Qualified Medicare Beneficiary Program: Enrollment Trends And Characteristics Of Low-Income Beneficiaries

Abstract

To address low take-up of Medicare Savings Programs for low-income Medicare beneficiaries, the Centers for Medicare and Medicaid Services issued regulations in 2023 aimed at streamlining the enrollment process. We analyzed 2016–22 data from the Medicare Current Beneficiary Survey to examine recent take-up of the largest and most generous Medicare Savings Program, the Qualified Medicare Beneficiary (QMB) program. We compared beneficiary characteristics and cost-related barriers to care among QMB enrollees and beneficiaries who were eligible but not enrolled. QMB take-up rose from 62 percent in 2016 to 66 percent in 2022. QMB-eligible beneficiaries who were Asian or Hispanic were more likely than eligible White beneficiaries to enroll in the program. Eligible beneficiaries younger than age sixty-five were more likely to enroll than those ages sixty-five and older, and beneficiaries in Medicaid expansion states were more likely to enroll than those in nonexpansion states. QMB enrollees were less likely than Medicare beneficiaries who were eligible but not enrolled to report delaying care because of cost or having problems paying for care. These findings suggest that additional policies and programs may be needed to increase QMB take-up and to improve health care access and affordability for millions of low-income Medicare beneficiaries.

To help reduce the cost burden of Medicare premiums and cost sharing for low-income Medicare beneficiaries, Congress created four Medicare Savings Programs between 1989 and 1998.^1-3 Of the nearly ten million beneficiaries enrolled in Medicare Savings Programs in 2021 (the most recent year for which data are available), most (8.0 million) were enrolled in the Qualified Medicare Beneficiary (QMB) program, wherein Medicaid covers Medicare Parts A and B premiums and cost sharing, including deductibles, coinsurance, and co-payments.⁴ To be eligible for the QMB program, beneficiaries must have incomes at or below 100 percent of the federal poverty level and assets of $9,660 or less for an individual or $14,470 or less for a couple in 2025.⁵ Among QMB enrollees in 2021, nearly four-fifths were enrolled in QMB Plus, meaning that they were also dually eligible for full Medicaid benefits via other Medicaid pathways, such as receipt of Supplemental Security Income (SSI) benefits.^4,6 The remaining QMB enrollees were in QMB Only, meaning they only had Medicaid coverage of their Medicare premiums and cost sharing.⁴ Among the other Medicare Savings Programs, the Specified Low-Income Medicare Beneficiary program and the Qualifying Individual program pay for Medicare Part B premiums only and do not cover cost sharing. The Qualified Disabled and Working Individual program covers Medicare Part A premiums only, with no cost-sharing coverage.

Background

Despite the potential of Medicare Savings Programs to improve access to needed care by reducing the cost burden of Medicare premiums and cost sharing, many eligible beneficiaries have not enrolled.^2,7-12 As the only Medicare Savings Program that covers both Medicare premiums and cost sharing, the QMB program is of particular interest for its potential to substantially improve the affordability of care for low-income beneficiaries. There is a robust literature documenting low Medicare Savings Program take-up, measured as the share of the eligible population that is enrolled. For example, a study based on 2009–10 data estimated that only about half of those eligible for the QMB program were enrolled, and take-up of other Medicare Savings Programs was lower.⁸ More recent estimates have found overall Medicare Savings Program take-up to be 68 percent and 63 percent, using 2013 and 2014 data, respectively.^11,12 Reasons for poor take-up include lack of awareness of the program, difficulty determining eligibility, and lack of resources to support beneficiaries through the application process.^7,11-14

It is possible that increases in Medicaid enrollment that resulted from the Affordable Care Act (ACA) Medicaid expansion could boost enrollment into the QMB program and other Medicare Savings Programs once these enrollees become Medicare eligible. The Medicaid and CHIP Payment and Access Commission examined changes in Medicare Savings Program enrollment before and after the ACA Medicaid expansion and found that enrollment increased between 2010 and 2021 by an average annual growth rate of 2.6 percent for QMB Plus and 4.7 percent for QMB Only; however, the extent to which the increased enrollment was attributable to increases in take-up or the number of beneficiaries eligible for QMB is yet unclear.⁴

Persistently low QMB take-up raises concerns about the affordability of care for low-income Medicare beneficiaries; increasing out-of-pocket expenses have been associated with lower use of outpatient visits and recommended cancer screenings.^15-19 To address these concerns, as well as low Medicare Savings Program take-up rates, the Centers for Medicare and Medicaid Services (CMS) finalized a rule in 2023 to streamline the Medicare Savings Program enrollment process. The rule includes requirements for states that automatically enroll SSI recipients in Medicaid to automatically enroll them in the QMB program by October 2024. It also requires states to align the definition of family size used for income calculations in eligibility determinations for the Medicare Savings Programs with the definition used in the Medicare Part D Low-Income Subsidy program,²⁰ and it codifies the statutory requirements for states to maximize their use of income and assets data from beneficiaries’ Part D Low-Income Subsidy applications (“leads” data) to initiate Medicare Savings Program applications by April 2026.^21-23 Better aligning eligibility rules for the Medicare Savings Program and Low-Income Subsidy program could reduce administrative burdens for beneficiaries related to the Medicare Savings Program enrollment process.^21,24

We analyzed 2016–22 data from the Medicare Current Beneficiary Survey (MCBS) to assess recent take-up of the QMB program, the largest and most generous Medicare Savings Program, and to compare beneficiary characteristics and cost-related barriers to care among eligible populations enrolled and not enrolled in the program.

Study Data And Methods

DATA SOURCES AND STUDY SAMPLE

We used pooled, cross-sectional data from the 2016–22 MCBS Survey File Limited Data Set files, which also include Medicare administrative data.²⁵ Our analysis included community-dwelling beneficiaries in all fifty states and Washington, D.C. (hereafter referred to as “states”), who were enrolled in Medicare for twelve months and who were likely eligible for the QMB program in each year. We identified eligibility based on annual, self-reported income and asset data in the MCBS. We calculated household income as a percentage of the federal poverty level on the basis of annual income and household size. Our sample included beneficiaries with incomes up to 100 percent of poverty and assets that did not exceed the federal Medicare Savings Program asset limit in each year. For beneficiaries in the fifteen states that had higher Medicare Savings Program income or asset limits, or both, during the period 2016–22, we applied the state- and year-specific limits (see online appendix exhibit A1).²⁶ Hereafter, beneficiaries who met these criteria are referred to as QMB eligible.

KEY MEASURES

In primary analyses, we defined QMB take-up as being QMB eligible and having one or more months of QMB enrollment in each year. Among QMB-eligible beneficiaries who did not enroll in QMB, we calculated the percentages of those with at least one month of enrollment in Medicare Savings Programs covering Medicare Part A or Part B premiums only, those who were full-benefit dual-eligible beneficiaries without the Medicare Savings Program, those who were non-dual-eligible beneficiaries receiving the Part D Low-Income Subsidy only, and those who were not enrolled in any of these financial assistance programs.

We obtained data on beneficiary characteristics from Medicare administrative data and MCBS survey data. Self-reported survey measures included marital status, education, English-speaking proficiency, annual income, health status, number of chronic conditions reported among twenty-two conditions listed in the MCBS, number of limitations in activities of daily living (walking, feeding, dressing, using the toilet, bathing, and getting into and out of bed or a chair), and cognitive impairment status (determined by whether beneficiaries reported in the MCBS that they had “difficulty concentrating/remembering/deciding,” had “trouble concentrating for more than half the days or nearly every day,” or reported having Alzheimer’s disease or dementia).

We used Medicare administrative data to identify beneficiaries’ sex; age; race and ethnicity; residential urbanicity; and other sources of insurance coverage, including Medicare Advantage, Medicare Part D, Part D Low-Income Subsidy, and other full dual enrollment in Medicaid without Medicare Savings Program enrollment. Data on Medicare supplemental coverage were self-reported, including Medigap or other supplemental coverage (such as employer-sponsored insurance or coverage from the Department of Veterans Affairs). We also characterized beneficiaries’ state policy characteristics by year, including whether they resided in a state that increased Medicare Savings Program income or asset limits above the federal limits (see appendix exhibit A1),²⁶ a state that auto-enrolled SSI recipients into Medicaid,^27,28 a state that expanded Medicaid to cover adults with income up to 138 percent of the federal poverty level,²⁹ or a state in which the Medicaid program included a medically needy pathway for elderly or disabled beneficiaries.^27,30-35

Last, we examined two measures of cost-related barriers to care reported in the MCBS: whether beneficiaries reported delaying care because they were worried about the cost and whether beneficiaries reported problems paying any medical bills during the year. Because the question about problems paying medical bills was not asked in 2016, our analysis of this outcome used 2017–22 data only.

ANALYSIS

We examined the percentage of QMB-eligible beneficiaries who enrolled in QMB for at least one month in each year, as well as the characteristics of the QMB-eligible population by take-up status. We accounted for the MCBS sampling design by using the weights provided in the Limited Data Set files. We compared beneficiary characteristics among QMB-eligible beneficiaries by QMB take-up status. For these beneficiary characteristics, we calculated the average among beneficiaries pooled during the period 2016–22 and tested for differences using chi-square tests. We then fit a multivariable linear probability model to examine the adjusted associations between beneficiary characteristics and QMB take-up. Last, we examined the percentage of QMB-eligible beneficiaries who reported each of the cost-related barriers to care by take-up status and the association between QMB take-up and the likelihood of reporting cost-related barriers to care, using separate multivariable linear probability models and adjusting for the beneficiary characteristics listed above.

SENSITIVITY ANALYSES

In sensitivity analyses, we examined the percentage of enrollees with twelve continuous months of QMB enrollment in each year. We examined QMB enrollment holding all states to their 2016 income and asset limits to examine whether time trends in take-up differed in the absence of state eligibility changes. Last, to address potential underestimates of self-reported income, we set QMB income and asset tests at 120 percent of their actual level to determine QMB eligibility.

LIMITATIONS

We acknowledge several limitations. First, our study design was descriptive and did not assess causal relationships between beneficiary or state characteristics and QMB take-up or between QMB take-up and cost-related barriers to care. Second, our study included only community-dwelling beneficiaries; therefore, the results are not generalizable to beneficiaries living in facilities. Third, to estimate QMB take-up, our analysis used self-reported annual income and asset data, which could be subject to reporting error and might not reflect within-year fluctuations in QMB eligibility, which is based on monthly income. Past work has found that survey respondents underestimate income relative to administrative sources;³⁶ therefore, we may have been more likely to undercount QMB-eligible beneficiaries at the upper end of the income scale. Nevertheless, self-reported income and asset data have been used to estimate QMB take-up previously because of the lack of alternative administrative data sources with comprehensive capture of this information linked to Medicare Savings Program enrollment.^8,9,11,12,37

Study Results

QMB TAKE-UP AMONG ELIGIBLE BENEFICIARIES OVER TIME

QMB take-up increased from 62 percent in 2016 to 66 percent in 2022 among QMB-eligible beneficiaries (exhibit 1). In 2022, among QMB-eligible beneficiaries, 8 percent were not enrolled in QMB but were enrolled in other Medicare Savings Programs that covered Medicare Part A or Part B premiums, 6 percent were enrolled in full-benefit Medicaid and not enrolled in any Medicare Savings Program, and 3 percent were enrolled in Medicare Part D Low-Income Subsidy only. Seventeen percent of QMB-eligible Medicare beneficiaries were not enrolled in any of these programs in 2022, a decline from 21 percent of QMB-eligible beneficiaries in 2016.

EXHIBIT 1. Enrollment in selected programs for low-income Medicare beneficiaries, as a percent of the eligible Medicare population, 2016–22.

SOURCE Authors’ analysis of the Medicare Current Beneficiary Survey Limited Data Set files, 2016–22. NOTES Beneficiaries with enrollment in multiple programs during a calendar year were classified into one mutually exclusive category using the following hierarchy: Qualified Medicare Beneficiary (QMB) program, Medicare Savings Program (MSP) covering Medicare Part A or Part B premiums only, full-benefit Medicaid and not in any MSP, and Medicare Part D Low-Income Subsidy (LIS) only. Results from 2020 do not sum to 100% because of rounding.

CHARACTERISTICS OF QMB-ELIGIBLE BENEFICIARIES

We examined pooled data for all QMB-eligible beneficiaries for the period 2016–22 (n = 9,736); 62 percent were female, and 35 percent were younger than age sixty-five (exhibit 2). Forty-six percent were White, 24 percent were Black, and 22 percent were Hispanic. Forty-one percent had less than a high school diploma; 21 percent had limited English-speaking proficiency; and 44 percent had incomes at or below 75 percent of poverty. Forty percent of QMB-eligible beneficiaries reported fair or poor health, 24 percent reported six or more chronic conditions, 43 percent had difficulty with at least one activity of daily living, and 36 percent reported cognitive impairment. More than half (54 percent) were enrolled in Medicare Advantage, 12 percent reported having Medigap or other supplemental insurance, and 94 percent were enrolled in Part D. Twenty-four percent of eligible beneficiaries lived in states that increased Medicare Savings Program income or asset limits above federal limits, and 57 percent lived in a Medicaid expansion state.

EXHIBIT 2. Characteristics of Medicare beneficiaries who were eligible for the Qualified Medicare Beneficiary (QMB) program, by QMB take-up status, 2016–22.

Characteristics	All QMB- eligible (N = 9,736)	Took up QMB (n = 6,434)	Did not take up QMB (n = 3,302)	p value
Age (years)
64 or younger	35%	39%	29%	<0.001
65–69	19	19	19
70–74	16	15	17
75–79	12	11	13
80 or older	18	16	21
Female sex	62	64	59	0.02
Race and ethnicity
American Indian or Alaska Native	1	1	1	<0.001
Asian	5	5	3
Black	24	23	25
Hispanic	22	27	14
White	46	41	55
Unknown or other	2	2	2
In a core-based statistical area (urban)	79	81	76	0.001
Medicare coverage
Medicare Advantage	54	56	51	<0.001
Traditional Medicare only	33	36	29
Traditional Medicare with Medigap	5	3	9
Traditional Medicare with other supplemental insurancea	7	5	11
Enrolled in Medicare Part D	94	100	84	<0.001
Married	22	16	32	<0.001
Less than a high school diploma	41	44	35	<0.001
Speaks English not well or not at all	21	25	13	<0.001
Annual income-to-FPL ratio
<50% FPL	14	12	16	<0.001
50–75% FPL	30	32	26
75–100% FPL	57	56	58
Self-reported health status				<0.001
Excellent or very good	27	23	32
Good	33	34	32
Fair or poor	40	43	35
No. of chronic conditions				<0.001
0	5	4	6
1–3	41	39	44
4–5	31	31	31
6 or more	24	26	20
No. of ADL limitations				<0.001
0	57	55	60
1–2	25	25	25
3–6	18	20	14
Any cognitive impairmentb	36	40	31	<0.001
State characteristics
Expanded MSP income or asset limits	24	23	25	0.08
Auto-enrolled SSI recipients in Medicaid	87	88	85	0.03
Expanded Medicaid to cover adults with income up to 138 percent of FPL	57	58	53	0.03
Had medically needy Medicaid program for aged and disabled adults	70	72	67	0.01

SOURCE Authors’ analysis of the Medicare Current Beneficiary Survey (MCBS) Limited Data Set files, 2016–22. NOTES Our analysis accounted for the sampling design of the MCBS, using the weights provided in the Limited Data Set files. Some results in the following categories do not sum to 100% because of missing data and/or rounding: age, race/ethnicity, Medicare coverage, annual income to federal poverty level (FPL) ratio, self-reported health status, number of chronic conditions, and number of activities of daily living (ADL) limitations. Percentages were calculated as the average of beneficiaries pooled during the period 2016–22. p values were obtained from chi-square tests that compared eligible beneficiaries by QMB take-up status. MSP is Medicare Savings Program. SSI is Supplemental Security Income. ^aOther supplemental could include retiree health coverage from an employer, regardless of Medigap. ^bIncludes dementia or Alzheimer’s diagnosis.

BENEFICIARY CHARACTERISTICS ASSOCIATED WITH QMB ENROLLMENT

Compared with QMB-eligible beneficiaries who did not take up QMB, QMB enrollees were more likely to be younger than age sixty-five, female, Asian or Hispanic, and unmarried; have less than a high school diploma; and have limited English-speaking proficiency (all p < 0.05) (exhibit 2). Beneficiaries who enrolled in QMB also were more likely than non-enrolled eligible beneficiaries to report fair or poor health and cognitive impairment (both p < 0.05). Last, QMB-eligible beneficiaries who took up versus did not take up QMB were more likely to be enrolled in Medicare Advantage and Medicare Part D and to live in states that expanded Medicaid (all p < 0.05). These findings were largely consistent in multivariable analyses (appendix exhibit A2).²⁶

ASSOCIATION BETWEEN QMB TAKE-UP AND COST-RELATED BARRIERS TO CARE

Among QMB-eligible beneficiaries, those who enrolled in QMB were less likely than those who did not enroll to report delaying seeking care because of worries about cost (10 percent versus 17 percent; p < 0.001) or having problems paying medical bills (11 percent versus 21 percent; p < 0.001) (exhibit 3). These results were consistent in multivariable analyses that assessed the difference for those who took up versus did not take up QMB in the probability of delaying care because of cost (8 percentage points less likely) or having problems paying medical bills (14 percentage points less likely) (appendix exhibit A2).²⁶

EXHIBIT 3. Cost-related barriers in access to care among Qualified Medicare Beneficiary (QMB) program–eligible Medicare beneficiaries, by QMB take-up status, 2016–22.

SOURCE Authors’ analysis of the Medicare Current Beneficiary Survey Limited Data Set files, 2016–22. NOTES Percentages were calculated as the average of beneficiaries pooled from the period 2016–22. Analysis of problems paying medical bills included 2017–22 data only, because the question was not asked in 2016. The differences between QMB eligibles enrolled versus not enrolled in QMB for each cost-related barrier to care were statistically significant (p < 0.001). p values were obtained from chi-square tests that compared eligible beneficiaries by QMB take-up status.

SENSITIVITY ANALYSES

Sensitivity analyses found that QMB take-up was lower compared with the primary analysis when we defined it as eligible beneficiaries’ enrollment in QMB for twelve continuous months (55 percent in 2016 and 60 percent in 2022; appendix exhibit A3)²⁶ and when we set QMB eligibility at 120 percent of the Medicare Savings Program income and asset limits (appendix exhibit A4).²⁶ Take-up rates holding states to their 2016 QMB income and asset limits were the same as in the main findings (appendix exhibit A5).²⁶

Results of sensitivity analyses examining differences in beneficiary characteristics between QMB-eligible beneficiaries enrolled in QMB continuously for twelve months versus those not continuously enrolled for this period were similar to those of the primary analysis (appendix exhibit A6).²⁶

Discussion

We found that about one-third of QMB-eligible beneficiaries were not enrolled in the program in 2022. Take-up was higher among beneficiaries who were younger than age sixty-five, Asian or Hispanic, or in fair or poor health. Among eligible beneficiaries, those who enrolled in QMB were less likely to report delaying care because of cost or having problems paying for care.

Our findings of limited QMB take-up and differences in characteristics of QMB enrollees and non-enrolled eligible populations are consistent with those of past studies.^9,37 Our estimate of 66 percent take-up in 2022 is higher than that of a 2009–10 study, which found a QMB take-up rate of 53 percent.⁸ These higher estimates could reflect policy changes that occurred between the two study periods. For example, the Medicare Improvements for Patients and Providers Act of 2008 increased the asset limit for QMB and other Medicare Savings Programs in 2010 to be more closely aligned with the Part D Low-Income Subsidy asset test, and it required the Social Security Administration to forward Part D Low-Income Subsidy applicants’ leads data to states to initiate a Medicare Savings Program application.²¹ Although a recent report suggests that the Medicare Improvements for Patients and Providers Act might not have been associated with increases in Medicare Savings Program take-up, it represented an initial step in better aligning Part D Low-Income Subsidy and Medicare Savings Program eligibility and enrollment.¹¹

Since 2014, most states have expanded Medicaid to cover adults with income up to 138 percent of the federal poverty level, pursuant to the ACA.²⁹ Although low-income Medicare beneficiaries are not eligible for Medicaid through this pathway, having Medicaid enrollment before entering Medicare may increase QMB take-up if Medicaid beneficiaries have greater awareness of available assistance or if they receive outreach and support to transition to the Medicare Savings Program once they enter Medicare. We found that QMB take-up was more likely among beneficiaries residing in Medicaid expansion states versus nonexpansion states; however, we did not assess the causal impact of Medicaid expansion on QMB take-up, and there could have been other differences between Medicaid expansion and nonexpansion states that contributed to this association, such as in Medicaid enrollment and renewal processes or resources available to beneficiaries that facilitate enrollment in Medicare Savings Programs. Notably, the highest QMB take-up rates in our study period occurred in 2020–22, when states halted Medicaid redeterminations during the COVID-19 public health emergency.^30,31 Future research is needed to assess whether these take-up rates persist after the 2023–25 Medicaid unwinding.

CMS estimates that the 2023 rule to streamline the Medicare Savings Program enrollment process will increase Medicare Savings Program enrollment by 860,000 people by boosting enrollment among those receiving SSI and the Part D Low-Income Subsidy.²² In 2022, 6 percent of QMB-eligible beneficiaries were enrolled in full-benefit Medicaid and not enrolled in any Medicare Savings Program (exhibit 1). Under the 2023 CMS rule, beneficiaries in this group who receive SSI are automatically enrolled in QMB in thirty-five states that automatically enroll SSI recipients into Medicaid. We also found that 3 percent of QMB-eligible beneficiaries in 2022 were enrolled in the Part D Low-Income Subsidy only. The CMS rule encourages, but does not require, states to fully align income and asset counting rules between the Medicare Savings Program and Part D Low-Income Subsidy. Greater state adoption of the Low-Income Subsidy eligibility rules for Medicare Savings Program eligibility determinations would facilitate enrolling a greater share of Low-Income Subsidy enrollees in the Medicare Savings Program, using leads data from their Low-Income Subsidy applications, and would reduce the potential burden of filing multiple applications for separate financial assistance programs with similar eligibility criteria. Because dual-eligible beneficiaries are automatically redetermined each year to be eligible for the Low-Income Subsidy, increasing Medicare Savings Program enrollment among Low-Income Subsidy enrollees could also have the benefit of reducing annual Low-Income Subsidy churn.³⁸

Nevertheless, we found that 17 percent of QMB-eligible beneficiaries in 2022 were not enrolled in Medicare Savings Programs, full-benefit Medicaid, or the Part D Low-Income Subsidy and thus would be unaffected by the 2023 CMS rule. Therefore, additional outreach to increase awareness of QMB is likely needed.^10,39 Research also is needed on which approaches to simplify enrollment processes and reduce administrative burden are most effective in increasing QMB take-up. The asset test for Medicare Savings Programs has been identified as a major source of administrative burden to beneficiaries and states; however, we found that QMB take-up among eligible beneficiaries in states with increased Medicare Savings Program income or asset limits was not higher compared with among those residing in states that used federal limits. These results are consistent with the findings of past studies.^4,8,39,40 Information on how states have implemented Medicare Savings Program expansions, particularly regarding outreach and education activities, and enrollment changes after expansion would provide valuable insights for future efforts to increase QMB take-up.

Our analysis found that 40 percent of QMB-eligible Medicare beneficiaries reported fair or poor health status, and most had multiple chronic conditions. Increasing QMB take-up has the potential to improve financial protections and access to care for this high-risk and high-need population. Findings from previous research suggest that QMB coverage is associated with increases in physician office visits.¹⁵ Further evidence suggests that reducing cost sharing increases the use of preventive services among beneficiaries with low incomes, whereas increases in cost sharing are associated with reductions in outpatient service use and increases in hospitalizations.^16,17,41 Additional work is needed to evaluate whether and the extent to which the financial assistance provided to beneficiaries in QMB affects health outcomes.

Conclusion

Approximately a third of eligible low-income Medicare beneficiaries did not enroll in QMB in 2022. Although new CMS rules could increase QMB take-up among beneficiaries receiving SSI or the Part D Low-Income Subsidy, our findings suggest that additional policies and programs may be needed to increase QMB take-up and improve health care access and affordability for millions of low-income beneficiaries.