As Congress considers ways to cut Medicaid spending to help finance the extension of federal tax cuts, a new KFF analysis finds that imposing a cap on federal spending per Medicaid enrollee—known as a “per capita cap”—could trigger a decrease in federal Medicaid spending over a 10-year period of $532 billion to almost $1 trillion, depending on how states respond to the cuts. An estimated 15 million people could lose Medicaid coverage by 2034, if states were to respond to the change by reducing their own Medicaid spending and curtailing eligibility.

The size of the cuts and which states would be hit hardest would depend on how states respond, and also whether the policy is combined with other changes, such as ending the 90% federal match for the Affordable Care Act’s Medicaid expansion.

The analysis shows that if states choose to increase their own Medicaid spending to offset federal reductions and maintain coverage and benefits, federal Medicaid spending would fall by $532 billion and state costs would increase by the same amount. State responses to federal cuts may vary. 

However, if Congress at the same time eliminates the enhanced federal matching rate for the Medicaid expansion (another significant policy being discussed, and which was examined by a previous KFF analysis), the impact would be bigger. Federal Medicaid spending cuts could range from $1 trillion–if states offset costs–to $2.1 trillion–if states respond by cutting state spending and eligibility over 10 years, with 30 million people losing Medicaid coverage by the end of the period.

In a party-line vote, the Republican-controlled House last night passed a budget resolution that would target cuts to Medicaid of up to $880 billion or more over a decade to help pay for tax cuts.

“The current proposals being discussed by Congress would lead to the largest Medicaid spending cuts and enrollment declines in the program’s history with an unprecedented cut in federal Medicaid funding to states,” said KFF President and CEO Drew Altman. “As our polling and focus groups with voters show, Americans, including many Trump voters, are not expecting, nor would they want, cuts to Medicaid, which would be felt across the country.” 

The analysis examines the potential impacts of two of the Medicaid proposals being discussed that would generate significant federal savings. It looks at the combination of them because lawmakers may pursue multiple, simultaneous changes to the program that would have interacting effects. Other proposals being discussed reportedly include imposing Medicaid work requirements, reducing the minimum federal matching rate, restricting states’ use of provider taxes to finance their share of Medicaid spending, and repealing certain Medicaid regulations. Future KFF analyses will examine these proposals as well.

Under the current system, the federal government reimburses states for a share of Medicaid enrollees’ costs, with no upper limit on expenditures, and states pick up the rest. Capping the federal contribution would force states to choose how to offset the funding reductions and could lead to increases in the uninsured rate and reduced revenue for health plans, hospitals and nursing homes.

Other key takeaways of the analysis include:

• Maintaining current Medicaid benefit levels and enrollment would require states to pay $1,500 more per enrollee under a per capita cap, or up to $2,300 more per enrollee under a per capita cap that is combined with the elimination of the enhanced Medicaid expansion match rate, by FY 2034.
• Decreases in Medicaid enrollment would vary by state and could be as high as 57% in some states if Congress implements both per capita cap financing and the elimination of the Medicaid expansion match.
• Spending reductions under a per capita cap would compound in future years as the per enrollee cap levels diverge further from spending levels expected without the cap, limiting states’ ability to meet changing program needs.

All states would be affected by a per capita cap, but the magnitude would vary due to differences in the mix of the types of Medicaid enrollees in each state. The effects would vary across eligibility groups because spending for different groups is expected to grow at different rates over time.  Also, the elimination of the enhanced Medicaid expansion match rate would only affect the 40 states and the District of Columbia that have adopted the expansion. 

For detailed state-by-state impacts of the proposed Medicaid financing changes on spending and enrollment, see Appendix Tables 1 and 2.

A related report released this week highlights the experiences and opinions of Medicaid enrollees who participated in five virtual focus groups that KFF conducted in January with enrollees who voted for President Trump or for Vice President Harris. Despite differences in political leanings, participants reported having favorable experiences with Medicaid and concerns about potential cuts to the program. While some Trump voters enrolled in Medicaid believed there is fraud in the program and were open to work requirements, they also didn’t think that President Trump would follow through on cuts to Medicaid because they believed he understood their financial struggles.

Medicaid covers one in five people in the U.S., and accounts for nearly $1 out of every $5 spent on health care. It covers 41% of all births, nearly half of children with special health care needs, and five in eight nursing home residents.

There are several options under consideration in Congress to significantly reduce Medicaid spending to help pay for tax cuts, with the recently passed House budget resolution targeting cuts to Medicaid of up to $880 billion or more over a decade. Medicaid is the primary program providing comprehensive health and long-term care to one in five people living in the U.S and accounts for nearly $1 out of every $5 spent on health care. Medicaid is administered by states within broad federal rules and jointly funded by states and the federal government, meaning restrictions in federal Medicaid spending could leave states with tough choices about how to offset reductions through cuts to Medicaid, cuts to other programs, or tax increases.

This analysis examines the potential impacts on states, Medicaid enrollees, and providers of implementing a per capita cap on federal Medicaid spending, which is one proposal that has been discussed in Congress. The analysis assumes, based on proposals floated in Congress, that the plan would cap federal Medicaid spending growth per enrollee for each of the five major eligibility groups at medical inflation. Given the likelihood of multiple, simultaneous changes to Medicaid and the interactive nature of those changes, the analysis also illustrates the effects of a per capita cap on Medicaid if implemented jointly with the elimination of the 90% federal match rate for the Affordable Care Act (ACA) expansion, another significant policy change that has been discussed in Congress and which was examined by a previous KFF analysis. Other proposals to reduce federal Medicaid spending have also been reportedly raised, including work requirements, a reduced federal matching rate, limits on provider taxes to finance the state share of Medicaid spending, and repeal of certain Medicaid regulations issued by the Biden administration. Future KFF analyses will examine these proposals as well.

Key takeaways

• The estimated effects of a per capita cap depend highly on what assumptions are made about policy specifications, future growth in Medicaid spending, and states’ responses to federal cuts.
• Capping per enrollee spending could reduce federal spending by $532 billion to nearly $1 trillion dollars between FY 2025 and FY 2034. Depending on how states respond, the policy could shift costs to states by $532 billion and leave total Medicaid spending unchanged or total Medicaid spending could decline by 14% (or $1.4 trillion).
• If per capita caps were implemented jointly with the elimination of the enhanced ACA expansion match rate, federal Medicaid spending could decline by $1 trillion to $2.1 trillion dollars between FY 2025 and FY 2034. Depending on how states respond, this combination of policies could shift costs to states by $1 trillion and leave total Medicaid spending unchanged or total Medicaid spending could decline by a quarter (or $2.6 trillion).
• If states reduced Medicaid eligibility proportionally to the cuts in federal spending, 15 million enrollees could lose Medicaid coverage under a per capita cap by the final year of the analysis, with that figure doubling to 30 million when combined with elimination of the ACA expansion match rate.
• States would face tough choices about Medicaid’s future under a per capita cap: maintaining current Medicaid benefit levels and enrollment would require states to pay $1,500 to $2,300 more per enrollee in the final year of the analysis.
• The effects of a per capita cap differ across eligibility groups because spending for different eligibility groups is expected to grow at different rates, with larger estimated effects among children and adults who are not eligible because they have a disability or are ages 65 and older.
• All states would be affected by a per capita cap, but the size of the effects vary due to differences in the mix of enrollees. Depending on state responses, state spending could increase up to 57% in some states if they chose to pay for the cuts. Enrollment could decrease by similar levels if states chose to restrict eligibility in response to the federal funding cuts.

Spending reductions under a per capita cap would compound over time as expected increases in Medicaid spending diverge from the caps. The effects would be smaller immediately following implementation, larger in FY 2034, the final year of this analysis, and continue to grow in the years after 2034. A per capita cap can provide lower and more predictable federal costs over time, but it would also fundamentally change the Medicaid federal-state partnership by eliminating the federal guaranteed match and transferring new financial risk to states. States would face significant challenges in efforts to pay for federal cuts and there would be pressure to reduce benefits and eligibility, with even larger effects if a per capita cap were paired with other Medicaid cuts. Beyond reduced Medicaid spending and enrollment, there could be increases in the number of people who are uninsured, fewer covered benefits for future Medicaid enrollees, and reduced revenues available for health plans and providers such as hospitals and nursing facilities.

What is the proposed policy change?

Medicaid spending is currently shared by states and the federal government with a guarantee to states for federal matching payments without a cap on federal expenditures. The percentage of costs paid by the federal government (known as the federal medical assistance percentage or “FMAP”) for most Medicaid enrollees is determined by a formula set in law designed to provide a higher federal match rate for states with lower per capita incomes. There are also higher match rates for certain services and populations like the ACA expansion group (90%). This leads to variation in the federal share of Medicaid spending across states. There is also considerable variation in per enrollee spending across states, due to state flexibility to determine eligibility levels, benefits, and provider payment, and across eligibility groups,reflecting differences in health care needs and utilization.

This analysis estimates the impact of implementing a per capita cap on the federal share of Medicaid spending. One proposal under consideration would “establish a per capita cap [on federal Medicaid expenditures] for each of the different enrollment populations set to grow at medical inflation.” While specifics on the implementation of this policy would be included in a legislative proposal, further details have yet to be released, and assumptions made here may differ from details included in proposed legislation. To estimate a per capita (i.e., per enrollee) cap policy, this analysis first establishes FY 2025 per enrollee spending as the base year estimate; then, starting in FY 2027, the analysis limits federal spending growth for the five major eligibility groups (children, adults, expansion adults, people with disabilities and aged 65+) to the consumer price index (CPI-U) plus 0.4 percentage points, which is KFF’s estimate of the difference between CPI-U and medical inflation (CPI-M) over the past 20 years (see Methods).

The analysis also estimates the combined impact of implementing a per capita cap and eliminating the ACA expansion federal match rate. The combined policy would include the same per capita caps on federal spending and also assumes that, starting in FY 2027, expenditures for people eligible through Medicaid expansion would be matched at each state’s traditional FY 2026 FMAP rate. This part of the analysis accounts for the interactive effects of these two policy changes; therefore, savings are lower than if each policy was modeled separately and the totals were added together (see Methods). There are a number of other Medicaid policy changes that have been suggested, and policy estimates would likely differ depending on the combination of policies and their interactive effects.

What is the potential impact on Medicaid spending?

This analysis does not make assumptions about specific state behavioral responses and instead examines how the impacts of the two policy alternatives vary based on two types of state responses to the cuts. The state responses are designed to illustrate the spectrum of potential policy change effects. However, in practice, each state is likely to respond to the policy change differently and spending impacts overall would likely fall within the range. While some states may choose to increase state spending to maintain current programs with substantially reduced federal funding, many would likely need to make programmatic cuts, making both the lower end and higher end estimates unlikely. The analysis does not explore people’s insurance coverage after losing Medicaid; some may enroll in another source of coverage, but many others would likely become uninsured. The estimates presented here are not directly comparable to the estimates of federal savings from the Congressional Budget Office (CBO) because CBO’s estimates account for people enrolling in other coverage and make assumptions about how states would respond in the aggregate. While the estimates assume that spending per enrollee by eligibility group would grow uniformly across states based on CBO’s national projections, it is likely that growth rates would vary by state, and that a cap would therefore have varying effects

Per Capita Cap

In analyzing the effects of a per capita cap on Medicaid spending, KFF considered the following two types of state responses.

• Pay for Federal Cuts: States maintain per enrollee spending and eligibility at current levels, picking up new costs due to the federal cap on per enrollee spending. Enrollment and total spending would remain constant while costs would shift from the federal government to the states. States would have to make offsetting cuts in programs other than Medicaid or raise revenues.
• Reduce Spending and Eligibility: States cap their share of per enrollee spending at medical inflation – e.g., by reducing payment rates to health care providers — and also reduce eligibility by the same percentage that federal spending is cut for each eligibility group to reflect the fact that the federal government is contributing less towards the cost. As a result, there would be decreases in enrollment and total, federal, and state spending. 

Capping per enrollee spending could shift costs to states or reduce total Medicaid spending by 14% ($1.4 trillion) over a 10-year period (Figure 1). If states offset the federal cuts, federal Medicaid spending could decrease by 8% or $532 billion over the 10-year period, and states would pay those costs, increasing the state share by 14% across all states. If instead, states cap their share of per enrollee spending and reduce Medicaid eligibility, federal Medicaid spending could decline by 15% or $989 billion, reflecting the fact that per enrollee spending is capped and enrollment goes down. State spending could decline by 12% or $450 billion. Combined, Medicaid spending could decrease by 14% or $1.4 trillion over the 10-year period. The analysis assumes states would reduce Medicaid eligibility proportionally to the cuts in federal spending, and the effects compound over time, meaning spending and enrollment effects are smaller in years just following implementation but grow over time. By the end of the 10-year period, enrollment could be 17% lower than current policy, meaning 15 million people could lose Medicaid coverage. Some who lose Medicaid may enroll in another source of coverage, but many others would become uninsured.

Per Capita Cap and Elimination of the ACA Expansion Match Rate

In analyzing the effects of a per capita cap on Medicaid spending if it were combined with the elimination of the ACA expansion match rate, KFF considered the following two types of state responses.

• Pay for Federal Cuts: States maintain per enrollee spending and eligibility at current levels, picking up new costs due to the federal cap on per enrollee spending and the loss of the higher match rate for expansion enrollees. Enrollment and total spending would remain constant while costs would shift from the federal government to the states.
• Reduce Spending and Eligibility: States cap their share of per enrollee spending at medical inflation, drop the ACA Medicaid expansion, and reduce eligibility by the same percentage that federal spending per enrollee is cut for each eligibility group, resulting in changes to enrollment and to total, federal, and state spending as well as per enrollee spending.

Capping per enrollee spending and eliminating the ACA expansion match rate could shift costs to states or reduce total Medicaid spending by a quarter ($2.6 trillion) over a 10-year period (Figure 2). If states pay for the cuts, federal Medicaid spending could decrease by 15% or $1 trillion over the 10-year period, and states would pay those costs, increasing the state share by 27% across all states. If instead, states cap their share of per enrollee spending and reduce Medicaid eligibility, federal Medicaid spending could decrease by 32% or $2.1 trillion, and state spending could decrease by 15% or $571 billion. Combined, Medicaid spending could decrease by 26% or $2.6 trillion over the 10-year period. Enrollment losses could increase each year proportionally with the increase in federal spending cuts, resulting in 30 million (or 36%) fewer people with Medicaid coverage by the end of the 10-year period.

Impacts on State Spending if States Pay for Federal Cuts to Medicaid

Under both policy alternatives, costs would shift significantly from the federal government to the states if states chose to maintain current benefits and eligibility, causing state Medicaid spending per enrollee to increase by $1,500 to $2,300 in FY 2034 (Figure 3). With federal Medicaid spending per enrollee capped, state spending per enrollee could increase from $5,500 under current policy to $7,000 with a per capita cap and $7,800 if the per capita cap were paired with the elimination of the ACA expansion match rate. By FY 2034, the federal share of Medicaid spending on average across states could fall from 64% to 54% with a per capita cap and 49% if the per capita cap were paired with elimination of the ACA expansion match rate.

All states would be affected by a per capita cap, but the specific effects vary depending on the mix of enrollees in a state, with state spending increasing by as much as 57% in some states if paired with the elimination of the ACA expansion match rate (Figure 4). If states opted to pay for the federal Medicaid cuts, state spending on Medicaid could increase anywhere from 6% to 57% over 10 years. Expansion states could experience much larger state cost increases, ranging from 18% to 57% if the per capita cap were paired with the elimination of the ACA expansion match rate, whereas state cost increases for all non-expansion states could remain at or below 21%. While this analysis assumes that growth rates for each eligibility group are uniform across states under current policy, it’s likely that growth rates would vary by state in any given period based on policy choices and underlying factors related to their economies and health systems.

Impacts on Medicaid Spending and Enrollment if States Respond to Federal by Cutting Spending and Eligibility

The effects of per capita caps differ across eligibility groups because spending for different eligibility groups grows at different rates, with larger effects among children and adults who are not eligible because they have a disability or are ages 65 and older. The effects vary by eligibility group because spending estimates under current policy assume that per enrollee spending for each group grows at different rates (aligning with CBO per enrollee spending growth assumptions), but the per capita cap policy would limit the growth for all groups using the same inflation rate. Spending grows at different rates largely based on the percentage of spending that comes from use of long-term care for each eligibility group. Although Medicaid enrollees who use long-term care have much higher per enrollee costs, CBO expects those costs to grow more slowly in future years. By FY 2034, the decline in per enrollee spending under a per capita cap could be: 24% for expansion adults, 20% for other adults, 19% for children, 11% for people eligible because of a disability, and 6% for adults ages 65 and older.

This analysis assumes that states would reduce eligibility in proportion with the cuts to federal spending, resulting in larger total spending and eligibility cuts among children and adults who are not eligible because they have a disability or are ages 65 and older (Figure 5). Under a per capita cap, total spending would drop to reflect the lower per person costs and the reductions in eligibility, which would be proportional to the cuts in federal spending. In FY 2034, an estimated 15 million fewer people could be covered by Medicaid including:

• 5.3 million children,
• 4.8 million adults eligible through the ACA expansion,
• 2.9 million parents and other adults under age 65,
• 1.3 million people with disabilities, and
• 0.6 million people ages 65 and older.

An additional 15 million expansion enrollees could lose Medicaid coverage (totaling about 20 million expansion enrollees by FY 2034) if the ACA expansion match rate is also eliminated. Some people losing Medicaid would be eligible for ACA marketplace coverage (those with incomes 100-138% of the poverty level) and others would be able to obtain employer-sponsored health insurance. But, others would become uninsured. Most people over age 65 and some who qualify for Medicaid based on a disability would generally be able to maintain Medicare coverage but could lose access to wrap-around services not covered by Medicare.

All states would be affected by a per capita cap, but the specific effects vary, with enrollment decreasing by as much as 57% in some states if paired with the elimination of the ACA expansion match rate in 2034 (Figure 6). If states respond to federal cuts by reducing eligibility, Medicaid enrollment could decrease anywhere from 12% to 57% in 2034. Expansion states could experience much larger enrollment declines if paired with the elimination of the ACA expansion match rate, ranging from 31% to 57%, whereas enrollment declines for all non-expansion states could remain at or below 17%.

What are other implications to consider?

Medicaid per capita caps lock in historical spending patterns, and the effects would compound overtime so the full implications would not be visible until after FY 2034. Per capita caps are initially set to reflect historical spending patterns that vary by state and eligibility group. Spending per enrollee will grow at the same rate for all states and eligibility groups, meaning that the low-spending states today will continue to be low-spending states indefinitely, and the same is true for spending across the eligibility groups. Moreover, the caps are typically designed to constrain federal Medicaid spending growth to a rate slower than is expected under current law, which is how they achieve federal savings. As time passes, the effects compound, limiting states’ ability to meet changing needs and demands. Federal spending reductions in FY 2034, the final year of the analysis, are larger than reductions over the full 10-year period (Figure 7), and the effects on enrollment and spending in future years would continue to grow.

The effects of a per capita cap on Medicaid spending and enrollment are also highly sensitive to policy design, inflation rates, and how states respond to the cuts; and estimates are highly sensitive to assumptions about those factors. Decisions about how to calculate the per enrollee allotments for the base year and how to grow the allotments over time would determine the magnitude of costs shifted to states. The effects also depend on how much Medicaid costs grow over time and how they compare with changes in the inflator used to calculate the caps. States’ responses to the cuts would determine how much total spending and enrollment changed. If per capita caps are paired with other cuts to Medicaid, effects would be larger, and there would be more pressure on states to respond with programmatic cuts. Estimates of the effects of different per capita cap proposals may find a wide range of outcomes depending on what assumptions are made to complete the analysis.

A per capita cap would fundamentally change the Medicaid federal-state partnership by eliminating the federal guaranteed match and transferring new financial risk to states. Per capita caps allow federal spending to rise with enrollment; however, spending would not rise to account for: increasing costs due to the emergence of new technology such as cell and gene therapies, changes in population health status that increase per enrollee spending, or increased provider payment rates enacted to address workforce shortages. Medicaid is currently a partnership between the federal and state governments with both entities sharing the financial risk. Under a per capita cap scenario, federal spending would be lower and more predictable. The federal government would only bear risk associated with enrollment changes and the states would assume 100% of the risks associated with other factors that affect health care spending.

Under any per capita cap policy, states would face challenges and there would be pressure to reduce benefits and eligibility, but the effects would be larger if a per capita cap were paired with other Medicaid cuts. To maintain current policy, states would have to increase state tax revenues or decrease spending on non-Medicaid services such as education, which is the largest source of expenditures from state funds. Given the size of the federal funding cuts, states would face significant challenges in efforts to replace the loss of federal funds, which would be exacerbated if paired with other reductions in federal funding for areas beyond Medicaid. The loss of federal revenues would create significant pressure to eliminate the Medicaid expansion and restrain the growth in per enrollee spending for other enrollees. Medicaid spending per enrollee already grows at a slower rate compared with private insurance and Medicare, leaving fewer options for cutting per enrollee costs. States would all respond differently but could reduce spending by cutting optional benefits (which includes prescription drugs and nearly all home care, also known as home and community-based services or HCBS), reducing provider payment rates, or cutting eligibility for the specific eligibility groups with higher per person spending, such as those linked to the use of long-term care.

If states are unable to maintain Medicaid eligibility, benefits, and payment rates, there could be increases in the number of people who are uninsured, reduced access to care, and significant reductions in payment rates to providers. Reduced Medicaid eligibility would mean an increase in the number of uninsured people, with the most notable increases likely among people eligible for coverage through the ACA expansion, though people with disabilities may be particularly vulnerable on account of their more extensive health needs. Enrollees with incomes between 100% and 138% of poverty could be eligible for coverage through the ACA marketplaces, but ACA coverage could soon become more costly for enrollees if the enhanced subsidies expire at the end of 2025, and few low-income people have access to insurance through their employer. An increase in the number of uninsured people could reverse gains in financial security, access to care, and health outcomes as well as lead to loss of revenues and increased uncompensated care costs for providers. Providers could also face revenue losses from lower Medicaid payment rates or coverage of fewer services, which could exacerbate issues with access to care for Medicaid enrollees and within the health care system more broadly.

Appendix

Promoting price transparency in health care is a policy approach with bi-partisan support in Congress and the public at large, and the first Trump administration finalized regulations that require group health plans and insurers to make detailed data with all their in-network payment rates available with the objective that such transparency would increase price competition and ultimately drive down health care costs.

This report documents how the vast troves of data reported in pursuit of those goals include misleading and unlikely prices, inconsistencies, and other oddities that pose significant challenges for researchers, industry and others seeking to make sense of the data.

The transparency regulations set out in significant detail the methods that payers should follow in reporting their rates, generating huge amounts of publicly available data since reporting began in 2022. The analysis, which relies on the extensive collection of this data downloaded and maintained by Turquoise Health, includes examples about each of the challenges identified and how they complicate efforts to use the data for its intended purposes.

The analysis is available through the Peterson-KFF Health System Tracker, an information hub dedicated to monitoring and assessing the performance of the U.S. health system.

Congressional Republicans and President Trump are in search of trillions of dollars in cuts to mandatory federal spending that could help offset the cost of extending expiring tax cuts. With spending on health programs accounting for a substantial share of federal spending, those programs are an obvious target to achieve overall spending goals in current budget reconciliation discussions. Medicaid has been the primary focus for federal spending cuts, but cuts to Medicare and the Affordable Care Act have also been floated. Cuts to discretionary spending, which includes funding for several federal health agencies, veterans health care, and global health, are not part of the reconciliation process, but the Trump Administration has taken unilateral actions to reduce this spending. Proposed cuts to federal spending on health programs and services have trade-offs and could increase the number of people without insurance; reduce access to health care; increase consumer costs for health care; and reduce payments for hospitals, nursing homes, and other providers. The effects would be felt by people of all incomes but would likely be concentrated among people with low incomes.

To provide context for ongoing discussions about federal spending, this brief analyzes current support from the federal government for health programs and services, including both spending and tax subsidies (that is, forgone tax revenues from provisions that reduce tax liability for people and businesses with qualifying health-related spending). The data come from the Office of Management and Budget, U.S. Treasury Department, and the Congressional Budget Office (see Methods).

Key takeaways

• The federal government spent $1.9 trillion on health care programs and services in fiscal year (FY) 2024, 27% of all federal outlays in that year, and collectively the largest category of federal spending.
• Forgone tax revenues to the federal government resulting from tax subsidies for employer sponsored insurance coverage (ESI) and a portion of the Affordable Care Act (ACA) premium tax credits together totaled $398 billion in FY 2024.
• Over 80% of all federal support for health programs and services, including spending and tax subsidies, goes to programs that provide or subsidize health insurance coverage, with 36% going to Medicare, 25% going to Medicaid and CHIP, 17% going to employment-based health coverage, and 5% going to subsidies for Affordable Care Act (ACA) coverage.
• Discretionary spending is a relatively small component of overall federal support for health programs and services. Over half (52% or $128 billion) of discretionary health spending paid for hospital and medical care for veterans. Discretionary health spending also provides funding for agencies such as the National Institutes of Health (NIH) (19% of discretionary health spending) and the Centers for Disease Control and Prevention (CDC) (4%), as well as global health (4%).

How does the federal budget support health programs and services?

Over one out of every four dollars in federal spending was used to pay for health programs and services in FY 2024 (Figure 1). The next largest categories are Social Security (21%), national defense (13%), and interest payments on the federal debt (13%). Combined, these four categories account for nearly three quarters of all federal spending.

Seventy percent of government support for health programs and services comes from mandatory spending, with tax subsidies accounting for 19% and discretionary spending accounting for 11% (Figure 2). There are three ways that the federal government provides support for health programs and services: mandatory spending (outlays), discretionary spending (outlays), and tax subsidies (forgone tax revenue, also called tax expenditures or tax preferences).

• Mandatory spending, also known as “direct spending,” is governed by regular legislative action in Congress and not determined through the annual appropriations or “spending” bills. Mandatory health spending includes nearly all Medicare spending ($839 billion), federal spending on Medicaid and CHIP ($584 billion in federal funding), and the refundable portion of the health insurance premium tax credit for coverage through the ACA Marketplaces ($111 billion in federal funding) (Appendix Table 1). (This does not include ACA premium tax credits that offset income taxes individuals owe to the federal government.)
• Discretionary spending is funded through annual appropriations bills. The largest component of discretionary health spending ($128 billion) is for veterans’ health care. Discretionary health spending also provides funding for agencies such as the National Institutes of Health (NIH) and the Centers for Disease Control and Prevention (CDC), as well as global health, or international health assistance, much of which is channeled through the Department of State and U.S. Agency for International Development (USAID). (For a more detailed discussion of federal funding for global health, including the potential impact of the Trump administration’s executive order pausing foreign aid, see KFF briefs, 10 Things to Know About U.S. Funding for Global Health, How Much Global Health Funding Goes Through USAID?, and The Status of President Trump’s Pause of Foreign Aid and Implications for PEPFAR and other Global Health Programs) (Appendix Table 2).
• Tax subsidies allow businesses and individuals to reduce the amount of taxes they must pay based on the amount of money they spend on qualifying health programs and services. Most of the foregone revenue to the federal government from health-related tax subsidies (85%) comes from the exclusion of employer contributions for health insurance, which totaled $384 billion in 2024. Other notable tax subsidies that support health care include the non-refundable portion of the tax credit for coverage through the ACA Marketplaces ($14 billion), contributions to medical and health savings accounts ($14 billion), deductions for medical expenses ($13 billion), and deductions for charitable contributions to health institutions ($10 billion) (Appendix Table 3).

What is the budget reconciliation process?

Congress will likely use reconciliation to make reductions in mandatory spending to help offset some of the cost of extending the expiring tax cuts. Budget reconciliation is a special legislative process used to make changes to taxes and mandatory spending that allows the Senate to pass legislation with only 50 votes rather than the customary 60 votes. There are limits to what kinds of legislation can be passed through budget reconciliation, and it can only be used for policies that make non-incidental changes to mandatory spending or revenues. Congress initially enacted the 2017 Tax Cuts and Jobs Act through reconciliation, and it is expected that the reconciliation process will be used again for an extension. That law made a set of broad but temporary tax cuts, which expire at varying years starting in 2025. The cost of extending these tax cuts has been estimated at $4.0 trillion between FY 2025 and FY 2034.

Reconciliation was originally intended to reduce budget deficits, and there are several unique rules governing the reconciliation process. The budget reconciliation process begins with the adoption of a budget resolution that is passed in both houses of Congress but not signed by the President. The budget resolution provides each Congressional committee with the dollar amount of budgetary changes that must be achieved over a specified budget “window,” which is usually either a five-year or ten-year period. These budget changes can either increase the deficit or decrease it. Budget resolutions often suggest specific policies to achieve budgetary changes, but those suggestions are not binding or enforceable. Committees of jurisdiction must meet the dollar targets in the budget resolution but have discretion as to how to meet the targets. A bill with all the details developed by committees would have to pass both houses of Congress and be signed by the president to become law.

To maintain privileged status in the Senate, a reconciliation bill cannot:

• Increase the deficit in years after the budget period (though it can increase the deficit during the budget period),
• Change Social Security spending or revenues,
• Be “extraneous” to the budget, meaning the reconciliation bill cannot include policies that have “merely incidental” fiscal impacts, or
• Make changes to discretionary spending.

How might a budget reconciliation package affect health insurance coverage?

There are four primary sources of health insurance coverage in the U.S.: Medicare, Medicaid, ACA Marketplaces, and employer sponsored coverage. Each of these sources of coverage could be subject to changes in a reconciliation bill. Of these four programs, Medicare accounts for the largest share (36%) of total federal support for health programs and services (including both spending and tax expenditures), followed by Medicaid and CHIP (25%), employer coverage (17%), and ACA Marketplaces (5%). Reconciliation could be used to reduce federal financial support for each of these programs because they are all funded through mandatory spending and/or financed with tax subsidies.

Existing discussions surrounding a reconciliation package have focused on reductions in mandatory federal health spending, the largest components of which are Medicare (52% of mandatory health spending) and Medicaid (36%) (Figure 3, Appendix Table 1). A 50-page menu of options for policies that could be included in a reconciliation package is largely focused on reducing federal spending to finance an extension of the 2017 tax cuts. Medicaid is the largest source of proposed cuts, but there are also options to reduce spending on Medicare and the ACA subsidies.

Medicaid: Medicaid covers 83 million low-income people, accounts for one fifth of health care spending in the U.S., and covers 61% of long-term care costs. Despite most adults having favorable views of Medicaid and only 13% thinking Medicaid cuts are a top priority, federal Medicaid funding ($584 billion in 2024, see Appendix Table 1) is at significant risk under Republican proposals to reduce federal spending by nearly one third over ten years. Changes to Medicaid under consideration include imposing a per capita cap on federal spending, reducing the federal government’s share of costs for the ACA expansion group, limiting provider taxes states use to help pay for their share of Medicaid costs, and imposing work requirements. Such policy changes would fundamentally alter how Medicaid financing works and large federal spending reductions would force states to make tough choices whether to raise new revenue, restrict the number of people covered, cover fewer benefits, or cut payment rates for physicians, hospitals, nursing homes, and other providers. Millions or even tens of millions of people could lose Medicaid coverage depending on how the policy was structured.

Medicare: Medicare provides health insurance coverage to nearly 68 million older adults and younger people with long-term disabilities and accounted for just over half (52% or $839 billion) of mandatory spending on federal health programs and services in FY 2024. Despite President Trump’s campaign promises not to cut Medicare, Republican lawmakers have put several Medicare savings proposals on the table in recent budget reconciliation talks, including implementing site-neutral payment policies, making changes to Medicare payment of uncompensated care, bad debt, and other hospital payments, and reforming Medicare payment for graduate medical education, among other changes. Altogether, these specific proposals could yield around $500 billion in 10-year savings, and would have the greatest impact on hospitals, with indirect effects on patients, depending on how hospitals responded to payment reductions.

ACA Marketplaces: Healthcare.gov and State-Based Marketplaces cover over 24 million people in 2025 (about 7% of the US population), most of whom are low-income. Federal support totaled $125 billion, including direct spending and tax subsidies in FY 2024, when about 19 million people received a subsidy.

ACA Marketplace subsidies are provided through the tax system, with most subsidized enrollees receiving an advanced payment of the premium tax credit, which they reconcile when they file their taxes the following year. Because incomes can be very volatile for ACA Marketplace enrollees (many of whom work shifts, are self-employed, or gig workers), predicting one’s income a year in advance can be difficult. The ACA currently limits how much an enrollee must pay back in the tax credit if their income is below four times the poverty level. However, a Ways and Means Committee document proposes to remove repayment limits for people who receive excess tax credits.

Since 2021, enhanced premium tax credits have lowered premium payments across all subsidized enrollees and made middle income people (over four times poverty) newly eligible for subsidies. These enhanced tax credits were originally passed as COVID relief and extended by the Inflation Reduction Act of 2022, but they are set to expire at the end of 2025. If the enhanced tax credits are not renewed by Congress, out-of-pocket premium payments for enrollees are expected to increase by over 75%, though this amount will vary by income and location. The cost to renew the subsidies would be $335 billion over ten years, according to CBO projections.

What health programs and services are discretionary, and subject to the annual appropriations process?

Only 11% of federal support for health care programs and services is discretionary spending, and over half of that amount (52%) pays for veterans’ health care (Figure 4, Appendix Table 2). In FY 2024, $128 billion in federal funding supported care for more than 7 million veterans. The next largest source of discretionary spending is the National Institutes of Health, which received $46 billion in 2024 (19% of discretionary spending). Smaller sources of federal discretionary spending include public health and social services emergency funding ($11 billion; 4%), global health ($10 billion; 4%), and the Centers for Disease Control and Prevention ($9 billion; 4%). (Global health totals presented here are outlays, which represent actual cash flows, and therefore do not match those presented in other KFF resources, such as the U.S. Global Health Budget Tracker, which highlight the budget authority totals as provided by Congress in annual appropriations and include some other funding components that are counted elsewhere in this analysis.)

Making changes to discretionary health spending through the appropriations process requires 60 votes in the Senate, meaning changes would require Democratic support, unless President Trump takes unilateral actions to reduce federal spending. Congress is supposed to pass appropriations bills by June 30 each year that provide funding for discretionary programs from October 1 through September 31 of the following year (the federal fiscal year). However, in most years, Congress does not pass the appropriations bills on time, and instead uses continuing resolutions to prevent lapses in federal funding for discretionary programs. The current continuing resolution funds the federal government through March 14, 2025. Extending funding for discretionary programs beyond that date will require a majority vote in the House and 60 votes in the Senate.

In the early weeks of President Trump’s second term in office, the administration has taken unilateral action to reduce federal funding, such as by laying off federal employees and issuing executive orders to freeze federal funding in various programs. Declining to spend appropriated funds is also known as “impoundment.” In 1974, Congress enacted the Impoundment Control Act in response to President Nixon’s attempts to refuse to spend Congressionally-appropriated funds. During the campaign, President Trump promised to “restore Impoundment Power,” and the administration characterizes the Impoundment Control Act as unconstitutional. The Trump administration also indicates that the current funding freezes are “programmatic delays” rather than deferrals. It is currently unclear the extent to which President Trump will be able to significantly reduce federal spending through these and other unilateral actions.

Key Points

• On January 24, 2025, President Trump directed the Secretary of State to “take all necessary actions to ensure U.S. taxpayer dollars do not fund organizations or programs that support or participate in the management of a program of coercive abortion or involuntary sterilization.” This provision of law, known as the Kemp-Kasten amendment, was first enacted by Congress in 1985 and has been included in appropriations language annually.
• Kemp-Kasten has often been used, as determined by presidents along party lines, to withhold U.S. funding to the United Nations Population Fund (UNFPA, the lead U.N. agency focused on global population and reproductive health).
• The U.S. government provided $194.4 million in FY 2023 for UNFPA – $30.6 million in core support and $163.8 million more for other project activities – making it the largest government contributor in that year.
• While framed broadly, Kemp-Kasten was originally intended to restrict funding to UNFPA specifically, after concerns arose about China’s population control policies and UNFPA’s work in China; to date, it has only been applied to UNFPA. Evaluations by the U.S. government and others have found no evidence that UNFPA directly engages in coercive abortion or involuntary sterilization in China, and more generally, UNFPA does not promote abortion as a method of family planning or fund abortion services.
• Kemp-Kasten has been used to withhold funding from UNFPA in 19 of the past 40 fiscal years. Under current law, any U.S. funding withheld from UNFPA is to be made available to other family planning, maternal health, and reproductive health activities.

What is the Kemp-Kasten Amendment?

The Kemp-Kasten amendment, first enacted in 1985, is a provision of U.S. law that states that no U.S. funds may be made available to “any organization or program which, as determined by the [p]resident of the United States, supports or participates in the management of a program of coercive abortion or involuntary sterilization.”¹  It was the congressional response to a Reagan administration decision in 1984 to temporarily withhold some funding from UNFPA and to begin conditioning its funding on assurances that the agency did not engage in or provide funding for abortion or coercive family planning. This policy change was made after concerns arose about whether UNFPA supported China’s coercive population policies.²  It was announced by the Reagan administration at the 2nd International Conference on Population in 1984, in conjunction with the “Mexico City Policy.”³  The Mexico City Policy required foreign NGOs to certify that they would not “perform or actively promote abortion as a method of family planning” with non-U.S. funds as a condition of receiving U.S. family planning assistance; the Trump administration recently expanded this restriction to include all U.S. global health assistance (see the KFF explainer on the policy).

What U.S. funding does Kemp-Kasten apply to?

Kemp-Kasten applies to all funds appropriated under the State and Foreign Operations appropriations act as well as any unobligated balances from prior appropriations. This includes all funding provided to the State Department and USAID, which, in turn, includes the vast majority of U.S. global health funding.⁵ 

When has Kemp-Kasten been in effect?

The Kemp-Kasten amendment has been in effect for 40 years. First enacted in 1985,⁶  its language has been included in the State and Foreign Operations appropriations act every fiscal year since then. (Although the provision is present in current law, language similar to Kemp-Kasten was also included in President Trump’s presidential memorandum reinstating the Mexico City Policy on January 24, 2025.⁷ ) While Congress has kept the amendment in place annually, it remains up to the president to determine whether or not to invoke Kemp-Kasten as a reason to withhold funding from an organization (see below).⁸ 

Though Kemp-Kasten technically could apply to funding provided to any organization or program (including U.S. NGOs, non-U.S. NGOs, multilateral organizations, and foreign governments), the U.S. government has issued determinations about only one organization, UNFPA, thus far. The U.S. played a key role in the launch of UNFPA in 1969 and was, until 1985, the largest government donor to the agency.⁹  However, the U.S. has withheld funding from UNFPA due to presidential determinations that it violated Kemp-Kasten as often as it has provided funding since 1985 (in 19 of the past 40 fiscal years, to date), and in some years, funding was also withheld from UNFPA based on other provisions of the law, such as the dollar-for-dollar withholding requirement¹⁰  (see below). These determinations have been made along party lines with only one exception – the first year of President George W. Bush’s administration (see Figure 1 and Table 1).

How much funding does the U.S. provide to UNFPA?

In 2023, the U.S. was the largest government donor to UNFPA, having contributed 11% of all contributions (funding amounts for 2024 are not yet available). Total funding from the U.S. in FY 2023 for UNFPA was $194.4 million – $30.6 million in core support as well as $163.8 million in non-core support (see Box 2).¹¹  See Figure 1 and Table 1 for historical funding data.

How is a determination about Kemp-Kasten made?

By law, it is up to the president to determine whether any organization or program should be ineligible for funding due to a violation of the Kemp-Kasten amendment (in practice, this authority has generally been delegated to the State Department). In most recent years, legislative language has also specified that this determination must be: 1) made no later than six months after the date of enactment of the law that includes the provision and 2) accompanied by the evidence and criteria used to make the determination.¹³ 

Most recently, on January 24, 2025 at the beginning of his second term, President Trump directed the Secretary of State to begin the process of Kemp-Kasten determination by taking “all necessary steps.” These determinations are usually made after the annual appropriations process is completed. For example, in 2017, the Trump administration’s determination was made on March 30, 2017, at the six month mark after the passage of the FY 2017 continuing resolution appropriations bill and was accompanied by a two-page justification memorandum.¹⁴ 

Has there ever been evidence that UNFPA supports coercive abortion or involuntary sterilizations?

To date, there has not been evidence that UNFPA supports coercive abortion or involuntary sterilizations. Several evaluations by the U.S. government (including one by an assessment team sent to China by the State Department in 2002) as well as other groups, such as the British All-Party Parliamentary Group on Population, Development, and Reproductive Health (in 2002) and the Interfaith Delegation (in 2003), have found no evidence of direct engagement by UNFPA in such activities in China or elsewhere.¹⁵  In addition, UNFPA does not promote abortion as a method of family planning or fund abortion services.¹⁶  In years when a determination has been made that UNFPA violated Kemp-Kasten, the U.S. government has stated that the determination was based on its conclusion that UNFPA support to or partnering with the Chinese government for other population and reproductive health activities was sufficient grounds for invoking the amendment to withhold funding. In the March 30, 2017, determination by the Trump administration, for example, the justification memorandum stated that: “While there is no evidence that UNFPA directly engages in coercive abortions or involuntary sterilizations in China, the agency continues to partner with the NHFPC [China’s National Health and Family Planning Commission] on family planning, and thus can be found to support, or participate in the management of China’s coercive policies for purposes of the Kemp-Kasten amendment.”

What other legislative requirements apply to U.S. funding for UNFPA?

In addition to Kemp-Kasten, there are several other provisions of law that Congress has enacted in recent years to set conditions on U.S. funding for the agency.¹⁷  These provisions:

• require UNFPA to keep U.S. funding to the agency in a separate account, not to be commingled with other funds;
• prohibit UNFPA from funding abortion;
• prohibit UNFPA from using any U.S. funds for their programming in China;
• reduce the U.S. contribution to UNFPA by one dollar for every dollar that UNFPA spends on its programming in China (“dollar-for-dollar withholding”); and
• in some years, state that not more than half of funding designated for the U.S. contribution to UNFPA is to be released before a particular date, which varies by fiscal year (this provision is not currently in effect).

What happens to funding that is withheld from UNFPA?

For several years, including FY 2017, FY 2018, FY 2019, and FY 2020, Congress has required that funding withheld from UNFPA be reallocated to USAID’s family planning, maternal, and reproductive health activities.¹⁸  The enactment of this provision first affected reallocation of FY 2002 funds.¹⁹  It is now typically included in the State and Foreign Operations appropriations act each year.²⁰ 

1. U.S. Congress, FY 2017 Consolidated Appropriations Act (P.L. 115-31), May 5, 2017; KFF, The U.S. Government and International Family Planning & Reproductive Health: Statutory Requirements and Policies, fact sheet. ↩︎
2. Congressional Research Service (CRS), The U.N. Population Fund: Background and the U.S. Funding Debate, RL32703, July 2010; “Policy Statement of the United States of America at the United Nations International Conference on Population (Second Session), Mexico City, Mexico, August 6-14, 1984,” undated. ↩︎
3. “Policy Statement of the United States of America at the United Nations International Conference on Population (Second Session), Mexico City, Mexico, August 6-14, 1984,” undated; United Nations Division of Economic and Social Affairs/Population Division, “United Nations Conferences on Population,” webpage, undated, http://www.un.org/en/development/desa/population/events/conference/index.shtml. ↩︎
4. “Policy Statement of the United States of America at the United Nations International Conference on Population (Second Session), Mexico City, Mexico, August 6-14, 1984,” undated. ↩︎
5. KFF, The U.S. Congress and Global Health: A Primer; and KFF U.S. Global Health Budget Tracker, available at: https://modern.kff.org/interactive/u-s-global-health-budget-tracker/. ↩︎
6. Via FY 1985 supplemental appropriations, per CRS, The U.N. Population Fund: Background and the U.S. Funding Debate, RL32703, July 2010. ↩︎
7. Specifically, in this memorandum, President Trump stated, “I further direct the Secretary of State to take all necessary actions, to the extent permitted by law, to ensure that U.S. taxpayer dollars do not fund organizations or programs that support or participate in the management of a program of coercive abortion or involuntary sterilization.” ↩︎
8. However, after UNFPA ended its program in China in 1997 but then began a new program there in 1999, this resulted in Congress withholding funding from UNFPA that year. ↩︎
9. CRS, The U.N. Population Fund: Background and the U.S. Funding Debate, RL32703, July 2010; PAI, Why the United States Should Maintain Funding for UNFPA, May 2015. ↩︎
10. In FY 1999, Congress prohibited UNFPA funding in response to the initiation of a new UNFPA program in China (this was unrelated to Kemp-Kasten), and in some other years when the U.S. made a contribution to UNFPA, UNFPA’s China program meant some UNFPA funding was withheld under the “dollar-for-dollar withholding” provision. ↩︎
11. KFF analysis of data from: Congressional Appropriations Bills, Press Releases, and Conference Reports; Federal Agency Budget and Congressional Justification documents and Operating Plans; ForeignAssistance.gov; Office of Management and Budget, personal communication; UNFPA, personal communication; Biden Administration, “Memorandum on Protecting Women’s Health at Home and Abroad,” presidential actions, Jan. 28, 2021, Biden White House Archives, https://bidenwhitehouse.archives.gov/briefing-room/presidential-actions/2021/01/28/memorandum-on-protecting-womens-health-at-home-and-abroad/. ↩︎
12. UNFPA, Annual Report 2013, 2014. ↩︎
13. Typically included in annual State and Foreign Operations appropriations since FY 2008, including in FY 2017 under the terms of the continuing resolution. CRS, The U.N. Population Fund: Background and the U.S. Funding Debate, RL32703, July 2010; KFF analysis of appropriations bills. ↩︎
14. State Department: Letter to Bob Corker, Chairman, Committee on Foreign Relations, from Joseph E Macmanus, Bureau of Legislative Affairs, State Department, dated April 3, 2017, and accompanying “Determination Regarding the ‘Kemp-Kasten Amendment,’” dated March 30, 2017, and “Memorandum of Justification for the Determination Regarding the “Kemp-Kasten Amendment,” undated. Available online (follows the article) at: https://www.buzzfeednews.com/article/jinamoore/the-us-wont-give-any-more-money-to-the-un-population-fund. ↩︎
15. CRS, The U.N. Population Fund: Background and the U.S. Funding Debate, RL32703, July 2010. ↩︎
16. UNFPA, “Frequently Asked Questions,” webpage, updated January 2025, http://www.unfpa.org/frequently-asked-questions#abortion. ↩︎
17. KFF, The U.S. Government and International Family Planning & Reproductive Health: Statutory Requirements and Policies, fact sheet. ↩︎
18. KFF analysis of Congressional Appropriations Bills. ↩︎
19. “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,” per KFF, The U.S. Government and International Family Planning & Reproductive Health: Statutory Requirements and Policies, fact sheet. ↩︎
20. The activities to which Congress directs reallocated funds varies by fiscal year; in FY 2003, for example, reallocated funding supported assistance to vulnerable children and victims of trafficking in persons. CRS, The U.N. Population Fund: Background and the U.S. Funding Debate, RL32703, July 2010. ↩︎

Options under consideration in Congress to significantly reduce Medicaid spending could have major implications for adults who live with mental illness. Nationwide, an estimated 52 million nonelderly adults live with mental illness, and Medicaid covers nearly one in three (29%) of them, or about 15 million adults. Changes to Medicaid under consideration include imposing a per capita cap on federal spending, reducing the federal government’s share of costs for the ACA expansion group, and imposing work requirements. Such policy changes would fundamentally alter how Medicaid financing works and large federal spending reductions would force states to make tough choices on whether to raise new revenue, restrict the number of people covered, cover fewer benefits, or cut payment rates for physicians, hospitals, and other providers. For people with mental illness, losing Medicaid coverage would reduce access to mental health treatment and other health care, which could have negative implications for their mental and physical health.

1. More than one in three adult Medicaid enrollees have a mental illness.

More than one in three nonelderly adults enrolled in Medicaid have a mental illness (35%), including 10% with a serious mental illness. These rates are higher than the rates among adults with private insurance or no coverage (Figure 1). Serious mental illness generally involves more severe symptoms and may include other neurological factors, both of which can complicate treatment and impact daily functioning. Data in Figure 1 come from the National Survey on Drug Use and Health (NSDUH), which categorizes respondents as having a probable mild, moderate, or serious mental illness through a combination of mental health scales and indicators of functional impairment. Generally, adults with severe symptoms and with greater impairments in daily functioning would meet the threshold for serious mental illness, while “any mental illness” includes adults who meet criteria for mild, moderate or serious mental illness (see methods). Among Medicaid adults, any mental illness is most prevalent among White adults, rural or small metro residents, those aged 26-34, and females (Appendix Table 1). Rates of any mental illness among adult Medicaid enrollees vary widely by state, from 22% in New Jersey to 51% in Iowa. Similarly, the percentage of adult Medicaid enrollees with a serious mental illness ranges from 4% in Mississippi to 22% in Wyoming and Missouri (Appendix Table 2).

2. Mental health treatment rates for Medicaid-enrolled adults are higher or similar to those with private insurance.

Nonelderly adults in Medicaid receive mental health treatment at rates that are higher or similar to those with private insurance, and much more often than those who are uninsured. In 2023, 59% of adult Medicaid enrollees with any mental illness received treatment – somewhat above the rate for privately insured adults (55%) and far higher than for those who are uninsured (37%). Across all coverage types, treatment rates increased with the illness severity, with adults diagnosed with serious mental illness reporting the highest rates of treatment. Although adults with mild mental illness covered by Medicaid or private insurance report similar treatment rates, Medicaid-enrolled adults with moderate and serious mental illnesses report higher treatment rates than those with private coverage. In every category, adults with either Medicaid or private insurance receive treatment at much higher rates than those who lack health coverage. These data indicate only that care was accessed, not the adequacy or quality of that care. In an effort to increase access to care, many state Medicaid programs have expanded covered services and adopted policies to address workforce shortages.

3. Adults enrolled in Medicaid experience a range of mental health diagnoses.

Anxiety and depressive disorders are the most frequently diagnosed mental illnesses among nonelderly Medicaid-enrolled adults (over 5 million diagnoses for anxiety alone). Serious mental illnesses – such as bipolar disorder and schizophrenia or other psychotic disorders – were diagnosed in over 2.3 million adult Medicaid enrollees (Figure 3). Many people experience overlapping mental health conditions (e.g., a person with depression may also have anxiety). Estimates of the number of people with specific types of mental illnesses come from Medicaid administrative data (i.e. claims data) which only capture diagnoses for people with a mental illness diagnosis recorded in their medical claims. This is not a measure of overall prevalence of mental illness because not everyone is screened, and diagnoses are not always recorded. Prevalence rates estimated through surveys are generally higher than the prevalence rates observed in claims data.

4. Medicaid enrollees with mental illness have higher rates of chronic conditions.

Medicaid enrollees diagnosed with mental illness have higher rates of chronic conditions and substance use disorder compared to those without a mental health diagnosis. Approximately two-thirds of adult enrollees with any mental illness have at least one other chronic condition – twice the rate of those without a diagnosed mental illness (Figure 4). Enrollees with serious mental illness have the highest rates of chronic conditions with 76% having at least one chronic condition. Chronic conditions are those that last at least a year and require ongoing medical care or limit daily activities, such as heart disease, diabetes, cancer, and mental illnesses. The most common chronic condition among Medicaid enrollees with mental illness is substance use disorder. These often overlap with mental illness, potentially due to shared risk factors. One in four enrollees diagnosed with any mental illness also have a diagnosed substance use disorder, and about 40% of those diagnosed with SMI do, compared to 5% of enrollees without a mental health diagnosis (Figure 4).

5. Average annual spending for enrollees with a mental illness is twice that of those without.

Average annual Medicaid spending per nonelderly adult enrollee is twice as high for those with any mental health diagnosis—about $14,000 per year—compared to roughly $7,000 for those without a mental health diagnosis. Spending is highest among adults diagnosed with a serious mental illness, at approximately $21,000 per enrollee per year—three times higher than the annual spending for those without a mental illness (Figure 5). Medicaid spending for adults with any mental illness accounts for one-third of the total Medicaid spending for non-elderly adults enrolled only in Medicaid. Higher rates of other chronic diseases among adults with mental illness may contribute to higher spending (Figure 3).

Appendix Tables

PEPFAR, the U.S. global HIV/AIDS program, is – for the first time in its two-decade history – facing significant challenges that could impede its ability to fulfill its mission. One of President Trump’s first actions was to issue an executive order to re-evaluate and realign foreign aid, requiring a 90-day pause in foreign aid funding while a review was undertaken, as well as dismissal of thousands of USAID staff and contractors, and other changes, effectively halting most programs, including for PEPFAR, around the world. Despite PEPFAR receiving a limited waiver to continue some services and the courts stepping in to provide temporary relief, services are still disrupted. In addition, PEPFAR’s current short-term authorization expires on March 25, 2025, about a month before the 90-day aid review is scheduled to be completed, and Congress is increasingly looking for program reforms and ultimately scale down.

The current situation, coupled with uncertainty about future changes, poses potential risks to health outcomes for a program that has been shown to have saved millions of lives and helped to build health infrastructure in sub-Saharan Africa; already, analyses have estimated that the foreign aid freeze and ensuing service disruptions have led to increases in HIV-related deaths and new HIV infections. This policy brief provides an overview of these recent events and ongoing challenges facing the program.

Background

Despite almost two decades of strong, bipartisan support and demonstrated success and impact, PEPFAR began facing growing headwinds in recent years. This was due to several external shifts, including pressures on the global economy; an increasingly crowded global health and development space; and shrinking resources to address the global HIV epidemic, with a concomitant and growing reliance on the United States, the largest donor to HIV.

Within the U.S., members of Congress were increasingly asking questions about PEPFAR’s “end game,” pressing the program for its plans for the future to scale down and transition services, programming and, ultimately, financing to countries. In addition, some members began raising concerns that PEPFAR was supporting abortion activities, which is prohibited under federal law and policy. Together, these issues rendered PEPFAR unable to secure a five-year reauthorization for the first time in its history. Instead, Congress reauthorized the program for just one year (from March 23, 2024, to March 25, 2025), pushing further decision until after the 2024 election. In the meantime, ongoing concerns about abortion prompted Senator Risch to put a $1 billion hold on the program in September 2024.

What to Watch

Against this backdrop, several recent developments have made PEPFAR’s future particularly precarious; many of these intersect with one another, adding uncertainty and complexity to an already shifting landscape. These include:

• Increasing Abortion Concerns. In January, just before President Trump assumed office, PEPFAR notified Congress that some funds had, in violation of the Helms amendment, mistakenly been used to support a limited number of abortions in Mozambique, where abortion is legal in certain circumstances. According to one report, this was discovered by U.S. officials in late October 2024 when routine compliance measures found that some PEPFAR-funded nurses had not received required training on the prohibition on paying for the performance of abortion in U.S.-funded services, with four found to have performed abortions. In a statement issued on January 17, PEPFAR enumerated the steps it took to halt and remediate the violation and to prevent future violations, including immediately suspending funding and obtaining reimbursement from the Government of Mozambique for the amount (reported to be $4,100 spent on the nurse’s salaries) and putting in place new, additional measures, such as “requiring an annual signed attestation by PEPFAR-funded clinical service providers to ensure compliance with U.S. funding restrictions.” While this was the first time there had been evidence of any PEPFAR violation of the prohibition on abortion, it raised significant additional concerns among members of Congress who called for further investigation. This has led to increased scrutiny of the program and is likely to figure significantly in any discussion of reauthorization and PEPFAR’s future more broadly.
• President Trump’s Foreign Aid Funding Freeze and PEPFAR’s Limited Waiver. Beginning on the first day of his second term, President Trump issued several executive actions that affected and continue to affect PEPFAR’s operations, most notably a January 20 Executive Order reevaluating foreign aid. The order called for a 90-day freeze on U.S. foreign aid, including PEPFAR funding, to allow for a review of foreign assistance programs for alignment with Trump administration policy. While the order focused on pausing funding for new obligations and disbursements, on January 24, Secretary of State Rubio issued a stop-work order on all existing operations (those already underway), with limited exception. This, coupled with actions affecting USAID, PEPFAR’s main U.S. government implementing agency (see below), effectively halted service delivery.Although PEPFAR was able to secure a limited waiver on February 1 to allow some services to continue, communication about the waiver has been slow or unclear and the payment system remained unavailable. In addition, the waiver was limited to care and treatment only, as well as PMTCT and PrEP for pregnant and breastfeeding women; no other prevention services, including PrEP for those at risk of HIV infection or already on PrEP, were allowed. Even where the waiver might have been communicated to implementers, the capacity to deliver services has already been negatively affected with thousands of aid workers having already lost their jobs.In a set of lawsuits brought by organizational recipients of U.S. foreign aid, the court issued a temporary restraining order on February 13, requiring the administration to end the foreign aid freeze. Still, the case continues to move through the legal process and the government has since indicated that, despite the requirement to resume funding, it acted within its authority to cancel most grants and contracts. It is also unknown how much of the damage already done to service capacity can be repaired, should funding resume. Finally, the administration’s foreign aid review is ongoing, and expected to conclude by April 19 with recommendations for programmatic changes and potential cuts. How PEPFAR fares in this process is unclear but could have significant implications for its future. It is also unknown how any recommendations about changes to PEPFAR under the foreign aid review will be met by Congress and/or figure into Congress’ own deliberations about reauthorization.

• The Potential Dismantling of USAID. PEPFAR is overseen by the State Department but implemented primarily by other U.S. government agencies – particularly USAID and CDC. In FY 2023, 60% of PEPFAR’s bilateral HIV assistance was obligated and implemented by USAID. Shortly after the foreign aid pause was announced, however, President Trump and others in the administration announced their intention to dissolve USAID and implemented a series of actions that affected its operations and capacity including: taking down its website and payment and data systems, letting go of hundreds of staff and plans to put thousands more on leave (although a federal judge has temporarily enjoined the administration from doing so), closing the USAID building, and appointing the Secretary of State as the Acting Administrator of USAID. As mentioned above, this has already significantly affected PEPFAR’s operations even with a limited waiver issued, including widespread reports that there are no USAID personnel to communicate about the waiver, answer questions about waiver implementation, or facilitate payment. More long-term, it remains unclear what will happen with the agency including the recommendations that will be made by the State Department and whether Congress will accept proposed changes. Given the current reliance of PEPFAR on USAID as one of its main implementing agencies, any change stands to impact programming.
• The Reinstatement of the Expanded Mexico City Policy. As was expected, President Trump reinstated the Mexico City Policy, marking the resumption of the expansion of the policy from his first term that, for the first time, included PEPFAR. The Trump administration’s 2017 expanded policy required foreign non-governmental organizations (NGOs) to certify that they would not “perform or actively promote abortion as a method of family planning” using funds from any source (including non-U.S. funds) as a condition of receiving U.S. government global family planning funding and most other bilateral U.S. global health assistance (prior to this time, only family planning assistance had been subject to the policy). An analysis of the impact of Trump’s expanded policy found significant decreases in services offered by PEPFAR implementing organizations, including reductions in HIV testing and counseling among other services. The newly reinstated policy calls for the Secretary of State, in coordination with the Secretary of Health and Human Services, to develop a plan to implement the policy, and it is possible that the plan will reach even further than during Trump’s first presidency, as some have called for.
• PEPFAR’s Reauthorization Uncertainty. As mentioned earlier, PEPFAR’s current authorization expires March 25, which means that eight time-bound provisions would lapse if Congress does not act to extend them. However, the PEPFAR program overall would continue, so long as funding is appropriated. Still, how concerns raised by some members of Congress about abortion and increasing pressure on PEPFAR to demonstrate more concrete plans about its future, will figure into reauthorization have yet to be seen. It is possible that the new measures announced by PEPFAR to assure compliance with U.S. law, the reinstatement of the Mexico City Policy, and, presumably, a new PEPFAR coordinator appointed by President Trump, will alleviate abortion concerns. At the same time, the current upheaval in PEPFAR services and disruptions around the world may accelerate Congress’ interest in playing a more active role in charting its future, as will its interest in seeing more active sustainability reforms – for example, Congress could assess whether to include new provisions in PEPFAR authorizing language, such as country co-financing or graduation requirements.
• Future Funding. Finally, while PEPFAR’s funding has been relatively flat for several years, it now enters a much more uncertain funding environment. Last year, during budget negotiations, both the House and Senate’s final FY 2025 appropriation bills for State, Foreign Operations, and Related Programs (which funds most of PEPFAR) included level funding for program. Still, Congress could not agree on an overall budget for FY 2025 and the entire government is operating under a continuing resolution through March 28, 2025, requiring Congress to either agree on a FY 2025 budget or adopt another continuing resolution (or the government shuts down), and reports already indicate that Republicans will be looking for deep cuts in the budget. This changes the overall funding calculus for PEPFAR potentially for FY 2025 but also beyond. Contributing to this is the likelihood that President Trump’s first budget request for FY 2026, which will be released soon, will call for funding cuts to global health, including PEPFAR, as it did during his first term – at that time, Congress balked at those cuts but it is much less certain how they will respond now, given current fiscal and political pressures.