Cost-Related Problems Are Less Common Among Beneficiaries in Traditional Medicare Than in Medicare Advantage, Mainly Due to Supplemental Coverage
This analysis is based on data from the Centers for Medicare & Medicaid Services’ Medicare Current Beneficiary Survey (MCBS), 2018 Survey File. The analysis includes beneficiaries enrolled in Part A and Part B for most months of the year and excludes those with Part A or Part B only and Medicare as a Secondary Payer for most months of the year. The analysis also excludes institutionalized beneficiaries since the analysis was based on questions asked of community residents only.
In this analysis, we define “cost-related problems” based on positive responses to any of the following four questions:
- Since (12 months prior), have you had any trouble getting health care that you wanted or needed because the cost was too high?
- Since (12 months prior), have you had any trouble getting health care that you wanted or needed because you did not have enough money?
- Since (12 months prior), have you delayed seeking medical care because you were worried about the cost?
- Since (12 months prior) have you had problems paying or were unable to pay any medical bills?
We used a bivariate analysis to assess differences in the share of beneficiaries reporting cost-related problems across groups. All numbers reported in the text, figures, and table are a result of this analysis and we indicate where differences were statistically significant at the 5% level.
To account for differences in demographic characteristics and health status of Medicare beneficiaries enrolled in Medicare Advantage, traditional Medicare with supplemental coverage, and traditional Medicare without supplemental coverage, we modeled the likelihood of reporting a cost-related problem using a multivariate logistic regression. Specifically, we estimate the following:
E[Yi] = g(X’β)
where g(·) is a logit link function; Y is equal to 1 if beneficiary i reported a cost-related problem, Xi is a vector of individual-level covariates, including coverage type (Medicare Advantage, traditional Medicare with supplemental coverage, and traditional Medicare without supplemental coverage), income, age category, race/ethnicity, self-reported health status, number of chronic conditions, whether a beneficiary had a cognitive impairment, number of limitations in activities of daily living, and Medicare-Medicaid dual status.
To further examine the relationship between cost-related problems and coverage types, we estimated the same model on subsets of the full sample, including those with Medicare Advantage, those in traditional Medicare with supplemental coverage, those with traditional Medicare without supplemental coverage, White beneficiaries, Black beneficiaries, Hispanic beneficiaries, beneficiaries with per capita income below $20,000, income $20,000-$39,999, and income $40,000 and above, beneficiaries in fair/poor health, and beneficiaries under the age of 65 with a long-term disability. Tables 2 to 6 provide the odds-ratios and confidence intervals from each of the models.
Due to sample size limitations (as estimated by power and sample size calculations), we do not report data on self-reported health status and age group among Medicare beneficiaries without supplemental coverage, or among Hispanic beneficiaries and other racial and ethnic groups by sources of coverage. Additionally, while the collection of race and ethnicity data in survey data has improved over time, sample size and other limitations hinder our ability to display results comparing the share of Medicare beneficiaries reporting cost-related problems for certain racial and ethnic groups or subgroups within certain racial and ethnic groups in our analysis, especially Asian adults, American Indian and Alaska Native adults, Native Hawaiian and Other Pacific Islander adults, and adults who identify as two more races. Throughout this brief, individuals of Hispanic origin may be of any race, but are classified as Hispanic for the analysis; all other groups are non-Hispanic.