About the Financial Well-being PUF Data
Background
Starting with the 2020 data year, the Centers for Medicare & Medicaid Services (CMS) began releasing an annual Medicare Current Beneficiary Survey (MCBS) Public Use File (PUF) on the Financial Well-being of Medicare Beneficiaries and accompanying technical documentation to allow researchers to access estimates on asset ownership, access to transportation, food insecurity, and access to health care for Medicare beneficiaries living in the community. Estimates on access to and use of the internet are available for the 2020 data year only.
The MCBS Financial Well-being PUF contains estimates created using the MCBS Survey File Limited Data Set (LDS). MCBS Limited Data Sets (LDS) are available to researchers with a data use agreement. MCBS Public Use Files (PUFs) are available to the public as free downloads. The MCBS Financial Well-being PUF uses estimate suppression to protect the confidentiality of Medicare beneficiaries by avoiding the release of information that can be used to identify individual beneficiaries.
More Information on MCBS Financial Well-being PUF Data
To learn more about the Financial Well-being PUF data, see the following resources:
Questionnaires
Download the MCBS Questionnaires.
Public Use File
Download the 2020-2021 MCBS Financial Well-being PUFs, including the technical appendix.
Limited Data Set
Download documentation for the MCBS LDS, including the Data User’s Guides and Methodology Reports, and learn more about the LDS DUA process.
About This Tool
The MCBS Financial Well-being Data Tool presents estimates from the 2020-2021 MCBS Financial Well-being PUFs with appropriate weights. Estimates represent the population of beneficiaries living in the community who were ever enrolled in Medicare at any time during the calendar year.
The Financial Well-being Data Tool is comprised of a series of dashboards related to Medicare beneficiaries’ financial assets, transportation access, and food insecurity. A dashboard for access to and use of the internet is also available for the 2020 data year. Estimates for various demographic, geographic, health factor, insurance coverage, and access to care subgroups are also presented.
The tool presents a visual approximation of the analysis; all conclusions should be verified through appropriate analysis of the datasets.
Methodology
Criteria for Inclusion in the Tool
The Financial Well-being Data Tool includes Medicare beneficiaries living in the community who completed only Community interviews during the year. This excludes beneficiaries for whom any Facility interview was completed. Beneficiaries who received a Community interview answered questions themselves or by proxy.
All estimates in the Financial Well-being Data Tool also exclude beneficiaries for whom Survey File 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.
Estimate suppression is used 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. Additionally, some estimates were suppressed as they do not meet minimum criteria for reliability. The reliability criteria applied for the estimates presented in the data tool states 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.
Parker, Jennifer D., Makram Talih, Donald 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).
To link to this article: https://www.cdc.gov/nchs/data/series/sr_02/sr02_175.pdf
Data Tool Measure Construction
Several measures included in the dashboards are created using derived variables. These derived variables combine information from one or more variables available in the Survey File LDS:
Area Deprivation Index: ADI is an indicator of the socioeconomic disadvantage of geographic areas. National rankings are based on the Census block group for the beneficiary’s primary residence address. ADI values in the 1-25th percentile are the least disadvantaged, while those in 75-100th percentile are the most disadvantaged.
Assets at Financial Institutions: Assets at financial institutions comprise a group of three assets held by if the beneficiary and/or their spouse/partner (if applicable): a checking account, a savings account, or certificates of deposit.
Chronic Condition Measures
Arthritis: Respondents were asked if a doctor or other health professional had ever told them that they had rheumatoid arthritis, osteoarthritis, or any other form of arthritis. Respondents who reported more than one condition were only counted once for the purposes of calculating the proportion of beneficiaries with arthritis. The proportion of beneficiaries with arthritis is not captured in the Number of Chronic Conditions measure for 2020 because variables for osteoarthritis and other types of arthritis are not available in the 2020 Survey File. Variables on all three arthritis conditions are once again available beginning with the 2021 Survey File.
Heart Disease: Respondents were asked if a doctor or other health professional had ever told them that they had myocardial infarction (heart attack), angina pectoris or coronary heart disease (CHD), congestive heart failure, or any other heart condition. Respondents who reported more than one condition are only counted once for the purposes of calculating the proportion of beneficiaries with heart disease.
Mental Condition: Respondents were asked whether a doctor or other health professional had ever told them that they had depression or a mental or psychiatric disorder other than depression. The mental condition measure counts the presence of at least one of these conditions. Beneficiaries who have more than one condition are only counted once for the purposes of calculating the proportion of beneficiaries with a mental condition.
Number of Chronic Conditions: Chronic conditions comprises a group of 14 health conditions measures: heart disease, cancer (other than skin cancer), Alzheimer's disease, dementia other than Alzheimer’s disease, depression, mental condition, hypertension, diabetes, arthritis, osteoporosis/broken hip, pulmonary disease, stroke, high cholesterol, and Parkinson’s disease. It is possible for a beneficiary to have "ever" been diagnosed with both Alzheimer’s disease and dementia (other than Alzheimer’s disease) as previous survey responses are carried forward into subsequent data years. For the purposes of the Number of Chronic Conditions measure, Alzheimer’s disease and dementia (other than Alzheimer’s disease) are counted as one chronic condition for beneficiaries diagnosed with both conditions. As the definition of mental condition encompasses depression, for the purposes of the Number of Chronic Conditions measure, depression and mental condition are counted as one chronic condition for beneficiaries diagnosed with both conditions. Additionally, as noted above, because variables for osteoarthritis and other types of arthritis are not available in the 2020 Survey File, the Number of Chronic Conditions measure does not capture arthritis for 2020 (i.e., comprises a group of 13 health conditions measures instead).
Osteoporosis/Broken Hip: Respondents were asked whether a doctor or other health professional has ever told them that they had osteoporosis or a broken hip. The osteoporosis/broken hip measure counts the presence of at least one of these conditions. Beneficiaries who have more than one condition are only counted once for the purposes of calculating the proportion of beneficiaries with osteoporosis/broken hip.
Couldn't Afford Balanced Meals: Respondents were asked if the following statement was often true, sometimes true, or never true in the last 12 months: The beneficiary or other adults in the beneficiary’s household couldn’t afford to eat balanced meals. Beneficiaries were coded as "Yes" if this was "often true" or "sometimes true."
Disability Status: Respondents were asked whether they have serious difficulty hearing, seeing, concentrating, remembering, or making decisions, walking or climbing stairs, dressing or bathing, or with errands. Beneficiaries who had no serious difficulties with these activities were included in the category “No disability.” Beneficiaries who had a serious difficulty in one area were categorized as “One disability” and those who had a serious difficulty in more than one area were categorized as "Two or more disabilities."
Dual Eligibility Status: Annual Medicare-Medicaid dual eligibility was based on the state Medicare Modernization Act (MMA) files. Beneficiaries were considered “dually eligible" and assigned a dual eligibility status if they were enrolled in Medicaid for at least one month. This information was obtained from administrative data sources.
Financial Assets: The assets at financial institutions, checking account, savings account, certificates of deposit, and stocks or mutual funds measures capture if the beneficiary, the beneficiary's spouse/partner, and/or the beneficiary and their spouse/partner jointly owned the respective asset. Similarly, the retirement accounts, receive Social Security, receive Supplemental Security Income (SSI), and receive pension measures capture if the beneficiary and/or the beneficiary’s spouse/partner received the respective asset.
Food Didn't Last and No Money to Buy More: Respondents were asked if the following statement was often true, sometimes true, or never true in the last 12 months: The food that the beneficiary or other adults in the beneficiary's household bought just didn’t last, and they didn't have money to get more. Beneficiaries were coded as "Yes” if this was "often true" or "sometimes true."
Food Insecure: Beneficiaries were categorized as food insecure if respondents reported any of the following five food insecurity measures: Food didn’t last and no money to buy more, cut size of meals or skip meals, eat less because not enough money for food, didn’t eat because not enough money for food, and couldn’t afford balanced meals.
Limited Driving to Daytime: Respondents were asked if the beneficiary gave up driving all together or limited driving to daytime. The limited driving to daytime measure is applicable to respondents who were asked if they have limited driving to daytime but have not given up driving all together.
Medicare Coverage: Beneficiaries were coded as having Medicare Advantage (MA) coverage if they had MA coverage for one or more months out of the calendar year. Otherwise, beneficiaries were coded as having traditional Fee-for-Service (FFS) coverage.
Estimates and Weights
Percentage estimates are calculated using the ever-enrolled survey weights supplied in the Survey File LDS. Additionally, estimates generated using data from Topical segments, which were fielded in the winter and summer rounds following the data year, used the special non-response adjustment weights that are specific to each Topical segment. Variance estimates (which are needed to derive standard errors and confidence intervals) are calculated using replicate weights supplied in the file. See documentation for MCBS weights.
Confidence Interval Calculation
Within the dashboard, the confidence intervals within each dot plot are adjusted using the Goldstein-Healy method as described in:
Wright, Tommy, Martin Klein, and Jerzy Wieczorek. “A Primer on Visualizations for Comparing Populations, Including the Issue of Overlapping Confidence Intervals.” The American Statistician 73, no. 2 (2019): 165-178. DOI: 10.1080/00031305.2017.1392359
To link to this article: https://doi.org/10.1080/00031305.2017.1392359
This adjustment is done so that the confidence intervals can be readily used to determine if two estimates within a chart are, statistically speaking, different (see the use of confidence intervals to compare groups).
Software Used
The MCBS Financial Well-being PUF Data Tool was created using R Shiny and D3.js.
How to Use This Tool
Each dashboard consists of a bar chart presenting outcome variables related to a theme. For example, the dashboard on Financial Assets shows the percent of beneficiaries who have different types of assets. The dashboard also shows a series of dot plots which dynamically update to show the breakdown of responses for a particular category.
If a category is missing from a dot plot (e.g., "85+ years" for "Didn't Eat Because Not Enough Money For Food" on the Food Insecurity dashboard), the estimate is suppressed. See Methodology for more information on the suppression criteria.
Interactivity
Select the text "Learn more" to open a drop-down box with instructions and a link to the methodology.
Click on a single bar within the bar chart to see how that topic differs within each group displayed in the dot plots. The dot plots include confidence intervals which can be used to identify potential meaningful differences.
Clicking on a bar in the bar chart adjusts the universe of beneficiaries in the dot plots. For example, clicking on the bar corresponding to "Home Ownership" in the Financial Assets dashboard updates the universe of beneficiaries in the dot plots to all beneficiaries who had a valid response to the question about owning their own home.
Click on the time period (e.g., 2021) to change the data source/survey year.
Scales on the dot plots are dynamic. Estimates should only be compared within groups in the dot plots, not across. For example, the “Male” group should not be compared to the “75-84 years” group.
Hover-Overs
Hovering over a particular bar or dot in the tool allows users to view additional details about the survey responses. For example, hovering over the bar corresponding to “Home Ownership” in the Financial Assets dashboard will provide the percent estimate for that category as well as the standard error.
Use Confidence Intervals to Compare Groups
Users can perform statistical hypothesis tests (p = 0.05) to determine if there is a meaningful difference between two percentages within a chart. The width of the intervals indicates the measure of uncertainty in the estimates.
Given the method used to construct the confidence intervals (see note in the Methodology section), if two confidence intervals within a chart overlap, then there is no meaningful difference between the percentages for those two groups. However, if two confidence intervals within a chart do not overlap, then there is a meaningful difference between the percentages for those two groups.
Confidence intervals are adjusted for direct comparison across groups; the confidence interval associated with each demographic group does NOT represent a univariate confidence interval for that group’s proportion. Confidence intervals should not be interpreted for a single demographic group, only used for hypothesis tests of differences between groups. However, the width of the confidence interval does provide a visual sense of the uncertainty of the estimate. All conclusions should be verified through appropriate analysis of the data.
For more information, contact us.