Evaluating Access to Care for Medicare Beneficiaries Younger Than 65 Years

ABSTRACT

Objectives: Individuals younger than 65 years can qualify for Medicare if they have long-term disabilities or certain qualifying conditions. These beneficiaries—particularly the non–dual-eligible population—may experience cost and access barriers to medical care. We examined the association between Medicare coverage type and reported barriers to care.

Study Design: Multivariable linear probability models assessed the association between self-reported Medicare coverage and patient-reported outcomes by dual-eligibility status.

Methods: Using 2012-2020 data from the Health and Retirement Study, we compared self-reported sociodemographic and health-related characteristics of non–dual-eligible and dual-eligible beneficiaries aged 50 to 64 years by Medicare coverage type at their baseline interview. We then examined the following self-reported outcomes: experiencing cost-related medication nonadherence, delaying care due to cost, not having a usual source of care, and having trouble finding a doctor.

Results: Among non–dual-eligible beneficiaries, enrollment in traditional Medicare (TM) plus supplemental coverage vs TM with no supplemental coverage was associated with lower reported rates of experiencing cost-related medication nonadherence (–7.5 percentage point [PP] change; 95% CI, –12.1 to –3.0), delaying care due to cost (–9.8 PP; 95% CI, –13.3 to –6.3), and having no usual source of care (–5.5 PP; 95% CI, –8.9 to –2.1). Compared with TM with no supplement, Medicare Advantage enrollment was associated with lower rates of delaying care due to cost (–4.2 PP; 95% CI, –7.6 to –0.7) and having no usual source of care (–5.2 PP; 95% CI, –8.2 to –2.3). Among dual-eligible beneficiaries, outcomes largely did not differ by coverage type. Switching from traditional Medicare to Medicare Advantage was associated with trouble finding a doctor for dual-eligible beneficiaries.

Conclusions: Enrollment in less generous Medicare coverage was associated with greater cost and access barriers to care for beneficiaries younger than 65 years.

Am J Manag Care. 2025;31(5):In Press

Takeaway Points

Enrollment in Medicare coverage without out-of-pocket protections was associated with a higher likelihood of reporting cost and access barriers to care for beneficiaries aged 50 to 64 years.

• Enrollment in traditional Medicare without supplemental coverage was associated with greater reporting of financial and access issues among those without Medicaid.
• Beneficiaries with Medicaid experienced minimal financial and access differences by coverage type.
• Switching from traditional Medicare to Medicare Advantage was associated with trouble finding a doctor for dual-eligible beneficiaries.

Although Medicare is widely known as coverage for older adults in the US, certain individuals younger than 65 years can qualify for coverage if they have a long-term disability or other qualifying condition (ie, end-stage renal disease [ESRD] or amyotrophic lateral sclerosis [ALS]). Individuals qualifying for Medicare via long-term disability (ie, they have a disability that makes them unable to engage in “substantial gainful activity” due to a physical or mental ailment expected to last 12 months or until death)¹ must receive Social Security and Disability Insurance (SSDI) payments for 24 months before gaining Medicare coverage. Notably, there is a 5-month waiting period to begin receiving SSDI payments. Those qualifying for Medicare due to ESRD or ALS receive coverage immediately after diagnosis.¹ Beneficiaries younger than 65 years comprised 13.1% of the Medicare population and 17.8% of Medicare spending in 2020.²

Due to their reasons for qualifying for Medicare, beneficiaries younger than 65 years are generally medically complex and face challenges navigating the Medicare program and the health care system.³ These younger Medicare beneficiaries are more likely to report worse physical and mental health,^1,3 identify as persons of color,⁴ and have lower incomes¹ compared with beneficiaries who qualify for Medicare due to age. Several studies have found that beneficiaries younger than 65 years experience less satisfaction with their coverage^3,5,6 as well as issues accessing and affording care.^5,7,8

Because younger Medicare beneficiaries may be particularly vulnerable to financial barriers and issues accessing care, Medicare plan enrollment choices may be particularly consequential for them. First, under traditional Medicare (TM), those without supplemental coverage (eg, Medigap) face 20% coinsurance for Part B services with no out-of-pocket limit. Most older beneficiaries who choose TM obtain supplemental coverage through Medigap, which limits or eliminates out-of-pocket spending. However, beneficiaries younger than 65 years largely do not qualify for federal guaranteed issue protections for Medigap plans³ and can be denied coverage or charged very high premiums for coverage depending upon state-specific Medigap regulations.

Second, enrollees can select Medicare Advantage (MA) plans, which are required to have out-of-pocket spending limits. However, these limits may apply only to in-network care (with higher limits required for out-of-network care, if available on the selected plan). Given the high health care needs of those who qualify for Medicare due to disability, these cost-sharing limits may still be too high. These challenges would be exacerbated among those who need to receive care outside their plan’s network.

Research on Medicare beneficiaries younger than 65 years remains limited. Prior studies^4-7 have not examined changes in Medicare plan enrollment choices for beneficiaries over time, nor have they focused on differences in care experience and related outcomes by plan enrollment choice. We sought to understand whether a beneficiary’s plan enrollment choice and switching coverage were associated with cost and access barriers to care. We separately considered experiences of enrollees aged 50 to 64 years who are dually eligible for Medicaid and those without Medicaid because most dual-eligible beneficiaries face no cost sharing for many services, which may affect plan choices and outcomes.

METHODS

Data Sources and Study Population

We used 2012-2020 data from the Health and Retirement Study (HRS). HRS is a nationally representative longitudinal survey of more than 43,000 adults 50 years and older. The survey is administered every 2 years, and the data provide information on several dimensions of older adults’ lives including finances, retirement, and health care experiences/utilization.⁹

Our study population included participants aged 50 to 64 years. We excluded participants enrolled in military health plans during the study period, younger than 50 years (because HRS is not representative of this population), and missing Medicare enrollment information during the study period. We also censored observations in periods after a participant changed dual-eligibility status. This study received an institutional review board waiver from Vanderbilt University Medical Center.

Study Measures

Our main independent variable was self-reported Medicare enrollment at the time of survey wave completion. In line with previous work, we categorized non–dual-eligible beneficiaries into 3 mutually exclusive groups: TM, TM plus supplemental coverage (eg, Medigap, employer-sponsored, or retiree benefits), or MA.¹⁰ Because dual-eligible beneficiaries face limited out-of-pocket costs, we categorized participants dually eligible for Medicare and Medicaid into 2 enrollment groups: MA and TM. Participants who were enrolled in Medicare and answered “yes” to current Medicaid coverage were dual eligible.

Our primary self-reported financial outcomes were cost-related medication nonadherence (ie, taking less medication than prescribed because of cost) and delayed care because of cost (ie, did not receive needed medical care because they could not afford it). Our primary self-reported access measures included not having a usual source of care (ie, not having a place to go when they are sick or need advice about their health) and having trouble finding a doctor (ie, trouble finding a doctor who would see them).

Covariates included sociodemographic and health-related characteristics self-reported at beneficiaries’ baseline survey: age, sex, race (Black, other, or White), Hispanic ethnicity (because Black and Hispanic beneficiaries have high rates of MA participation),¹¹ partnered/marital status, education (less than high school or more than high school), any consumer debt (ie, debts such as credit card balances, medical debts, life insurance policy loans, etc), survey wave–specific income quintile, health status (excellent/very good/good or fair/poor), hospitalizations, number of comorbidities, Center for Epidemiological Studies Depression Scale score, and survey wave fixed effects.

Statistical Analysis

We separately compared self-reported sociodemographic and health-related characteristics of non–dual-eligible and dual-eligible beneficiaries by Medicare coverage type at their baseline interview using χ² tests for categorical variables and analysis of variance for continuous variables.

We estimated adjusted linear probability models with survey wave fixed effects to assess the association between Medicare enrollment and cost and access outcomes. We present unadjusted proportions comparing outcomes by enrollment status and adjusted percentage point (PP) changes with 95% CIs for all outcomes. In all analyses, we stratified our models by dual-eligibility status because dual-eligible beneficiaries have lower out-of-pocket costs, which may influence their Medicare enrollment decisions. We clustered SEs at the survey participant level to account for repeated observations. We also performed stratified analyses by marital status, reported debt, chronic conditions, and income quintile to assess whether results differ by these subgroups.

Sensitivity Analysis

We conducted a sensitivity analysis to focus on the association between switching coverage types and outcomes of interest. For this analysis, we restricted the cohort to participants in at least 2 consecutive survey waves to examine whether changes in Medicare enrollment were associated with our outcomes of interest. Specifically, beneficiaries who change from less generous to more generous coverage may experience improvements in outcomes over time. Among this subsample, we used linear probability models with a first-difference estimator¹² to account for individual time-invariant characteristics and to examine changes in access and financial burden that may be accompanied by a beneficiary switching coverage. To be able to compare the findings from this modeling approach to our main analysis, we repeated our original modeling approach (that has no first-difference estimator) in the same study subpopulation.

Analyses were conducted using Stata/MP 17.0 (StataCorp LLC); results were 2-sided and statistically significant at P < .05.

RESULTS

Medicare Enrollment

Our main analysis included 3249 participants (mean [SD] age, 57.8 [4.4] years; 59.2% female), of whom 56.7% were Medicaid dual eligible at their baseline survey. Beneficiaries participated in a median (IQR) of 3 (2) survey waves. Enrollment in MA increased from 38.1% in 2012 through 2016 to 44.2% in 2018 through 2020 among non–dual-eligible beneficiaries while participation in TM only decreased (37.9% in 2012-2016 and 30.6% in 2018-2020) (eAppendix Figure [eAppendix available at ajmc.com]). MA coverage also grew over the same period for dual-eligible participants (69.2% in 2012-2016 and 74.5% in 2018-2020).

Cohort Characteristics

When comparing characteristics by enrollment group among non–dual-eligible and dual-eligible beneficiaries, we observe several differences (Table 1). Among non–dual-eligible participants, TM beneficiaries (mean [SD] age, 58.3 [4.3] years) and MA beneficiaries (58.6 [4.7] years) were slightly younger on average than beneficiaries with TM plus supplemental coverage (59.9 [4.5] years; P < .001). Respondents enrolled in MA were more likely to be Black (42.3%) compared with TM beneficiaries (TM with no supplemental: 38.0%; TM plus supplemental: 27.0%; P < .001). There also were differences in the top income quintile by enrollment group; participants in TM with supplemental coverage were most likely to be in the fifth income quintile (50.7%) compared with those in MA (23.1%) and TM with no supplement (14.7%; P < .001). Among dual-eligible participants, MA dual-eligible beneficiaries were slightly younger than TM dual-eligible enrollees (mean [SD], 56.8 [4.0] vs 57.5 [4.4] years; P = .002) and more likely to be Hispanic (28.8% vs 23.2%; P = .019).

Main Analysis

In unadjusted analyses, non–dual-eligible participants in TM with no supplemental coverage reported the highest rates of experiencing cost-related medication nonadherence (30.2% in TM with no supplement, 26.5% in MA, and 18.5% in TM plus supplemental coverage), delaying care due to costs (24.2%, 17.4%, and 8.6%, respectively), having no usual source of care (18.2%, 10.5%, and 8.0%), and having trouble finding a doctor (7.7%, 6.4%, and 4.7%) (Figure 1).

Similar patterns were observed in adjusted analyses. Compared with TM with no supplement, enrollment in TM plus supplemental coverage was associated with a –7.5 PP (95% CI, –12.1 to –3.0) difference in cost-related medication nonadherence (Table 2), a 29% relative decrease from the mean. Participants in MA or TM plus supplemental coverage were less likely to report delays in accessing care due to cost, representing a 24% and 56% relative decrease from the sample mean, respectively. Both groups were also less likely to report having no usual source of care relative to beneficiaries in TM with no supplement, with a –5.2 PP (95% CI, –8.2 to –2.3) difference (a 41% relative decrease) in reporting access issues among MA participants and a –5.5 PP (95% CI, –8.9 to –2.1) difference (a 44% relative decrease) among those enrolled in TM plus supplemental coverage. We observed no association between Medicare coverage type and having trouble finding a doctor.

When examining unadjusted outcomes, dual-eligible MA enrollees reported lower rates of cost-related medication nonadherence, at 17.1% in MA compared with 20.5% in TM, but fairly similar rates in delaying care because of costs (14.8% in MA vs 13.5% in TM). TM enrollees reported higher rates of not having a usual source of care than MA enrollees (17.9% vs 12.9%) and lower rates of trouble finding a doctor (6.8% vs 8.6%) (Figure 2).

In adjusted analyses of dual-eligible beneficiaries, there was no association between Medicare coverage type and financial measures or trouble finding a doctor. However, MA enrollees were less likely to report having no usual source of care relative to TM enrollees.

When examining outcomes among certain subgroups (eg,marital status, reported debt, chronic conditions, and income quintile), results largely aligned with the primary analysis: Less generous coverage was associated with worse outcomes among the non–dual-eligible population. Among the dual-eligible population, we observed similarities to the primary analysis; however, certain subgroups had an increased likelihood of having trouble finding a doctor in MA relative to TM (eAppendix Tables 1-4).

Sensitivity Analysis

Among the non–dual-eligible beneficiaries, we found a few differences between switchers and nonswitchers based on initial coverage. Among participants whose baseline coverage was TM, switchers were older (mean age, 63.2 vs 60.8 years; P < .001) and more likely to be in the fifth income quintile (23.3% vs 14.2%; P < .001). Switchers initially in TM with supplemental coverage were less likely to be in the top income quintile (37.9% vs 57.5%; P < .001) (eAppendix Table 5). Dual-eligible switchers initially in TM were more likely to have less than a high school education (85.6% vs 78.0%; P = .037), and those in MA at their baseline survey were older (62.0 vs 59.8 years; P < .001) (eAppendix Table 6).

The multivariate models yielded similar, more pronounced results (eAppendix Table 7). Our first-difference models largely indicated no association between switching Medicare coverage and our outcomes. However, participants who switched from MA to TM were more likely to report no usual source of care (6.9 PP; 95% CI, 0.4-13.4) (eAppendix Table 8).

Among dual-eligible participants, results were also mostly the same. However, MA enrollment was associated with a 3.7-PP (95% CI, 1.2-6.1) increase in reporting trouble finding a doctor relative to dual-eligible beneficiaries in TM (eAppendix Table 9). Dual-eligible participants who switched from TM to MA were 5.1 PP (95% CI, 0.5-9.7) more likely to report trouble finding a doctor (eAppendix Table 10).

DISCUSSION

In this cohort study of Medicare beneficiaries aged 50 to 64 years, we found that individuals without out-of-pocket protections either through a Medicare supplemental insurance policy (eg, Medigap), an MA plan, or Medicaid were more likely to report cost and access barriers to care. For non–dual-eligible participants, we also found that switching coverage was largely unassociated with cost or access barriers, except for when switching from MA to TM. MA enrollment may bring financial relief to beneficiaries due to plan out-of-pocket maximums. However, previous work comparing beneficiaries who switched from TM to MA relative to beneficiaries who remained in TM found minimal differences in financial burden between the groups.¹³ These results may indicate that once a beneficiary has made their enrollment choice, switching to another plan may not alleviate perceived cost and access burdens. We also found that switching from TM to MA among these dual-eligible beneficiaries was associated with an increased likelihood of reporting trouble finding a doctor.

Our results highlight how important protections from high out-of-pocket costs are for beneficiaries who largely cannot work and generally have lower incomes relative to the older Medicare population.¹ In 2023, median household income for beneficiaries younger than 65 years was $23,900, and roughly 25% lived on annual incomes less than $15,000.¹⁴ In addition to low incomes, most individuals qualifying for Medicare due to long-term disabilities encounter long wait times before qualifying for benefits, meaning they may be uninsured while living with a debilitating condition.¹⁵ They are also largely unable to access supplemental coverage in the Medigap market,³ which likely contributes to lower uptake of Medigap policies relative to the older Medicare population.¹⁶ Although guaranteed issue and community rating protections are available to individuals in the commercial market since passage of the Affordable Care Act, they have not been extended to the Medigap market and vary significantly by state. This can leave younger beneficiaries financially vulnerable when choosing coverage under TM.

Policy makers should consider state or federal options that would extend community rating and guaranteed issue requirements for supplemental coverage to Medicare beneficiaries younger than 65 years.¹⁷ Alternatively, reforms to TM’s cost-sharing structure that reduce financial burden and lessen the need for Medigap¹⁸ could make the choice between choosing TM and MA less financially consequential for beneficiaries. However, there are issues that should be considered, including whether broader Medigap access could induce adverse selection and lead to sicker beneficiaries disproportionately enrolling in TM once their health care needs increase.

Our analysis also highlights the role MA plans play in caring for the younger Medicare population. Consistent with other studies, our analysis showed that MA beneficiaries—regardless of dual-eligibility status—were more likely to report a usual source of care, which is supported by other studies.⁴ These results indicate that plan incentives to coordinate care for enrollees may improve patients’ access to primary care services to manage their conditions.^19,20 However, in sensitivity analyses, we found that dual-eligible beneficiaries in MA were more likely to report trouble finding a doctor when switching from TM to MA. These findings may reflect how MA plans’ ability to restrict coverage to in-network providers could narrow access to some physicians, especially when switching from TM’s almost universal coverage of all physicians.

Our results align with those of other studies of the Medicare population younger than 65 years showing that beneficiaries report not receiving needed care^5,7 and having issues accessing care.^8,15 However, our work builds on this literature by exploring the role of supplemental coverage in TM in financially protecting beneficiaries aged 50 to 64 years and provides novel evidence of how non–dual-eligible and dual-eligible beneficiaries likely make different enrollment decisions and the financial and access implications of those choices.

Limitations

There are several limitations to our analysis. First, our results examine the association between enrollment and cost and access barriers only among adults aged 50 to 64 years, leaving out younger Medicare beneficiaries who may make different enrollment choices than this age group and reducing generalizability to the entire Medicare population younger than 65 years. Second, we were unable to ascertain the specific reason for Medicare entitlement in our cohort, biasing our multivariable results. Because individuals qualifying for ESRD were excluded from enrollment in MA until 2021,²¹ they likely made different enrollment choices than individuals qualifying due to disability during our study period.²² However, our first-difference estimates could account for this. Third, we did not ascertain whether there were negative health consequences that arose from reported barriers to accessing medical care. Because this population already reports poor or fair health,³ they may be more vulnerable to the poor health outcomes associated with delaying care. Fourth, although the HRS is a nationally representative survey across several time periods, the data are self-reported and susceptible to bias (eg, recall bias, social desirability bias). However, the survey’s methods for data collection minimize this bias.⁹ Fifth, there may be unobserved differences in enrollment groups that likely resulted in residual confounding, and results should be taken as associations. Lastly, we have limited details on the source of the participants’ supplemental coverage. Because access to Medigap coverage is limited for those younger than 65 years, the portion of beneficiaries who can access supplemental coverage likely differs from the general population younger than 65 years.

CONCLUSIONS

In this cohort study of Medicare beneficiaries aged 50 to 64 years, beneficiaries exposed to TM’s cost-sharing structure reported issues with affording care. Our study findings highlight the need for continued focus on these beneficiaries’ experiences within the Medicare program. 

Author Affiliations: Department of Health Policy, Vanderbilt University School of Medicine (EMA, SAJ, LMK, SBD), Nashville, TN; Division of Geriatric Medicine, Department of Medicine, University of Colorado Anschutz Medical Campus (LHN), Aurora, CO; Department of Medicine, University of North Carolina at Chapel Hill School of Medicine (WAW), Chapel Hill, NC; Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill (WAW), Chapel Hill, NC; University of Michigan School of Nursing (CRF), Ann Arbor, MI; Department of Health Management and Policy, University of Michigan School of Public Health (CRF), Ann Arbor, MI; Rogel Cancer Center, University of Michigan (CRF), Ann Arbor, MI; Vanderbilt-Ingram Cancer Center (SBD), Nashville, TN.

Source of Funding: This study was funded by the Leukemia and Lymphoma Society. Dr Jazowski was supported by the Agency for Healthcare Research and Quality (T32 HS026122). The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality. Dr Nicholas was supported by the National Institute on Aging. Dr Friese was supported by the National Cancer Institute (P30 CA046592). Dr Dusetzina was supported as a research scientist by the National Cancer Institute (P30 CA068485).

Author Disclosures: Dr. Nicholas completed this paper as part of her grant-funded research and reports consulting fees from Acumen, LLC. Dr Wood reports research funding to his institution from Pfizer Inc and Genentech and consulting fees from the ASH Research Collaborative, Koneksa Health, and Teladoc. Dr Dusetzina reports funding from Arnold Ventures and the Commonwealth Fund and is a member of the Institute for Clinical and Economic Review (ICER) Midwestern Comparative Effectiveness Public Advisory Council, the Medicare Payment Advisory Commission (MedPAC), and the editorial board of The American Journal of Managed Care. This work does not necessarily represent the official position of ICER or MedPAC. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (EMA, SAJ, LHN, WAW, CRF, SBD); acquisition of data (SBD); analysis and interpretation of data (EMA, SAJ, LHN, LMK, WAW, CRF, SBD); drafting of the manuscript (EMA); critical revision of the manuscript for important intellectual content (EMA, SAJ, LHN, LMK, WAW, CRF, SBD); statistical analysis (EMA, SAJ, LHN, LMK); obtaining funding (LHN, SBD); and supervision (SBD).

Address Correspondence to: Emma M. Achola, BA, Department of Health Policy, Vanderbilt University School of Medicine, 2525 West End Ave, Ste 1275, Nashville, TN 37203. Email: emma.m.achola@vanderbilt.edu.

REFERENCES

1. Cubanski J, Neuman T, Damico A. Medicare’s role for people under age 65 with disabilities. KFF. August 12, 2016. Accessed March 6, 2023. https://www.kff.org/medicare/issue-brief/medicares-role-for-people-under-age-65-with-disabilities

2. A Data Book: Health Care Spending and the Medicare Program. Medicare Payment Advisory Commission; 2023. Accessed May 5, 2024. https://www.medpac.gov/wp-content/uploads/2023/07/July2023_MedPAC_DataBook_SEC_v2.pdf

3. Cubanski J, Mohamed M, Neuman T, Claxton G. Overall satisfaction with Medicare is high, but beneficiaries under age 65 with disabilities experience more insurance problems than older beneficiaries. KFF. October 26, 2023. Accessed November 10, 2023. https://www.kff.org/medicare/issue-brief/overall-satisfaction-with-medicare-is-high-but-beneficiaries-under-age-65-with-disabilities-experience-more-insurance-problems-than-older-beneficiaries/

4. Johnston KJ, Wen H, Pollack HA. Comparison of ambulatory care access and quality for beneficiaries with disabilities covered by Medicare Advantage vs traditional Medicare insurance. JAMA Health Forum. 2022;3(1):e214562. doi:10.1001/jamahealthforum.2021.4562

5. Koma JW, Fuglesten Biniek J, Cubanski J, Neuman T. Access problems and cost concerns of younger Medicare beneficiaries exceeded those of older beneficiaries in 2019. Health Aff (Millwood). 2023;42(4):470-478. doi:10.1377/hlthaff.2022.00501

6. Bernal OA, McClintock HF, Kurichi JE, et al. Patient satisfaction and perceived quality of care among younger Medicare beneficiaries according to activity limitation stages. Arch Phys Med Rehabil. 2019;100(2):289-299. doi:10.1016/j.apmr.2018.09.114

7. Johnston KJ, Wen H, Joynt Maddox KE, Pollack HA. Ambulatory care access and emergency department
use for Medicare beneficiaries with and without disabilities. Health Aff (Millwood). 2021;40(6):910-919. doi:10.1377/hlthaff.2020.01891

8. Cubanski J, Neuman P. Medicare doesn’t work as well for younger, disabled beneficiaries as it does for older enrollees. Health Aff (Millwood). 2010;29(9):1725-1733. doi:10.1377/hlthaff.2009.0962

9. Fisher GG, Ryan LH. Overview of the Health and Retirement Study and introduction to the special issue. Work Aging Retire. 2018;4(1):1-9. doi:10.1093/workar/wax032

10. Jazowski SA, Achola EM, Nicholas LH, Wood WA, Friese CR, Dusetzina SB. Comparing Medicare plan selection among beneficiaries with and without a history of cancer. Health Aff Sch. 2024;2(2):qxae014. doi:10.1093/haschl/qxae014

11. Meyers DJ, Mor V, Rahman M, Trivedi AN. Growth in Medicare Advantage greatest among Black and Hispanic enrollees. Health Aff (Millwood). 2021;40(6):945-950. doi:10.1377/hlthaff.2021.00118

12. Schaffer ME. XTIVREG2: Stata module to perform extended IV/2SLS, GMM and AC/HAC, LIML and k-class regression for panel data models. Statistical Software Components, Boston College Department of Economics. November 2005. Updated June 26, 2020. Accessed January 9, 2024. https://ideas.repec.org/c/boc/bocode/s456501.html

13. Park S, Meyers DJ, Trivedi AN. Association of Medicare Advantage enrollment with financial burden of care: a retrospective cohort study. Ann Intern Med. 2024;177(7):882-891. doi:10.7326/M23-2480

14. Cottrill A, Cubanski J, Neuman T, Smith K. Income and assets of Medicare beneficiaries in 2023. KFF. February 5, 2024. Accessed May 13, 2024. https://www.kff.org/medicare/issue-brief/income-and-assets-of-medicare-beneficiaries-in-2023/

15. Iezzoni LI. Quality of care for Medicare beneficiaries with disabilities under the age of 65 years. Expert Rev Pharmacoecon Outcomes Res. 2006;6(3):261-273. doi:10.1586/14737167.6.3.261

16. Clerveau G, Ochieng N, Cubanski J, Neuman T. A snapshot of sources of coverage among Medicare beneficiaries. KFF. December 13, 2023. Accessed December 13, 2023. https://web.archive.org/web/20231130194322/https://www.kff.org/medicare/issue-brief/a-snapshot-of-sources-of-coverage-among-medicare-beneficiaries/

17. Ginsburg PB, Lieberman SM. Improving access to Medigap when beneficiaries leave Medicare Advantage. Health Affairs Forefront. April 26, 2024. Accessed September 10, 2024. https://www.healthaffairs.org/content/forefront/improving-access-medigap-beneficiaries-leave-medicare-advantage

18. McWilliams JM. The future of Medicare and the role of traditional Medicare as competitor. N Engl J Med. 2024;391(8):763-769. doi:10.1056/NEJMsb2313939

19. Agarwal R, Connolly J, Gupta S, Navathe AS. Comparing Medicare Advantage and traditional Medicare: a systematic review. Health Aff (Millwood). 2021;40(6):937-944. doi:10.1377/hlthaff.2020.02149

20. Drzayich Antol D, Schwartz R, Caplan A, et al. Comparison of health care utilization by Medicare Advantage and traditional Medicare beneficiaries with complex care needs. JAMA Health Forum. 2022;3(10):e223451. doi:10.1001/jamahealthforum.2022.3451

21. Morgan PC, Wreschnig LA. Medicare Advantage (MA) Coverage of End Stage Renal Disease (ESRD) and Network Requirement Changes. Congressional Research Service; January 11, 2021. Accessed May 14, 2024. https://crsreports.congress.gov/product/pdf/R/R46655

22. Nguyen KH, Oh EG, Meyers DJ, Kim D, Mehrotra R, Trivedi AN. Medicare Advantage enrollment among beneficiaries with end-stage renal disease in the first year of the 21st Century Cures Act. JAMA. 2023;329(10):810-818. doi:10.1001/jama.2023.1426