50 sample persons or with a numerator of zero sample persons are suppressed in the Chartbook. Some estimates are suppressed because they do not meet minimum criteria for reliability, which are explained below.

Statistical Reliability

The Chartbook only displays statistics that meet reliability criteria. This reliability is assessed using two different sets of criteria, depending on the type of estimate. For proportions, the Clopper-Pearson method was used to compute confidence intervals for each estimate. Estimates with a confidence interval whose absolute width is at least 0.30, with a confidence interval whose absolute width is no greater than 0.05, or with a relative confidence interval width of more than 130 percent of the estimate are suppressed in the Chartbook.

For more information: Parker, J. D., M. Talih, D. J. Malec, et al. National Center for Health Statistics Data Presentation Standards for Proportions. National Center for Health Statistics. Vital Health Stat 2, no. 175 (2017). Available from:

For other estimates, relative standard errors (RSEs) are calculated as the standard error of the estimate divided by the estimate itself (percentage), and the result is then converted to a percentage value by multiplying the decimal value by 100. Estimates with a relative standard error of greater than 30 percent are suppressed in this Chartbook because they do not meet the standards of reliability or precision.

Key Definitions

The definition of each measure presented in a given chart in the Chartbook can be found within the Chart Tools by selecting View Key Definitions below the chart.

Below are definitions of additional terms used throughout the Chartbook.

  • Beneficiary: Beneficiary refers to a person receiving Medicare services who may or may not be participating in the MCBS. Beneficiary may also refer to an individual selected from the MCBS sample about whom the MCBS collects information. Beneficiaries must meet at least one of three criteria for Medicare eligibility (is aged 65 years or over, is under age 65 with certain disabilities, or is of any age with End-Stage Renal Disease) and be entitled to health insurance benefits. (Source:

  • Brief Interview for Mental Status (BIMS): The Brief Interview for Mental Status (BIMS) is a short interview used in the Long-Term Care Minimum Data Set (MDS) for assessing the mental status of long-term care facility residents. The interview contains a series of questions related to immediate recall of three words, temporal orientation (to correct month, year, and day), and delayed recall of three words, with a maximum score of 15 across the three components. Lower BIMS scores indicate greater cognitive impairment.

  • Claim-only event: A claim-only event is a medical service or event known only through the presence of a Medicare Fee-for-Service claim from administrative data. This means that the event represented in the data could not be reconciled with a corresponding survey-reported event.

  • Community interview: Survey administered for beneficiaries living in the community (i.e., not in a long-term care facility such as a nursing home) during the reference period covered by the MCBS interview. An interview may be conducted with the beneficiary or a proxy.

  • End-stage renal disease (ESRD): ESRD is the state of kidney impairment that is irreversible, cannot be controlled by conservative management alone, and requires dialysis or kidney transplantation to maintain life.

  • Facility interview: Survey administered for beneficiaries living in facilities, such as long-term care nursing homes or other institutions, during the reference period covered by the MCBS interview. Interviewers conduct the Facility interview with staff members located at the facility (i.e., Facility respondents); beneficiaries are not interviewed if they reside at a facility.

  • Fee-for-Service (FFS) payment: FFS is a method of paying for medical services in which each service delivered by a provider bears a charge. This charge is paid by the patient receiving the service or by an insurer on behalf of the patient.

  • Long-Term Care Minimum Data Set (MDS): The Long-Term Care MDS is part of the federally mandated process for clinical assessment of all residents in Medicare and Medicaid certified nursing homes. For more information, please visit .

  • Proxy: Beneficiaries who were too ill, or who could not complete the Community interview for other reasons, were asked to designate a proxy, someone very knowledgeable about the beneficiary's health and living habits. In most cases, the proxy was a close relative such as the spouse/partner or a son or daughter. In a few cases, the proxy was a non-relative like a close friend or caregiver. In addition, a proxy was utilized if a beneficiary had been reported as deceased during the current round's reference period or if a beneficiary who was living in the community in the previous round had since entered into a long-term care facility. Proxy interviews are only used for the Community interview, as the Facility interview is conducted with a staff member located at the facility (see definition of Facility interview).

  • Respondent: Respondent refers to a person who answers questions for the MCBS; for Community interviews, this person can be the beneficiary or a proxy. If the respondent is a proxy, they answer questions about the beneficiary rather than themselves.

  • Round: The MCBS data collection period. There are three distinct rounds each year: winter (January through April), summer (May through August), and fall (September through December).

  • Survey-reported event: A survey-reported event is a medical service or event reported by a respondent during an interview. The event may have been matched to a Medicare Fee-for-Service claim from administrative data, or it may be a survey-only event, in which case it was not matched to a Medicare claim and is only known through the survey.

  • Total expenditures: Total expenditures in this Chartbook equal the sum of expenditures by Medicare, Medicaid, private insurance, out-of-pocket, and other sources, as defined below. These expenditures include only health care goods and services purchased directly by individuals during the data collection year. They exclude public program administration costs, the net cost of private health insurance, research by nonprofit groups and government entities, and the value of new construction put in place for hospitals and nursing homes. Data regarding expenditures is collected in Community and Facility interviews as well as obtained from administrative data sources.

    • Medicare expenditures equal Medicare program payments for Fee-for-Service beneficiaries and payments by Medicare Advantage plans to health care providers. For beneficiaries enrolled in Fee-for-Service, Medicare expenditures are based on claims and survey-reported data. For beneficiaries enrolled in Medicare Advantage, Medicare expenditures are based on survey-reported data.

    • Medicaid expenditures consist of payments for services made by state Medicaid programs. Medicaid covers coinsurance amounts, copayments, deductibles, and charges for some non-Medicare covered services not paid for by other public or private insurance plans.

    • Private insurance expenditures consist of payments made by individually-purchased private insurance plans and employer-sponsored private insurance plans, plus payments reported by or imputed for beneficiaries enrolled in private health maintenance organizations. The definition applies to beneficiaries who only completed Community interviews during the year and beneficiaries who completed at least one Community interview and for whom at least one Facility interview was conducted during the year. For beneficiaries for whom a Facility interview was completed, private insurance expenditures consist of payments made by private health insurance plans, whose sources (i.e., individual purchase or employer-sponsored) are unknown. No distinction was made during the collection of the Facility data as to the source of private health insurance plans.

    • Out-of-pocket expenditures consist of direct payments to providers made by the beneficiary, or by another person on behalf of the beneficiary. These payments are for coinsurance amounts, copayments, deductibles, balance billings, and charges for non-Medicare covered services not paid for by public or private insurance plans. These expenditures do not include premiums or Part D deductibles. A coinsurance is the percentage of covered hospital or medical expense, after subtraction of any deductible, for which an insured person is responsible. For example, after the annual deductible has been met, Medicare will generally pay 80 percent of approved charges for services and supplies covered under Medicare Part B. The remaining 20 percent of the approved charge is the coinsurance amount, for which the beneficiary is liable. A copayment is a form of cost-sharing whereby the insured pays a specific amount at the point of service or use (e.g., $10 per doctor visit). A deductible is an initial expense of a specific amount of approved charges for covered services within a given time period (e.g., $100 per year) payable by an insured person before the insurer assumes liability for any additional costs of covered services. For example, in 2017, Medicare Part A helped to cover hospital, skilled nursing facility, home health care, and other services after the beneficiaries paid initial expenses ‐ known as the deductible ‐ of up to $1,260.

    • Other source expenditures consist of payments made by other public health plans and private liability insurance plans. For beneficiaries who completed a Community interview, examples of other public sources of payment include state pharmaceutical assistance programs and payments for beneficiaries who received medical services from the Department of Veterans Affairs. For beneficiaries for whom a Facility interview was conducted, examples of other public sources of payment include payments from state, county, or community departments of mental health, state supplemental assistance and welfare programs, and Black Lung funds.

Measure Construction

The construction of each measure presented in a given chart in the Chartbook can be found within the Chart Tools by selecting View Measure Construction below the chart. Each entry provides the measure construction logic, describes the numerator and denominator for the measure, and lists the LDS release (Survey File or Cost Supplement File) and LDS segment(s) used to construct the measure. Note, because the MCBS uses different questionnaires to collect data for beneficiaries living in the community and beneficiaries living in facilities, two measure construction entries are included for certain measures to clarify the construction of the measure for each population.

Below is additional information about measure construction used in the Chartbook.

The measure construction entries contain universal pseudo-code (not specific to any programming language or statistical package) for the construction of each measure's numerator. Data users will also need to construct the appropriate denominator using the information provided in the measure construction entries and LDS codebooks.

All estimates in the Chartbook exclude beneficiaries for whom LDS data are missing for a given measure. Missing variable data, which encompass Don't Know, Not Ascertained, and Refused responses, are excluded from both the numerator and denominator in the calculation of each measure. For some measures, additional variable values are excluded. In these cases, data users are instructed in the measure construction logic field to designate specific values as missing.

For measures that are constructed using multiple variables, if there are data for at least one of the variables, the record should be included in construction of the measure. If data are missing for all variables used in construction, the record should be excluded from construction of the measure.

Suggested Citation

To cite the MCBS Chartbook:

Centers for Medicare & Medicaid Services. Medicare Current Beneficiary Survey Chartbook. Last modified 2024. Retrieved from

Terms of Use

The MCBS Chartbook was created by NORC at the University of Chicago (NORC) under a contract with the Centers for Medicare & Medicaid Services (CMS) Office of Enterprise Data and Analytics (OEDA). This communication was printed, published, or produced and disseminated at U.S. taxpayer expense. For information about using the MCBS Chartbook website, see the Terms and Conditions and Privacy Policy .

Contact Us

If you have questions about using the MCBS Chartbook website, please email NORC at

If you have technical questions about the MCBS, including questions about the estimates presented in the charts and tables or how to access and analyze MCBS data, please email

You can also find more information about the MCBS on the CMS website: