Enrollment in Medicare Advantage, the private plan alternative to traditional Medicare, has increased steadily over the past two decades, with more than half of eligible beneficiaries enrolled in Medicare Advantage since 2023. The growth in enrollment has implications for federal spending, because according to the Medicare Payment Advisory Commission (MedPAC), Medicare payments to private plans are higher than spending for similar beneficiaries in traditional Medicare. In 2026, payments are 14% more per person, which translates into an additional $76 billion in federal spending this year. While payments per person relative to traditional Medicare are similar to a decade ago (15% higher), the impact on federal spending was substantially lower at $24 billion because enrollment was substantially lower with about one-third of eligible beneficiaries enrolled in a Medicare Advantage plan at that time.

Given the enrollment and spending trends, policymakers have become increasingly focused on how Medicare pays private plans, though without broad agreement on how or when to move forward with any changes to how Medicare Advantage plans are paid. In part, the difficulty stems from concerns about the effects of payment changes on beneficiaries’ choice among plans and access to supplemental benefits, such as coverage of dental, vision and hearing.

To better understand trends in the growth of the Medicare Advantage program, this brief provides current information about enrollment, including by plan type and firm. A second, companion analysis describes Medicare Advantage premiums, out-of-pocket limits, supplemental benefits offered, and prior authorization requirements in 2026.

Highlights for 2026:

• More than half (55%) of eligible Medicare beneficiaries are enrolled in Medicare Advantage in 2026. While a growing share of Medicare beneficiaries are enrolled in a Medicare Advantage plan, the pace of the increase in enrollment continued to slow in 2026.
• In 2026, nearly one quarter (23%) of Medicare Advantage enrollees are in a special needs plan (SNP), reflecting a steady increase in recent years. Most (85%) of the net increase in Medicare Advantage enrollment between 2025 and 2026 across all plan types was among SNPs, up from 48% in the prior year.
• More than three-quarters of SNP enrollment (78%) is in plans designed for people who are dually eligible for Medicare and Medicaid (D-SNPs), with plans for people with certain chronic conditions (C-SNPs) continuing to see a surge in enrollees in 2026 as in 2025. Enrollment in C-SNPS increased by 45% between 2025 and 2026, rising to 20% of SNP enrollment.
• Medicare Advantage enrollment is highly concentrated among plans owned by a small number of parent organizations, with UnitedHealth Group leading the market, and, together with Humana, accounting for nearly half (46%) of all Medicare Advantage enrollees nationwide, the same as in 2025, and consistent with the pattern in prior years. However, market shares for the leading parent organizations changed with UnitedHealth Group dropping to 26% (down from 29%), and Humana increasing to 20% (up from 17%). In absolute numbers, Humana had the largest growth in enrollment, with 1.3 million more enrollees in 2026 than in 2025. In contrast, enrollment in UnitedHealth Group plans decreased by nearly 647,000 from 2025 to 2026. 

More than half of eligible Medicare beneficiaries are enrolled in Medicare Advantage in 2026.

In 2026, more than half (55%) of eligible Medicare beneficiaries – 35.2 million out of 64.2 million Medicare beneficiaries with both Medicare Parts A and B – are enrolled in Medicare Advantage plans (Figure 1). The share of Medicare beneficiaries in Medicare Advantage plans varies by state and by county (Appendix Figures 1 and 2). Medicare Advantage enrollment as a share of the eligible Medicare population has jumped from 19% in 2007 to 55% in 2026.

Between 2025 and 2026, total Medicare Advantage enrollment grew by about 1.1 million beneficiaries, or 3% – a similar growth rate to the prior year (4%). The Congressional Budget Office (CBO) projects that the share of all Medicare beneficiaries enrolled in Medicare Advantage plans will rise to 63% by 2034 and remain at 63% for 2035 and 2036 (Figure 2).

In 2026, about six in ten Medicare Advantage enrollees are in individual plans that are open for general enrollment.

About 6 in 10 Medicare Advantage enrollees (61%), or 21.4 million people, are in plans generally available to all Medicare beneficiaries for individual enrollment (Figure 3). That is an increase of 0.2 million enrollees compared to 2025, though individual plans comprised a slightly smaller share of total Medicare Advantage enrollment in 2026 (61%) compared to 2025 (62%), and their share of enrollment has declined since 2010 when they comprised 71% of all enrollees. The decline in the share of enrollment in individual plans is due to faster enrollment growth in special needs plans (SNPs), especially since 2018.  

Special needs plans (SNPs) comprise a growing share of Medicare Advantage enrollment.

Nearly 8.2 million Medicare beneficiaries are enrolled in special needs plans (SNPs), an increase of more than 900,000 enrollees since 2025, and accounted for 85% of the net increase in Medicare Advantage enrollment over the last year. SNPs restrict enrollment to beneficiaries with significant or relatively specialized care needs, or who qualify because they are eligible for both Medicare and Medicaid. SNPs comprise a growing share of Medicare Advantage enrollment, accounting for 23% of enrollees in 2026 compared with 21% of enrollees in 2025 (Figure 4).

The increase in SNP enrollment is consistent with the increasing number of SNP plans available on average and more dual-eligible individuals having access to these plans since the Bipartisan Budget Act of 2018 made SNPs a permanent part of the Medicare Advantage program. Growth in SNP enrollment may also stem from the sharp increase in rebate payments, which have more than doubled since 2017 and are higher than for individual plans, which enables SNPs to offer extra benefits that are attractive to dual-eligible individuals.

Further, the Medicare Advantage Value-Based Insurance Design Model (VBID), which launched in 2017, gave plans the ability to provide additional interventions to enrollees such as reduced cost sharing and extra benefits based on their socioeconomic status (e.g., LIS eligibility), along with other targeting criteria. This meant many enrollees in SNPs, including those dually eligible for Medicare and Medicaid, may have been eligible for additional benefits. For example, plans enrolled in the model were able to offer benefits such as food and produce, assistance with utilities, and transportation for non-medical needs, among others, to a broader group of enrollees, when many of those benefits were typically reserved for enrollees who met the stricter definition of eligibility for Special Supplemental Benefits for the Chronically Ill (SSBCI). However, CMS discontinued this model in 2025 due to the “unprecedented cost” of this model, including increased rebates and increased Part D expenditures.

D-SNPs. Most SNP enrollees (78%) are in plans for beneficiaries dually enrolled in both Medicare and Medicaid (D-SNPs), a decline from 83% in 2025. While D-SNPs are designed specifically for dually-eligible individuals, among the 3.9 million dually-eligible enrollees with full benefits enrolled in Medicare Advantage plans, 28% were enrolled in Medicare Advantage plans that are generally available to all beneficiaries (not designed specifically for the dually-eligible population).

C-SNPs. Another 20% of SNP enrollees are in plans for people with severe chronic or disabling conditions (C-SNPs) – an increase from 16% in 2025. C-SNP enrollment in 2026 (about 1.7 million people) is 45% higher than it was in 2025 – an increase of about 518,000 enrollees. Nearly all (97%) C-SNP enrollees are in plans for people with diabetes or cardiovascular conditions in 2026, the same as in 2025.

I-SNPs. And 2% of SNP enrollees in 2026 are in plans for beneficiaries requiring a nursing home or institutional level of care (I-SNPs), the same as 2025. Enrollment in I-SNPs increased slightly in 2026 124,000 enrollees, an increase of about 9,000 enrollees since 2025.

SNP enrollment varies across states. In the District of Columbia and Puerto Rico, SNP enrollees comprise about half of all Medicare Advantage enrollees (51% in DC and 49% in PR). In eleven states, SNP enrollment accounts for at least a quarter of Medicare Advantage enrollment: 46% in MS, 40% in AR, 37% in NY, 36% in LA, 33% in FL, 32% in SC, 30% in GA, 27% in OK and AL, 26% in CT, and 26% in RI. In the remaining 39 states, fewer than a quarter of Medicare Advantage enrollees are in SNPs, including Alaska and Vermont, which have no SNP enrollment.

Enrollment in a D-SNP comprises the largest share of enrollment in SNPs in all states except New Hampshire where C-SNP enrollment is higher, though overall SNP enrollment only represents 2% of Medicare Advantage enrollment in New Hampshire. C-SNP enrollment represents a relatively large share of Medicare Advantage SNP enrollment in Illinois, Utah, South Carolina, Delaware, Nevada, and Oregon, where 40% or more of SNP enrollment is in C-SNPs.

Slightly less than one in five (16% or about 5.7 million) Medicare Advantage enrollees are in a group plan offered to retirees by an employer or union.

Group enrollment as a share of total Medicare Advantage enrollment is 16%, the lowest share since 2010, though it has generally fluctuated between 17% and 20% of enrollment. At the same time, the number of enrollees has increased from 1.8 million in 2010 to 5.7 million in 2026. The 2026 enrollment in group plans is a slight decrease from 2025 (a decrease of about 31,000 enrollees), the first time since 2010 that enrollment in this type of plan has declined year-to-year (Figure 5). With a group plan, an employer or union contracts with an insurer and Medicare pays the insurer a fixed amount per enrollee to provide benefits covered by Medicare. For example, 13 states provided health insurance benefits to their Medicare-eligible retirees exclusively through Medicare Advantage plans in 2024.

As with other Medicare Advantage plans, employer and union group plans may provide additional benefits and/or lower cost sharing than traditional Medicare and are eligible for bonus payments if they obtain required quality scores. The employer or union (and sometimes the retiree) may also pay an additional premium for these supplemental benefits. Group enrollees comprise a quarter or more of Medicare Advantage enrollees in seven states: 100% in AK, 60% in VT, 34% in MI, 31% in NJ, 28% in MD and WV, and 25% in IL.

Medicare Advantage enrollment is highly concentrated among a small number of parent organizations.

The average Medicare beneficiary is able to choose from Medicare Advantage plans offered by 8 parent organizations in 2026, the same as in 2025 and 2024, and nearly three in ten (29%) can choose among Medicare Advantage plans offered by 10 or more parent organizations.

UnitedHealth Group and Humana account for nearly half of all Medicare Advantage enrollees nationwide in 2026.

Despite most beneficiaries having access to plans operated by several parent organizations, Medicare Advantage enrollment is highly concentrated among a small number of parent organizations (Figure 6). UnitedHealth Group Inc. accounts for 26% of all Medicare Advantage enrollment in 2026 (a decline from 29% in 2025), or 9.3 million enrollees. Humana Inc. accounts for 20% of all Medicare Advantage enrollment in 2026 (an increase from 17% in 2025), or 7 million enrollees. Together, UnitedHealth Group Inc. and Humana Inc. account for nearly half (46%) of all Medicare Advantage enrollees nationwide, the same share as in 2025. In more than a quarter of counties (28% or 889 counties), these two organizations account for at least 75% of Medicare Advantage enrollment. These counties include East Baton Rouge (Baton Rouge), LA (81%), Travis County (Austin), TX (78%), Jackson County (Kansas City), MO (76%), and Palm Beach, Florida (75%).

Three other parent organizations comprise more than 5% of Medicare Advantage enrollment: CVS Health Corporation (12%), Kaiser Foundation Health Plan Inc. (6%), and Elevance Health Inc, (5%). (In contrast, as of 2025, CVS Health and Centene Corporation dominate the market for stand-alone prescription drug plans (PDP) that supplement traditional Medicare, with both organizations accounting for just over half of enrollment in PDPs).

Humana and Kaiser Permanente were the only large insurers to increase enrollment from 2025 to 2026.

In absolute numbers, Humana had the largest growth in enrollment, with 1.3 million more beneficiaries enrolled in a plan sponsored by Humana in March 2026 than in March 2025 (Figure 7). Of the largest insurers (5% of enrollment or more), Kaiser Permanente had the second largest growth in enrollment, with an increase of nearly 87,000 beneficiaries between March 2025 and March 2026. However, some smaller insurers had even larger growth – Devoted Health had an increase of nearly 258,000, SCAN Group had an increase of 136,000, and Aware Integrated had an increase of 89,000 beneficiaries between March 2025 and March 2026 (data not shown).

In contrast, enrollment in UnitedHealth plans declined, decreasing by nearly 647,000 beneficiaries between March 2025 and March 2026, and Elevance had the second largest enrollment drop, losing 346,000 beneficiaries, while enrollment in CVS plans declined by nearly 29,000.

Methods

This analysis uses data from the Centers for Medicare & Medicaid Services (CMS) Medicare Advantage Enrollment, Benefit and Landscape files for the respective year. KFF uses the Medicare Enrollment Dashboard for enrollment data for March 2024-2026, and the CMS Chronic Conditions Data Warehouse Master Beneficiary Summary File (MBSF) for March for earlier years. For other analyses of enrollment in this brief, KFF uses March data; enrollment data is only provided for plan-county combinations that have at least 11 beneficiaries; thus, this analysis excludes approximately 400,000 individuals who reside in a county where county-wide plan enrollment does not meet this threshold. Trend analysis begins in 2007 because that was the earliest year of data that was based on March enrollment. Data from the Medicare enrollment dashboard (used to calculate penetration rates) may change retroactively as CMS updates and reconciles administrative enrollment records. As a result, historical enrollment counts and penetration rates may differ from earlier downloads of the same data.

KFF calculates the share of eligible Medicare beneficiaries enrolled in Medicare Advantage, meaning they must have both Part A and B coverage. In previous years, KFF calculated the share of Medicare beneficiaries enrolled in Medicare Advantage by including Medicare beneficiaries with either Part A and/or B coverage. We modified our approach in 2022 to estimate the share enrolled among beneficiaries eligible for Medicare Advantage who have both Medicare Part A and Medicare B.

Additionally, in previous years, KFF had used the term Medicare Advantage to refer to Medicare Advantage plans as well as other types of private plans, including cost plans, PACE plans, and HCPPs. However, cost plans, PACE plans, and HCPPs are excluded from this analysis in addition to MMPs. In this analysis, KFF excludes these other plans as some may have different enrollment requirements than Medicare Advantage plans (e.g., may be available to beneficiaries with only Part B coverage) and in some cases, may be paid differently than Medicare Advantage plans. These exclusions are reflected in all data displayed trending back to 2007.

Beginning with the analysis of 2025 Medicare Advantage enrollment, KFF has relied on the parent organization field reported to CMS to identify plans sponsored by the same insurer. Previously, KFF had supplemented these data with publicly available information about acquisitions, mergers, and business relationships. The previous approach led to fewer total plan sponsors.

Medicare projections for 2027-2036 are from the February Congressional Budget Office (CBO) Medicare Baseline for 2026. Using the CBO baseline, Medicare enrollment is based on individuals who are enrolled in Part B, which is designed to include only individuals who are eligible for Medicare Advantage and exclude those who only have Part A only (~5.7 million people in 2026) and cannot enroll in Medicare Advantage. However, it may include some individuals who have Part B only and also are not eligible for Medicare Advantage.

Enrollment counts in publications by firms operating in the Medicare Advantage market, such as company financial statements, might differ from KFF estimates due to inclusion or exclusion of certain plan types, such as SNPs or employer group health plans.

Appendix

It has now been over a year since President Trump signed executive orders eliminating federal diversity, equity, inclusion, and accessibility (DEIA) programs and related initiatives, reshaping policies and priorities across federal agencies, research, and education. Additionally, the Trump administration’s workforce reshaping initiatives, including reductions in force, the Deferred Resignation Program, early retirement incentives and a hiring freeze have led to major declines in the federal workforce which have reduced resources available to support efforts to address disparities. While some efforts have been halted or modified by court rulings, the Trump administration has continued to take steps to remove staff, programs, funding, data collection, and other activities tied to concepts of diversity, equity, or disparities. Together, these actions signal a significant shift in federal approaches to addressing inequities and may have broad implications for efforts to monitor and address disparities in health and health care, as well as for the diversity of the workforce, including in health care.

This brief provides an update on the status of the administration’s actions to eliminate DEI-related initiatives one year later and examines emerging impacts and implications for racial health disparities.

Reductions to Federal Agency Staff and Programs

Federal workforce reductions have led to the elimination or scaling back of programs that support data collection, research, and interventions that have historically included a focus on addressing disparities or groups at higher risk of poor health outcomes. Since January 2025, the Trump administration’s workforce reshaping initiatives have led to significant federal workforce declines, with more than 420,000 people separating from the federal workforce as of May 14, 2026. Within the Department of Health and Human Services (HHS), staffing losses have exceeded 20,000 employees since January 2025. At the Centers for Disease Control and Prevention (CDC), an estimated 15% of the workforce, or about 3,000 employees, have departed, with some reports suggesting even larger reductions. With these reductions, many programs that support research and public health interventions appear to have been terminated. Although the full scope of program cutbacks is unclear due to the lack of a comprehensive public record, staffing reductions across agencies have left key programs understaffed or unable to maintain core functions, including efforts focused on addressing health disparities and advancing health equity. For example:

• Reporting suggests that layoffs within the CDC’s Division of Reproductive Health in late March and early April reduced the workforce by about two-thirds, disrupting multiple programs focused on maternal and infant health, an area of health with longstanding disparities. These cuts included the elimination of the Pregnancy Risk Assessment Monitoring System (PRAMS), a primary source of data on health behaviors and outcomes before, during, and after pregnancy that has been widely used to study maternal mortality, including disparities, leaving its future uncertain.
• At HHS, the Office of Climate Change and Health Equity was removed from the agency’s website, and its staff were reportedly placed on administrative leave. Staffing reductions within the CDC’s Division of Environmental Health Science and Practice also led to the elimination of the Environmental Public Health Tracking Program, which monitored data on issues such as cancer clusters and weather-related illnesses. Together, these changes may limit federal capacity to monitor and address environmental and climate-related health disparities, particularly among low-income communities and communities of color that are often disproportionately exposed to environmental hazards and extreme weather events.
• The elimination of the National Survey on Drug Use and Health team halted a key federal data source on substance use and mental health trends, while CDC staff supporting the National Youth Tobacco Survey were also eliminated. Both surveys have historically been used to assess differences in behavioral health and tobacco use across populations and to help guide prevention and treatment efforts for groups disproportionately affected by these issues.

Elimination of Grants and Research Initiatives

Efforts to eliminate DEI across the federal government have reduced support for research related to disparities, including through funding cuts, changes to grant review and award processes, and the loss of leadership and infrastructure that support clinical trials and participation. Federal agencies have reduced or eliminated funding for research initiatives that included DEI-related goals or focused on specific populations, and grant review processes have been revised to flag or exclude applications containing DEI-related terms or focus areas including discrimination, diversity, equity, and race. For example, the recently formed and now defunct Department of Government Efficiency (DOGE) was tasked with auditing and canceling DEI-related federal research grants. Recent deposition hearings indicate that staff members used ChatGPT to propose cuts to roughly 1,400 grants under the National Endowment for the Humanities, including examples that were not related to DEI. In addition to the grants reviewed under DOGE, broader administrative actions led to the termination of more than 2,300 National Institute of Health (NIH) grants by late June 2025, with nearly 1,100 grants remaining terminated as of May 4, 2026. One study found that the National Institute of Minority Health and Health Disparities lost both the largest share of grants and the largest share of funding across all NIH institutes and centers. Research shows that grant terminations were more likely to impact American Indian or Alaskan Native (AIAN), Asian, Black, Hispanic or Native Hawaiian or Pacific Islander (NHPI) researchers compared to White researchers. These actions also affected 160 NIH-funded clinical trials, more than half of which (57%) included project terms related to racial and ethnic minority populations, such as Black, Latino, Indigenous, Asian, and other historically underserved groups. At the same time, the loss of research leadership and staff has disrupted the research pipeline and contributed to additional grant terminations. HIV research has been particularly impacted. Reporting indicates that at least 145 NIH-funded HIV research grants, totaling nearly $450 million, were terminated in early 2025, including studies focused on HIV prevention, access to pre-exposure prophylaxis (PrEP), and populations disproportionately affected by HIV, which include groups of color. A newly proposed rule issued by the Office of Management and Budget (OMB) would revise federal grant requirements by increasing political review of awards, requiring alignment with presidential priorities, and creating new mechanisms for modifying or terminating existing funding, with an emphasis on prohibiting federal support for programs, funding preferences, or award requirements that advance DEI efforts. If finalized, the proposal could expand federal grantmaking agencies’ authority to review and terminate awards deemed inconsistent with the administration’s interpretation of civil rights laws and federal priorities, which may affect research and programs designed to address racial, ethnic, gender, and other disparities.

Elimination of Public Information and Resources

Efforts to eliminate DEI across the federal government have also reduced the availability and integrity of public information and data, including through the suspension of national surveys, the removal of DEI-related data elements, and changes to climate data and evidence. As previously noted, several national surveys were suspended, delayed, or scaled back, limiting the availability of timely population-level data. These include the Pregnancy Risk Assessment Monitoring System (PRAMS) and the National Survey on Drug Use and Health (NSDUH). There are also reports that the National Intimate Partner and Sexual Violence Survey and the National Youth Tobacco Survey (NYTS) may have also been eliminated, however, their official status remains unclear. These surveys have historically provided data used to identify disparities and inform public health policy and interventions. In addition, key surveys and data systems removed or modified questions and data elements related to race, ethnicity, gender identity, sexual orientation, and other demographic measures, reducing the ability to measure differences across populations. Some publicly available datasets were also modified or taken offline, limiting access for researchers, policymakers, and the public, while some federal websites removed or archived reports, dashboards, and tools that previously highlighted disparities or equity-focused analyses. Climate-related data and resources were similarly removed, reframed and misrepresented, or made more difficult to access. For example, the Trump administration determined that official government websites would no longer host national climate assessments, including the most recent assessment released in 2023, which found that climate change disproportionately affects the health, livelihoods, and security of people of color, with Indigenous populations at particular risk. In addition, concerns have been raised about the scientific rigor and peer review process of the Climate Working Group report recently released under the administration.

Education and Workforce

Amid the executive orders, threats by the Trump administration to withdraw funding from schools with DEI programs, and charges from the Equal Employment Opportunity Commission (EEOC) for discrimination against White people, concerns have grown about the potential impacts on diversity among students and the future workforce. Early last year, the Department of Education issued guidance, commonly referred to as the “Dear Colleague” letter, directing schools and other entities receiving federal education funding to stop using what it described as “racial preferences” in admissions, programming, and other activities. This followed the Supreme Court’s 2023 decision ending race-conscious admissions, which led to a shift in enrollment patterns across higher education institutions, with declines in the share of students of color at highly selective universities. Research also suggests that the ruling has already contributed to declines in the number of Black, Hispanic, and AIAN students entering medical school, raising concerns about the future diversity of the physician workforce. These trends may further exacerbate existing disparities in representation. KFF analysis shows that Hispanic, Black, AIAN, and NHPI people remain underrepresented among physicians relative to their share of the population, with the largest gap among Hispanic people, who comprise 20% of the U.S. population but only 7% of the physician workforce. More broadly, the EEOC has increasingly emphasized enforcement related to alleged discrimination against White workers, which some argue could discourage workplace diversity initiatives. In addition, broader immigration restrictions and enforcement actions may further affect workforce diversity, particularly in health care and research fields that rely heavily on immigrant workers and internationally trained professionals.

Legal Challenges and Protections

In some cases, implementation of DEI-related cuts has been temporarily halted or mitigated by lawsuits and congressional action. As of April 2026, Congress has rejected many of the Trump administration’s proposed reductions to federal health programs through the FY 2026 appropriations process. For example:

• The legislation provided HHS with approximately $116 billion in funding, about $33 billion more than proposed in President Trump’s FY 2026 budget request, including increases for several public health and behavioral health programs, such as reproductive and community health programs, that support populations disproportionately affected by poor health outcomes, including people of color and lower-income groups.
• It also maintained the Substance Abuse and Mental Health Services Administration (SAMHSA) as an independent agency and increased SAMHSA funding by $65 million to approximately $7.4 billion, despite earlier proposals to reduce funding by roughly $1 billion. The legislation also included guardrails intended to ensure that SAMHSA funds are distributed as appropriated, preserving support for mental health and substance use programs that often serve low-income and underserved communities.
• Similarly, Congress maintained CDC funding at approximately $9.2 billion, rather than the nearly 50% reduction proposed in the administration’s FY 2026 budget request, and directed HHS to maintain staffing levels needed to carry out CDC programs, although the agency’s Social Determinants of Health program was eliminated.
• The legislation also preserved funding for reproductive health programs, including Title X and the Teen Pregnancy Prevention Program, and increased investments in maternal health programs across HRSA, CDC, and NIH, which have historically focused on addressing longstanding racial disparities in maternal and infant health outcomes.
• In addition, while the administration proposed substantial reductions to NIH funding, Congress ultimately approved increased funding for the agency, and several court challenges temporarily halted some funding cuts and grant terminations, which had a noted focus on DEI-related research.

Separate legal challenges have also limited implementation of some workforce and education-related actions. Federal courts and Congress blocked or delayed aspects of the Trump administration’s workforce reduction efforts, including rulings against certain reductions in force and layoffs implemented during agency restructuring efforts. Courts also blocked implementation of the Department of Education’s “Dear Colleague” guidance related to race-conscious practices in schools, and the Department later withdrew the guidance following legal challenges and court rulings. These actions suggest that the scope and long-term effects of the administration’s policies may continue to evolve as litigation and congressional oversight proceed.

This fact sheet summarizes the major statutory requirements and policies pertaining to U.S. global family planning/reproductive health (FP/RH) efforts over time and identifies those currently in effect. These laws and policies collectively serve to direct how U.S. funds are spent, to where and which organizations funds are provided, and generally shape the implementation and define the scope of U.S. global FP/RH activities. It includes U.S. laws and annual requirements enacted by Congress through appropriations bills (statutory provisions) as well as executive branch policies and guidance specific to FP/RH (policy provisions). Each category lists provisions in chronological order.

Since early 2025, the Trump administration has frozen or eliminated funding for most FP/RH activities, although Congress has continued to appropriate funding for this work. See the KFF fact sheet of the status of FP/RH efforts.

Table 1

Statutory Requirements and Policies for U.S. Global FP/RH Efforts (as of FY 2026)

1. As noted in CRS, Abortion and Family Planning-Related Provisions in U.S. Foreign Assistance Legislation and Policy, July 2022: “No restrictions exist on funding for the medical evacuation of Peace Corps volunteers who decide to have an abortion. Under existing policy, the Peace Corps covers the cost of evacuation to a location where ‘medically adequate facilities’ for obtaining an abortion are available and where abortions are legally permissible.” ↩︎
2. In most recent years, a provision is included requiring that any Kemp-Kasten determination that is made must be accompanied by the evidence and criteria used to make the determination. ↩︎
3. USAID defines a target/quota as “a predetermined figure that a service provider or referral agent is assigned or required to affect or achieve” for the purposes of the Tiahrt Amendment. It states that “the key to interpreting ‘incentives’ is to see whether they are provided in exchange for accepting a method (in the case of a client) or linked to achievement of a predetermined target or quota (in the case of program personnel).” USAID Global Health eLearning Center, “FP Legislative & Policy Requirements (Updated),” online course, February 2009, authored by Debbie Gueye, MSI. ↩︎
4. USAID Global Health eLearning Center, “FP Legislative & Policy Requirements (Updated),” online course, Feb. 2009, authored by Debbie Gueye, MSI. ↩︎
5. Although such reallocation began in practice in FY 2002, it was first authorized by Congress in legislation beginning in FY 2004 with reference to FY 2002 and FY 2003 funds. ↩︎
6. Informed Choice: Effective access to information on family planning choices and to the counseling, services, and supplies needed to help individuals choose to obtain or decline services; to seek, obtain, and follow up on a referral; or simply to consider the matter further. Voluntarism: Decision to use a specific method of family planning or to use any method of family planning is based upon the exercise of free choice and is not obtained by any special inducements or any element of force, fraud, deceit, duress or other forms of coercion or misrepresentation. USAID Global Health eLearning Center, “FP Legislative & Policy Requirements (Updated),” online course, Feb. 2009, authored by Debbie Gueye, MSI. ↩︎
7. This policy was first instituted via presidential memorandum in 1984 by President Reagan. In 1993, it was rescinded by President Clinton, although it was briefly applied legislatively in 1999 (see “Status” column). In 2001, it was reinstated by President Bush, who expanded its applicability in 2003 to include family planning funds at the State Department (see “Applies to” column) with some exemptions. In 2009, it was rescinded by President Obama. In 2017, it was reinstated by President Trump, who expanded its applicability to include the vast majority of global health assistance furnished by all departments and agencies. In 2021, it was rescinded by President Biden. ↩︎
8. Note that, with one exception, has been applied via Executive action. The exception was in FY 2000, when President Clinton agreed to a one-year legislative codification with a partial waiver of restrictions as part of a broader arrangement to pay the U.S. debt to the United Nations. See P.L. 106-113, Sec. 599D, and PAI, Global Gag Rule Timeline, July 12, 2011. ↩︎
9. “M16. VOLUNTARY POPULATION PLANNING ACTIVITIES – MANDATORY REQUIREMENTS (MAY 2006) …
   b. Prohibition on Abortion-Related Activities:
   (1) No funds made available under this award will be used to finance, support, or be attributed to the following activities: (i) procurement or distribution of equipment intended to be used for the purpose of inducing abortions as a method of family planning; (ii) special fees or incentives to any person to coerce or motivate them to have abortions; (iii) payments to persons to perform abortions or to solicit persons to undergo abortions; (iv) information, education, training, or communication programs that seek to promote abortion as a method of family planning; and (v) lobbying for or against abortion. The term “motivate,” as it relates to family planning assistance, must not be construed to prohibit the provision, consistent with local law, of information or counseling about all pregnancy options.
   (2)No funds made available under this award will be used to pay for any biomedical research which relates, in whole or in part, to methods of, or the performance of, abortions or involuntary sterilizations as a means of family planning. Epidemiologic or descriptive research to assess the incidence, extent, or consequences of abortions is not precluded.” ↩︎
10. For example, one provision related to abortion included in the State Department Standard Terms and Conditions is:
    “AA. Prohibition on use of funds for performance or research respecting abortions or involuntary sterilization
    The recipient agrees that in accordance with 22 USC 2151b(f) Population planning and health programs no foreign assistance funds provided by the award shall be used to:
    (1) pay for the performance of abortions as a method of family planning or to motivate or coerce any person to practice abortions (Helms Amendment, 1973).
    (2) pay for the performance of involuntary sterilizations as a method of family planning or to coerce or provide any financial incentive to any person to undergo sterilizations (Involuntary Sterilization Amendment, 1978).
    (3) pay for any biomedical research which relates, in whole or in part, to methods of, or the performance of, abortions or involuntary sterilization as a means of family planning (Biden Amendment, 1981).
    Furthermore, the recipient agrees in accordance with the Department’s annual appropriation bill, that no funds provided by the award may be used to lobby for or against abortion (Siljander Amendment, 1981).” ↩︎

On April 30, 2026, North Carolina Governor Josh Stein signed legislation that includes Medicaid policy changes and closes an estimated $319 million shortfall in funding for the state’s Medicaid program for FY 2026. Many of the legislation’s Medicaid policy changes are related to implementation of the 2025 federal reconciliation law. The 2025 reconciliation law requires states to condition Medicaid eligibility for adults in the Affordable Care Act (ACA) Medicaid expansion group and enrollees in partial expansion waiver programs (Georgia and Wisconsin) on meeting work requirements starting January 1, 2027. The 2025 reconciliation law also limits states’ ability to raise the state share of Medicaid spending through provider taxes, restricts state-directed payments (SDPs) to hospitals, nursing facilities, and other providers, and increases barriers to enrolling in and renewing Medicaid coverage. As states are preparing to implement the reconciliation law provisions, many states are facing more tenuous budget situations with slowing revenue growth and broader reductions in federal funding.

This policy watch examines the current budget context in North Carolina, the state’s recently passed legislation, the state’s Medicaid Advisory Committee (MAC) meetings, and data from KFF’s Medicaid work requirements tracker to provide initial insight into how North Carolina is preparing to implement certain Medicaid provisions of the 2025 reconciliation law and how other policy changes may affect coverage and access to care. While some of the issues North Carolina is facing are unique to that state, others are likely to be faced by other states as they implement federal changes to Medicaid in the midst of other fiscal challenges.

What is the budgetary context as North Carolina prepares to implement the 2025 reconciliation law’s provisions?

North Carolina is facing a more tenuous fiscal climate like in other states, and state legislators have not yet enacted a comprehensive state budget for the FY 2025-27 biennium. In the past year, revenue volatility and rising costs have led to slowing state revenue growth following a period of record-breaking revenue and expenditure growth for states after the initial pandemic-induced economic downturn. In North Carolina, scheduled tax cuts have been projected to drive declines in state revenue, and debates over whether to proceed with the cuts have contributed to a budget stalemate in the legislature.

In August 2025, Governor Stein signed a stopgap funding bill  that appropriated $600 million from the state general fund for Medicaid, but it left a $319 million shortfall for FY 2026 in funding for the cost of services for non-expansion (traditional) enrollees. The shortfall and budget stalemate led to rate cuts and the elimination of GLP-1 coverage, both of which were eventually restored. The Medicaid agency ceased “Healthy Opportunities Pilots” program services in FY 2026 due to a lack of appropriations. The pilots covered certain non-medical services that target social needs, including housing, nutrition, transportation, and interpersonal relationship supports to specific and limited enrollees, and evaluations of the “Healthy Opportunity Pilots” 1115 waiver showed lower costs over time and largely positive outcomes. The Medicaid agency also implemented changes to reduce administrative expenses, including reducing temporary staff and contractors, ending certain contracts, pausing quality improvement projects, and scaling back compliance and quality oversight activities. The legislation signed in April 2026 appropriated $319 million to close the shortfall for FY 2026 and made changes aimed at addressing financing pressures associated with new federal limits on provider taxes, which the state uses to help finance its Medicaid expansion and hospital state directed payment program (which increases payment rates for hospitals).

What are some of the Medicaid policy changes included in North Carolina’s recent legislation?

Eligibility and Cost Sharing

North Carolina’s new legislation includes more restrictive standards for how the state will implement work requirements than is required in current law. At a minimum, the 2025 reconciliation law requires states to look back one month immediately preceding the application month and one month between renewal periods to confirm compliance with the requirements. North Carolina’s legislation requires the state to confirm compliance for the three months preceding the application month. At renewal, the state must confirm compliance for at least three of the six months since the last determination of eligibility. The North Carolina legislation also prohibits the acceptance of self-attestation as the only evidence in verification of eligibility requirements (unless required by federal law or regulation, or pursuant to a court order). States await guidance from CMS as to whether self-attestation of medical frailty, parent/caretaker status, or other exemptions or work status can be accepted, but most states report plans to accept self-attestation if allowed. 

The legislation increases the frequency of data checks to identify changes in circumstances for Medicaid enrollees from quarterly to monthly. The state will review information on earned and unearned income, employment status and changes in employment, residency status, enrollment status for other public assistance programs administered by the state and outside of the state, financial resources, incarceration status, and lottery and gambling winnings. States are required to follow up on reported changes that potentially affect eligibility and give individuals an opportunity to respond before taking adverse action. In North Carolina, when data indicates an individual is no longer eligible, enrollees only have 10 days in advance of case closure to submit documentation verifying ongoing eligibility. Increasing the frequency of periodic data checks with insufficient response times can lead to procedural disenrollments and exacerbate churn.

The Medicaid agency will be required to set Medicaid copayments at the highest allowable amounts for both traditional Medicaid enrollees and ACA expansion enrollees. Current federal rules limit cost sharing in Medicaid because of enrollees’ low income and limited ability to pay out of pocket costs. The maximum allowable cost sharing varies by type of service and enrollee income. North Carolina has set current cost sharing amounts, regardless of enrollee income, at $4 per service. Starting July 1, 2027, the legislation requires the Medicaid agency to increase current cost sharing amounts for services where the maximum allowable amount is more than $4 and to increase cost sharing for ACA expansion adults with income 100-138% FPL to up to 10% of the cost of the service, except for prescription drugs and non-emergency use of the emergency department. Beginning in October 2028, when states are required to implement mandatory cost sharing of up to $35 per service for ACA expansion adults with income between 100%-138% FPL, the state will be required to set cost sharing amounts at $35 per service, except for prescription drugs, for all non-exempt services for this group. 

The legislative text implementing the changes to Medicaid eligibility for certain lawfully residing immigrants effectively ends the state’s long-standing optional Medicaid coverage for lawfully residing children and pregnant immigrants without a five-year waiting period. The law limits Medicaid coverage for immigrants to coverage that is required under federal law. However, North Carolina is one of 40 states that have taken up the option to extend Medicaid and/or CHIP coverage to children and/or pregnant adults who are lawfully residing and waive the five-year wait for these groups. The 2025 reconciliation law imposed additional eligibility restrictions for many lawfully present immigrants but allows states to maintain the option to cover lawfully residing children and pregnant adults. By limiting coverage for immigrants to only what is required by federal law, the state law effectively ends this optional coverage as of October 1, 2026. In a recent Medicaid Advisory Committee (MAC) meeting, the Medicaid agency indicated it was working with the legislature to make “corrections” and restore coverage for these populations.

The legislation requires the Medicaid agency to report certain Medicaid applicants and enrollees for whom it cannot verify citizenship or “satisfactory” immigration status to the Department of Homeland Security. These include applicants and enrollees who, after a reasonable opportunity period, have not verified satisfactory immigration status or whose final verification indicates that they do not have a satisfactory immigration status and are not lawfully present. This group would include those found ineligible based on immigration status and individuals receiving Emergency Medicaid services (where Medicaid pays hospitals for emergency care provided to ineligible immigrants who would otherwise be eligible for Medicaid based on their income).

Medicaid Financing

The legislation increases intergovernmental transfers (IGTs) from public hospitals, reducing reliance on the state’s hospital taxes for financing the nonfederal share of Medicaid spending. The 2025 reconciliation law imposes significant new restrictions on states’ ability to generate Medicaid provider tax revenue, including prohibiting all states from establishing new provider taxes or from increasing existing taxes, as well as reducing existing provider taxes for states that have adopted the ACA Medicaid expansion. North Carolina uses provider taxes to help finance the nonfederal share of Medicaid spending. State law requires the nonfederal share for the expansion program to be fully funded by certain non-general fund sources, including hospital taxes and hospital IGTs, and requires the end of expansion coverage if those sources cannot fully fund the nonfederal share. The state estimates $14.3 million in one-time administrative costs for the current state fiscal year and $44.4 million in recurring annual administrative costs (including both state and county expenditures) for the expansion program to implement work requirements and six-month eligibility redeterminations that existing financing mechanisms did not account for. The increased public hospital IGTs aim to offset the financing of some of the existing costs under the state’s hospital tax, as well as to help finance the new administrative costs.

By increasing reliance on IGTs as a financing source, the state may aim to retain higher hospital SDPs under new federal provider tax limits, but new federal requirements for state directed payments are expected to require further changes. North Carolina’s Healthcare Access and Stabilization Program (HASP), a hospital SDP program launched alongside Medicaid expansion in 2023, is also financed through hospital taxes and IGTs. An earlier state report indicated the new federal provider tax limits would eliminate all or most of HASP SDPs. The state has been using HASP payments to incentivize hospitals to relieve medical debt, and as of October 2025, more than $6.5 billion in debt had been relieved for more than 2.5 million North Carolinians under the initiative.

There is significant uncertainty about how federal regulations and state legislation may affect the state’s plan for financing the nonfederal share of Medicaid spending, including for the Medicaid expansion and HASP. New proposed rules on state directed payments and forthcoming provider taxes may affect the state’s financing plans. The state’s legislation created a “trigger” to end the new funding should HASP payments fall below certain thresholds or a change in federal law or regulation result in at least a 20% decrease to IGTs.

For coverage of Medicare benefits, people face a choice between traditional Medicare and private Medicare Advantage plans. While there are many distinguishing features between these coverage types, one key benefit of Medicare Advantage is an annual cap on out-of-pocket costs for medical benefits, which traditional Medicare does not include. In 2026, the out-of-pocket limit for Medicare Advantage plans may not exceed $9,250 for in-network services and $13,900 for a combination of in-network and out-of-network services, but plans can have lower out-of-pocket limits than the maximum.

Policymakers have long considered whether to add a similar out-of-pocket cap in traditional Medicare. In 1988, Congress enacted a Medicare out-of-pocket cap, but the law was repealed one year later principally due to concerns about the financing mechanism. The Medicare Payment Advisory Commission (MedPAC) has recommended changes in traditional Medicare to give beneficiaries better protection against high out-of-pocket spending, including an out-of-pocket cap as part of a broader redesign that also includes combining Part A and Part B deductibles and simplifying cost sharing. Traditional Medicare remains the only major form of health insurance that does not include a cap on out-of-pocket spending, though most beneficiaries in traditional Medicare have additional financial protection through supplemental coverage, including Medicaid, employer- or union-sponsored retiree coverage, or a Medigap policy (which may require additional premiums).

This brief analyzes out-of-pocket limits in Medicare Advantage plans in 2026, variation by plan type, the distribution of enrollees facing different out-of-pocket limits, and trends over time. Of note, this analysis does not show the share of enrollees that reach their plan’s out-of-pocket limit in any year because spending data are not available. The analysis includes Medicare Advantage plans generally available for individual enrollment, reflecting coverage for 21.3 million Medicare beneficiaries in 2026, excluding special needs plans and employer-and union-sponsored group plans (See Methods for details).

Takeaways

• In 2026, the average enrollment-weighted out-of-pocket limit for Medicare Advantage enrollees is $5,421 for in-network services and $9,825 for in-network and out-of-network services combined. The average out-of-pocket limit for in-network services is higher for preferred provider organizations (PPOs) ($6,592) than health maintenance organizations (HMOs) ($4,636).
• Just over one in ten Medicare Advantage enrollees (13% or 2.8 million people) are in plans with limits of $3,000 or less for in-network services in 2026, more than two-thirds (68%) are in plans with limits above $3,000 and up to $7,000, and about one in five (19%) are in plans with limits above $7,000.
• About one in 10 (9%; 1.8 million) Medicare Advantage enrollees are in plans with the maximum out-of-pocket limit for in-network services ($9,250). Among the 8.6 million enrollees in PPOs, roughly one in five (22%; 1.8 million) are in plans with the maximum out-of-pocket limit for in-and out-of-network services combined ($13,900).

• The average out-of-pocket limit for in-network services decreased by nearly $600 from 2017 to 2023 ($5,253 to $4,685), before increasing by about $700 from 2023 to 2026 ($4,685 to $5,421).

In 2026, the Average Enrollment-Weighted Out-of-Pocket Limit for Medicare Advantage Enrollees Is Lower for HMOs Than PPOs

The average out-of-pocket limit faced by Medicare Advantage enrollees in 2026 is $5,421 for in-network services and $9,825 for in-network and out-of-network services combined (Figure 1). These averages are lower than the maximum allowable out-of-pocket limits ($9,250 and $13,900, respectively).

Differences in out-of-pocket limits may reflect how Medicare Advantage plans choose to allocate rebate dollars, which are extra payments they receive from the federal government beyond the cost of providing Part A and Part B services. Plans may use rebates to reduce beneficiary cost sharing, including lowering out-of-pocket limits, but they can also use these funds for other purposes, such as offering supplemental benefits not covered by traditional Medicare, offering a rebate against the Part B premium, or lowering Part D premiums.

Out-of-pocket limits for in-network services vary by plan type. HMOs, which have 12.8 million Medicare Advantage enrollees in 2026, generally offer no coverage of services from out-of-network providers, but offer a lower out-of-pocket limit for in-network services than PPOs. PPOs, which have 8.6 million enrollees in 2026, cover services delivered by in-network and out-of-network providers, but require higher cost sharing for out-of-network providers. While PPO enrollees have broader access to out-of-network providers than HMO enrollees, they also face a higher out-of-pocket limit, even for in-network services. Specifically, the average enrollment-weighted out-of-pocket limit for in-network services is $6,592 for PPOs and $4,636 for HMOs.

The distribution of enrollment between HMOs and PPOs, as well as the average out-of-pocket caps for in-network services by type of plan, vary across states (Appendix Table 1). Due in part to PPOs (which have higher average in-network out-of-pocket limits than HMOs) comprising a larger share of enrollment in rural areas, the average out-of-pocket cap is about $800 higher for Medicare Advantage enrollees in rural areas than in urban areas ($6,078 vs $5,291).

In 2026, Roughly One in Ten Medicare Advantage Enrollees Are in Plans With Limits of $3,000 or Less for In-Network Services and About One in Five Are in Plans With Limits Above $7,000

Among all Medicare Advantage enrollees in individual plans, just over one in ten (13%, 2.8 million) are in plans with out-of-pocket limits of $3,000 or less for in-network services (Figure 2). Nearly all of these enrollees (99%) are in an HMO. More than two-thirds (68%, 14.4 million) of Medicare Advantage enrollees are in plans with out-of-pocket limits above $3,000 and up to $7,000. About one in five (19%, 4.1 million) are in plans with limits above $7,000, including 1.8 million enrollees who are in a plan with the maximum in-network out-of-pocket limit of $9,250. PPOs enrollees account for the majority of these enrollees at the maximum cap (66%).

Among the 8.6 million Medicare Advantage enrollees in PPOs, about one in ten (9%) have a combined limit for in-network and out-of-network services at or below $6,000. More than two-thirds (66%; 5.7 million) of enrollees are in plans with a combined limit for in-network and out-of-network services between $6,000 and $12,000. One quarter (25%; 2.1 million) are in plans with an out-of-pocket limit for in-and out-of-network services combined above $12,000, including 1.8 million enrollees in plans with the maximum out-of-pocket limit for in-and out-of-network services combined ($13,900) (Figure 2).

Average Out-of-Pocket Limits for In-Network Services Generally Declined Between 2017 and 2023 but Have Increased Since Then

The average limit for in-network services decreased by nearly $600 from 2017 ($5,253) to 2023 ($4,685), before increasing by about $700 from 2023 to 2026 ($5,421) (Figure 3, Appendix Table 2). The average limit for in-network and out-of-network services combined has fluctuated over time but increased overall by about $750 between 2017 ($9,073) and 2026 ($9,825).

The maximum allowable out-of-pocket cap has generally increased over time, consistent with projected beneficiary out-of-pocket spending in traditional Medicare, which CMS uses to calculate maximum out-of-pocket limits in Medicare Advantage. However, the maximum cap decreased by $100 between 2025 and 2026. The average out-of-pocket limit faced by enrollees each year has always been lower than the maximum allowable limit. This gap has generally widened over time, but shrank somewhat between 2025 and 2026.

Appendix Tables

Health care costs and affordability recently topped the public’s list of economic anxieties in the U.S., according to KFF polling. Even people with insurance say they struggle to afford health care costs, including people with Medicare. While Medicare provides health insurance coverage to 70 million people age 65 or older and younger adults with long-term disabilities, having Medicare coverage does not insulate beneficiaries from health-related affordability challenges, such as not getting needed medical care due to costs or incurring medical debt. In a 2026 KFF Health Tracking Poll, half (49%) of all Medicare beneficiaries ages 65 and older say they expect their health care costs to become less affordable in the next year.

This brief presents key facts and analysis about affordability of health care costs among people with Medicare, including younger adults with long-term disabilities, drawing on data from various sources (see methods for additional information).

Many Medicare Beneficiaries Have Low Incomes and Modest Savings, Which Can Make It Hard To Afford Medical and Long-Term Care Services

• One in four Medicare beneficiaries (16.5 million people) had income below $24,600 per person in 2024, or about 160% of the federal poverty level that year (Figure 1). These estimates take into account income from Social Security, pensions, retirement account (IRA) withdrawals, and other sources. At the upper end of the income spectrum, the top 5% of Medicare beneficiaries (3.3 million people) had incomes above $169,700 per person. Many beneficiaries also have relatively low levels of savings, with one in four Medicare beneficiaries having savings below $18,950 per person in 2024. Financial resources are even lower among some subgroups of beneficiaries, including Black and Hispanic beneficiaries. For example, one in four Black beneficiaries had income below $20,150 per person in 2024, while one in four Hispanic beneficiaries had income below $14,150 per person.

• About 6 million people ages 65 and older (10%) were living in poverty in 2024 under the official poverty measure, meaning they had income of $15,050 or less that year (if single), while 17.3 million (28%) had incomes below 200% of poverty. Poverty rates among older adults are higher under an alternative measure (the supplemental poverty measure, or SPM) that accounts for certain expenses that are higher among older people, including out-of-pocket medical expenses, with 15% of older adults living in poverty in 2024 under the SPM.
• In 2025, nearly a quarter (23%) of all Medicare beneficiaries who received Social Security benefits relied on Social Security income for 90% or more of their total per capita income, while about one-third (32%) relied on Social Security for at least 75% of their income (Figure 2). In 2025, the average per capita Social Security income was $20,168 among those who depended on Social Security for at least 90% of their income and $20,670 among those who depended on Social Security for at least 75% of their income.

Medicare Premiums and Other Health Expenses Consume a Sizeable Portion of Income and Household Budgets for People With Medicare

• Out-of-pocket health care spending for Medicare premiums and out-of-pocket costs for health care services accounted for 36% of Medicare beneficiaries’ average Social Security income per person in 2023 (Figure 3). In 2023, Medicare beneficiaries spent an average of $6,459 out of pocket on health care costs and had average Social Security income of $17,718 (though most older adults have other sources of income beyond Social Security that can be used to pay their out-of-pocket health care costs). Out-of-pocket spending includes premiums for Medicare supplemental coverage (Medigap) for those who enroll in traditional Medicare and buy private supplemental insurance.

• Health care spending accounted for 14% of Medicare beneficiaries’ total household spending in 2024. Medicare households spent a larger share of their total household spending on health care than non-Medicare households in 2024 (6%) (Figure 4).

• More than 7 million Medicare beneficiaries enrolled in Medicare Part B in 2024 spent more than 10% of their annual per capita income on Part B premiums alone. The standard Part B premium is paid by beneficiaries in both traditional Medicare and Medicare Advantage and has roughly doubled in the last decade, from an annual amount of $1,259 in 2015 to $2,435 in 2026.

The Choice Between Traditional Medicare and Medicare Advantage Has Cost Implications for Beneficiaries

• Out-of-pocket spending for Medicare-covered hospital and physician services is not capped in traditional Medicare, while Medicare Advantage plans are required by law to provide an out-of-pocket limit for these services. In 2025, the average out-of-pocket limit for Medicare Advantage enrollees was $5,320 for in-network services and $9,547 for in-network and out-of-network services combined. The maximum allowed limits were $9,350 for in-network services and $14,000 for in-network and out-of-network services combined in 2025, and increased to $9,250 and $13,900, respectively, in 2026. (Figure 5). This cap protects Medicare Advantage enrollees from unlimited health care costs for Medicare-covered services, but cost-sharing requirements can lead to high out-of-pocket expenses before reaching a plan’s annual limit. People enrolled in Medicare Advantage plans typically pay no additional premium.

• Medicare Advantage plans can reduce cost sharing, offer benefits not covered under traditional Medicare (such as dental and vision services), and lower Part B and Part D premiums, using extra payments they receive from the federal government beyond the cost of providing Part A and Part B services. In 2026, these extra payments (known as rebates) amounted to about $2,660 per person on average in 2026. For Medicare-covered services under Part A and Part B, Medicare Advantage plans can charge different amounts than what someone in traditional Medicare would pay, though they often reduce cost sharing for Medicare-covered benefits, and they are prohibited from charging more for certain services, such as Skilled Nursing Facility (SNF) stays. Plans are required to provide coverage that is at least as generous overall as traditional Medicare, even though what enrollees pay out of pocket can vary depending on the type of plan, the type and quantity of services they use, and whether they see a provider that participates in the plan’s network.

• Medigap limits traditional Medicare beneficiaries’ exposure to out-of-pocket medical costs, and the most popular Medigap policies cover virtually all cost sharing for Medicare-covered services. However, Medigap underwriting rules make these policies difficult, if not impossible, for beneficiaries with pre-existing conditions to obtain outside of limited enrollment opportunities. For beneficiaries with modest incomes, the monthly Medigap premium ($217 per month, on average, in 2023) could be unaffordable (Figure 6).

• More than 3 million Medicare beneficiaries in traditional Medicare have no additional coverage that helps with Medicare cost-sharing requirements, such as Medigap, employer coverage, or Medicaid. This leaves them at risk of facing high out-of-pocket costs if they need a lot of medical services, or high-cost services, because there is no limit on out-of-pocket spending for Parts A and B in traditional Medicare.

Paying for Health Care, Including for Services Not Covered by Medicare, Can Be a Financial Burden

• Medicare beneficiaries can face high out-of-pocket costs for dental, hearing, and vision care services, which are not covered under traditional Medicare (Figure 7). In 2023, traditional Medicare beneficiaries spent $1,107 on dental services and $564 on hearing services, on average (Figure 7). While most Medicare Advantage plans cover dental, hearing, and vision benefits, Medicare Advantage enrollees also face out-of-pocket costs for these services, spending an average of $571 on dental services and $212 on hearing services.

• About one in five (22%) Medicare-age adults reported having some type of debt as a result of medical or dental bills in 2022 (Figure 8). Among older adults with medical debt, about 4 in 10 say they cut back on other household spending and used up or all or most of their savings as a result of their health care debt.

• Long-term services and supports, which are not covered by Medicare, are unaffordable for all but the highest-income Medicare beneficiaries. In 2025, the median annual costs of common long-term services and supports in the U.S were $80,080 for full-time non-medical caregiver services (including home health aide and homemaker services for people who need help with instrumental activities of daily living such as preparing meals and doing laundry), $129,575 for a private room in a nursing home, and $305,760 for round-the-clock home health aide services (Figure 9). These costs greatly surpass median income ($43,200) and savings ($110,100) among people with Medicare. Long-term care services are generally not covered by Medicare, regardless of whether a beneficiary is enrolled in traditional Medicare or Medicare Advantage.

Medicaid Helps Make Medicare Affordable for About 12 Million Low-Income Medicare Beneficiaries

• In 2025, about 12 million people with Medicare also had Medicaid coverage (“dual-eligible individuals”).
  • More than seven in ten (72%) dual-eligible individuals, or 8.5 million people, are eligible for the full range of Medicaid benefits not otherwise covered by Medicare, including long-term services and supports, vision and dental services, and non-emergency medical transportation (also known as wraparound services) (Figure 10).These “full-benefit” dual-eligible individuals also usually receive additional help through the Medicare Savings Programs (MSPs), which cover Medicare premiums, and in most cases, cost sharing for people with limited income and assets.
  • The remaining 3.4 million dual-eligible individuals (“partial-benefit” individuals) do not receive coverage of the full range of Medicaid benefits, but do receive payments to cover Medicare premiums, and, in most cases, cost sharing through the Medicare Savings Programs. Under federal guidelines, Medicare beneficiaries with income up to 135% of the federal poverty line and assets below specified levels ($9,950 for an individual and $14,910 for a couple in 2026) qualify for financial assistance from their state Medicaid program through a Medicare Savings Program. States can choose to cover Medicare beneficiaries with income or assets above the federal limits, and in 2026, 18 states do so. Without financial support from Medicaid, the Part B premium alone would consume 15% of income for a dual-eligible individual with monthly income of $1,350, not including other out-of-pocket costs.

Medicare Part D Offers Protection Against High Out-of-Pocket Prescription Drug Costs, but Prescription Drug Costs Remain a Concern for Older Adults

• Virtually all Medicare beneficiaries (96%) use prescription drugs and most are enrolled in Medicare Part D drug plans for prescription drug coverage. On average, Medicare Part D enrollees who did not receive low-income subsidies spent an average of $872 on prescription drugs in 2023, and people in fair or poor health spent roughly twice as much, on average ($1,671) (Figure 11). The Part D low-income subsidy (LIS) provides substantial financial support to income below 150% of poverty (under $23,940 for an individual in 2026) and limited assets (under $18,090 for an individual). Part D enrollees with LIS had average out of pocket drug costs of $156 in 2023, roughly one fifth the amount paid by those without these low-income subsidies.

• Under changes made by the Inflation Reduction Act, Part D enrollees have a cap on out-of-pocket drug spending equal to $2,100 in 2026. Prior to this change, enrollees could have faced several thousand dollars in out-of-pocket costs for expensive medications each year. The Inflation Reduction Act also includes other changes that make prescription drugs more affordable for people with Medicare, including a $35 monthly cap on insulin, expanded eligibility for full benefits under the Part D Low-Income Subsidy, and $0 cost sharing for adult vaccines under Medicare Part D. Additionally, the law requires the federal government to negotiate lower drug prices for some high-cost drugs covered under Medicare through the Medicare Drug Price Negotiation Program.
• Even with a cap on overall out-of-pocket costs, prescription drug costs are a concern for Medicare beneficiaries, with half (50%) of Medicare beneficiaries age 65 and older worried about being able to afford prescription drugs for themselves or their family in 2026, according to a March 2026 KFF health tracking poll.

The Growth in Medicare Spending Will Lead to Higher Medicare Premiums, Deductibles, and Copayments Over Time

• People with Medicare will face higher premiums and other out-of-pocket costs for Medicare benefits over time since Medicare premiums and deductibles are directly tied to increases in spending on Medicare-covered services. Higher Medicare spending is driven in part by rising prices, higher utilization, and growth in spending on Medicare Advantage. Between 2026 and 2034, the Medicare Part B premium and deductible are projected to increase by roughly 70%—from $2,435 to $4,170 for the Part B premium (annualized) and from $283 to $486 for the Part B deductible, while the Part A deductible is projected to grow by about 30% (from $1,736 to $2,212) (Figure 12).

• Rising Medicare premiums and cost sharing will consume a growing portion of Medicare beneficiaries’ financial resources. On average, premiums and cost sharing for Medicare Part B and Part D accounted for 25% of the average Social Security benefit in 2025, and this share is projected to increase to 33% in 2040. Including other premium and cost-sharing amounts would increase these shares even further, though as noted earlier most older adults can tap other sources of income beyond Social Security to pay for health care expenses.

For the latest analysis on the medical frailty exemption and key takeaways from the published interim rule, see our updated brief.

The 2025 reconciliation law requires states to condition Medicaid eligibility for adults in the Affordable Care Act (ACA) Medicaid expansion group and enrollees in partial expansion waiver programs (Georgia and Wisconsin) on meeting work requirements starting January 1, 2027 or sooner at state option. The law specifies mandatory exemptions, including individuals who are “medically frail.” To ease the burden on individuals, the law directs states to use available information “where possible” to verify compliance with Medicaid work activities or exemption status, without requiring additional documentation from individuals. Given the abbreviated implementation timeline, states are moving forward with key decisions over how to implement the medical frailty exemption even as they wait for formal guidance from the Centers for Medicaid and Medicare Services (CMS), which is required to issue an interim final rule by June 1, 2026. This brief describes early state plans to operationalize the medical frailty exemption and identifies key issues that they are facing and may be answered in the forthcoming guidance.

What is the Medical Frailty Exemption?

The reconciliation law requires states to exempt from work requirements an individual who is “medically frail or otherwise has special medical needs.” The law specifies this includes individuals who are blind or disabled; have a physical, intellectual, or developmental disability that limits their ability to perform one or more activities of daily living (ADL); have a substance use disorder or a “disabling” mental disorder; and those with “serious or complex” medical conditions. This definition closely aligns with an existing federal medical frailty definition that CMS uses for states choosing to set “alternative benefit plans” that differ from the traditional Medicaid benefit package. In that context, CMS used the definition as a minimum standard for medical frailty but allows states flexibility to define medical frailty beyond the statutory and regulatory definitions. To identify individuals who are medically frail, CMS noted in the 2013 final rule that it expected states to use Medicaid claims data and encouraged them to use self-attestation and health screeners for applicants and new enrollees where the state does not have information on their current health status or historic encounter data.

How do States Plan to Operationalize the Medical Frailty Exemption?

Most states have not yet finalized a medical frailty definition, likely reflecting ongoing uncertainty over how much flexibility states will have as they await June CMS guidance. In a recent KFF survey (fielded January-March 2026), about half of states (22) subject to work requirements report having a current state medical frailty definition, though it is unclear whether these definitions align with the medical frailty provisions in the reconciliation law (Figure 1). Most of these definitions were not developed with work requirements in mind and may need to be updated to reflect the new federal requirements. Thirty-three states indicated they had not yet determined what definition they plan to use at the time the survey was fielded. If states are given flexibility by CMS to define who is medically frail and exempt from work requirements, six states would prefer to use a state definition (either an existing definition or a new definition,) while four states would use a federal definition.

States plan to use a variety of methods to verify medical frailty status, including using data to automate the process where possible. Most states reported plans to use Medicaid claims data (32) to verify medical frailty exemption status, while Georgia indicated it would not use claims data, and the remaining ten states had not yet made a decision at the time of the survey. New applicants and recent enrollees may not have claims data on file, requiring states to use other forms of verification.  All 32 states reporting plans to use claims data also reported plans to use other sources, such as data from other programs (e.g., enrollment in a behavioral health managed care plan or HCBS program), managed care utilization or claims data, and/or managed care case management data. Twenty-nine states said they will seek confirmation from a treating provider. Many states are developing health assessment screeners to collect information to identify medical exemptions, and 11 states said they will use the health screeners to verify medically frail status. Most states (30) also reported wanting to allow applicants and enrollees to self-attest to their medically frail status if verification data are not available and self-attestation is permitted by CMS. However, some states reported they do not plan to accept self-attestation (AR, HI, IN, MT) or have since passed legislation barring the use of self-attestation (NC and UT).

Nebraska, which became the first state to implement new work requirements on May 1, 2026, has released resources outlining how it will process medically frail exemptions, as well as an index of diagnosis and procedure codes. Examples of conditions covered in the thousands of codes include types of cancer, HIV, diabetes, heart disease, and certain types of mental health conditions and substance use disorders, among others. For existing enrollees, Nebraska is reviewing medical claims data prior to renewal and will auto-exempt an individual if indicated based on diagnosis/procedure codes. If Nebraska cannot determine medical frailty through claims data or other data sources, existing enrollees will need to submit a self-declaration form. Not all individuals will be able to be auto-exempted due to limitations of claims data such as:

• A lack of data on file for new applicants and recent enrollees;
• A lack of data on file due to providers not consistently coding or not using the codes identified by the state;
• Selected diagnoses and procedures potentially not being comprehensive of all individuals with target conditions or not capturing functional limitations;
• A lack of claims within the look-back period despite an individual having an exempted chronic condition or disability; and
• Delays between clinical care and claims presenting.

The state also intends to use the self-declaration form to identify medically frail exemption status for new applicants. The self-declaration form requires individuals to provide a description of their health condition, the contact information of their treating provider, and a note if services were provided while enrolled in Medicaid. No additional documentation is required at this time (e.g., a note from a provider or other proof of a condition). Nebraska indicates that it plans to move from self-declaration to more automated verification methods consistent with federal guidance over time, and plans to continue to review and update the index of ICD-10, CPT, and HCPCS codes as federal direction is clarified.

Early insights from other states highlight the variety of ways states are planning to use multiple sources of information to identify medical frailty. For example:

• California plans to automatically exempt individuals who are eligible for certain programs (e.g., home care—also known as home- and community-based services or HCBS) and has been exploring more timely data sources such as managed care plan care management systems in conjunction with evaluating which diagnostic codes could be used to establish medical frailty.
• Pennsylvania has outlined a plan to first check health care claims, followed by information from electronic health records exchanged through regional health information organizations, managed care organization (MCO) health assessments and case management information, and finally information provided by the applicant. Pennsylvania explains its approach is designed with data limitations in mind, and aims to reduce paperwork burden and streamline the process for beneficiaries.
• South Dakota has shared early plans to use clinical code lists and a medically frail health screener. Indicating lingering uncertainty on CMS’s stance regarding self-attestation, the state notes it may have to use physician attestation at renewal if screeners cannot be repeated and claims data is still not available.
• Several states also have reported that they would be interested in adding new categories such as homelessness to the medical frailty category but understand from communications with CMS that states may not have flexibility to do so.

How Will the CMS Guidance Address Key Issues and Potential Implications?

Although states are moving forward with operationalizing the medical frailty exemption, the interim final rule that CMS is required to release by June 1, 2026, is expected to address key issues related to defining and verifying medical frailty. Because of the time required to make system changes, states have been moving forward with implementing work requirements based on informal communication and guidance CMS has shared. Any significant changes to that informal guidance could require states to change course and could have implications for states’ readiness to implement work requirements as well as the costs associated with implementation. Key issues that may be addressed in the guidance, and that states report waiting on, include:

Defining Medical Frailty

• Defining medical frailty. CMS may provide a federal definition of medical frailty in the guidance. The law suggests the need for a Secretary-provided definition in its list of exemptions by including anyone “who is medically frail or otherwise has special medical needs (as defined by the Secretary).” It is not clear if CMS will rely on the existing definition or develop a new definition that more directly relates to work requirements. The statute includes people who are both medically frail and those with “serious or complex medical conditions,” which includes people with significant medical needs that may or may not limit their ability to engage in qualifying activities. For these individuals maintaining access to health coverage protects against serious health consequences if treatment is interrupted and may enable them to work.
• State flexibility to define medical frailty. Whatever federal definition CMS establishes, another important consideration is whether the definition serves as a minimum standard that states can use or expand on or whether CMS will require states to use the federal definition with little or no flexibility to interpret how it will be applied. For example, CMS may specify a set of diagnoses codes for states to use in their claims data analysis, but the effect of that list will depend greatly on whether states are allowed to add additional codes to CMS’ list. This may be particularly relevant in states that use state-specific diagnosis or treatment codes in their Medicaid billing systems. 
• Identifying people with functional limitations and mental health conditions. As noted, the law required people “with a physical, intellectual or developmental disability that significantly impairs their ability to perform 1 or more activities of daily living” be exempted. Guidance may provide additional insight into how states are expected to capture functional limitations in medical frailty definitions, as it may prove more difficult to identify these individuals using claims data. While there is a diagnostic code that could help to capture this, it is generally underutilized, with one barrier being a lack of financial incentive. More comprehensively identifying enrollees with these functional disabilities may require technical assistance for states and providers (i.e. in terms of utilizing different codes). Guidance may also provide clarity on identifying individuals with mental health conditions who may also be difficult to identify using claims data, as well as provide clarity on the exact diagnoses that would qualify.

Verifying Medical Frailty

• Data sources for verifying medical frailty. States will likely be expected to use claims data to automate identification of medically frail individuals. The guidance may provide information on other data sources states can use or would be expected to use, such as electronic health records, MCO health assessments and case management information, and pharmacy data. States may also be required to use SNAP data and other program data, such as HCBS enrollment or enrollment in a behavioral health managed care program, to identify medically frail individuals. Accessing some of these data sources may raise privacy concerns that the guidance may or may not address.
• Data look-back. The guidance may also specify the length of time states are permitted to look back to identify individuals with qualifying conditions when using data or verification from providers, including whether there are different look-back periods permitted at application versus renewal. The length of the look-back period will affect how many individuals are captured under the medically frail exemption (longer look-back periods are likely to pick up more people who may qualify as medically frail).
• Exemption re-verification. The legislation does not currently clarify how frequently medical frailty exemptions may last before states are expected to re-verify individuals’ exemption status. States may need to reverify exemption status at every renewal, or states may be able to create an internal flag and/or permanently exempt some people if their health condition or disability status is unlikely to change.
• Self-attestation. It is unclear if states will be permitted to accept self-attestation or if they will be prohibited from using it (altogether or in certain circumstances). Some state officials have noted the limitations of existing data sources, particularly claims data, for identifying new applicants who are medically frail as well as enrollees at their first renewal. Self-attestation could be subject to CMS audits, and guidance may outline under what circumstances states are expected to use self-attestation (e.g., only initial application, at renewal). If self-attestation is permitted, questions remain whether attestation alone will suffice or whether enrollees will be required to also provide supplemental information (e.g. provider sign-off, clinical or pharmacy records, etc.).
• Health screeners. States may look to use health screeners at application and renewal, especially in the absence of self-attestation. States may have flexibility to create their own health screeners, or there may be a federal template.
• Confirmation from providers. When using provider verification, the guidance may specify any requirements, including what information must be collected and whether providers will be required to assess whether an individual’s condition limits their ability to work. Relying on provider confirmation could increase administrative burden (on the clinical workforce, individuals, and states), particularly for providers that treat large shares of Medicaid patients, and, depending on the information requested, could raise ethical concerns among providers.
• Exemption hierarchy. When individuals qualify under multiple exemptions or when they would qualify for a medical frailty exemption but also are engaged in qualifying work activities, states will need to develop standards for how to operationalize which status is prioritized (e.g., checking exemption status and work status in a specified order, choosing to use whichever is longer lasting), especially if certain conditions will have longer exemption periods. The guidance may establish a hierarchy for states to follow or may give states the ability to set their own hierarchies.