This analysis of government data finds that people with medical debt are much more likely to have other forms of financial distress than those without medical debt, like having no “rainy day” fund, overdrawing a checking account, or relying on costly loans.

Medical debt is associated with financial vulnerability across a range of other indicators and can cause people to delay or forgo needed medical care due to cost.

The analysis relies on data from the 2021 National Financial Capabilities Survey. The survey uses information from more than 27,000 adults in each state and D.C.

The analysis is available through the KFF-Peterson Health System Tracker, an online information hub that monitors and assesses the performance of the U.S. health system.

This analysis of government data estimates that people in the United States owe at least $220 billion in medical debt. Approximately 14 million people (6% of adults) in the U.S. owe over $1,000 in medical debt and about 3 million people (1% of adults) owe medical debt of more than $10,000.

The share of adults with medical debt varies considerably across the U.S. Hawaii (2.3%) and D.C. (2.7%) have the lowest share of adults with medical debt. States with the highest share of adults with medical debt include South Dakota (17.7%), Mississippi (15.2%), North Carolina (13.4%), West Virginia (13.3%), and Georgia (12.7%).

The analysis is based on data from the 2021 Survey of Income and Program Participation, a nationally representative survey that asks every adult in a household whether they owed money for medical bills and how much they owe.

The report also examines variations in medical debt by age, race and ethnicity, health status, and for people with a disability.

The analysis is available through the KFF-Peterson Health System Tracker, an online information hub that monitors and assesses the performance of the U.S. health system.

About 12.9 million people received health coverage under both Medicare and Medicaid in 2021. Medicare-Medicaid enrollees, known as dual-eligible individuals, are a diverse group, as nearly half are people of color and nearly 40% are under age 65. However, this group of people share some common characteristics, such as limited financial means and health care needs that are more wide-ranging and complex than the average Medicare enrollee.

Prior KFF research has described the complex landscape of coverage options available to dual-eligible individuals when enrolling in Medicare and Medicaid. Medicare, the primary source of health insurance coverage for dual-eligible individuals, may be provided under traditional Medicare or a Medicare Advantage plan. Medicaid, which typically wraps around Medicare, covers the cost of Medicare premiums and in many cases, cost sharing assistance. Full dual-eligible individuals are also eligible for benefits not otherwise covered by Medicare, such as long-term services and supports. Dual-eligible individuals may receive Medicaid benefits through fee-for-service or managed care, and coverage and eligibility vary by state. Separate eligibility requirements, benefits, and rules for Medicare and Medicaid sometimes contribute to what has been described as a “fragmented and disjointed system of care for dual eligibles” which may lead to difficulty in navigating care among dual-eligible individuals.

In 2023, 5.2 million dual-eligible individuals were enrolled in a Medicare Advantage plan designed specifically for dual-eligible individuals, known as Dual-Eligible Special Needs Plans (D-SNPs). D-SNPs are required to provide greater coordination of Medicare and Medicaid benefits than other Medicare Advantage plans to improve coordination across programs and patient outcomes. D-SNPs typically provide benefits not otherwise available in traditional Medicare and generally do not charge a premium.

This brief highlights 10 things to know about D-SNPs, including national and state enrollment trends, plan availability, insurer participation, benefits, and prior authorization rates and denials, based on data from various sources (see methods for details). While D-SNPs can help with coordination for dual-eligibles in Medicare and Medicaid and they are growing rapidly, fewer than one in ten D-SNP enrollees were in fully integrated dual eligible (FIDE) special needs plans (SNPs), raising questions about how well coverage and care is being coordinated between Medicare and Medicaid. In addition, gaps in data make it difficult to assess the quality of D-SNPs, prior authorization rates and denials by type of service, and the extent to which extra benefits are used.

1. About 3 in 10 (29%) dual-eligible individuals enrolled in D-SNPs in 2021.

In 2021, nearly 3 in 10 (29%) dual-eligible individuals were enrolled in a D-SNP. Between 2010 and 2021, the share of dual-eligible individuals enrolled in D-SNPs nearly tripled from 11% to 29%.

As of 2023, most D-SNP enrollees (57%) were in coordination-only (CO) plans that are required to provide a minimum level of coordination between Medicare and Medicaid but are not fully integrated within the same plan. Another 35% of D-SNP enrollees were in what is known as highly integrated dual eligible (HIDE) SNPs that meet the requirements of CO plans and also require coverage of long-term services and supports and behavioral health. The remaining 8% were in plans that were fully integrated dual eligible (FIDE) special needs plans (SNPs), which coordinate care for dual-eligible individuals within a single managed care organization.

2. The share of dual-eligible individuals enrolled in D-SNPs varies by state, ranging from 5% (Nevada) to 58% (Hawaii) in 2021.

In 2021, D-SNP enrollment as a share of dual-eligible individuals ranged from 5% in Nevada to 58% in Hawaii, as compared to 29% nationwide. In seven states, 40% or more of all dual-eligible individuals were enrolled in D-SNPs: Alabama (42%), New York (42%), Florida (46%), Tennessee (46%), Arizona (47%), and Hawaii (58%), as well as in Puerto Rico (98%, not shown). Conversely, less than 10% of dual-eligible individuals were enrolled in D-SNPs in Nevada (5%), where D-SNPs were first available in 2021, and Montana (8%).

The share of dual-eligible individuals enrolled in D-SNPs more than doubled in 19 states between 2018 and 2021. States with relatively low D-SNP enrollment relative to the dual-eligible population in 2018, such as Oklahoma, Iowa, and West Virginia, had the highest percentage growth in share of dual-eligible individuals enrolled in D-SNPs between 2018 and 2021. For example, the share of dual-eligible individuals enrolled in D-SNPs in Oklahoma increased from 1% in 2018 to 12% in 2021. Several states, such as Arizona (45% vs. 47%), California (10% vs. 10%), Hawaii (57 vs. 58%), Massachusetts (16% vs. 19%), and Minnesota (30% vs. 34%), maintained similar shares of dual-eligible individuals enrolled in D-SNPs between 2018 and 2021. New Mexico, Oregon, and Utah were the only states where a smaller share of dual-eligible individuals enrolled in D-SNPs in 2021 compared to 2018. Insurers in Alaska, Illinois, New Hampshire, Wyoming, Vermont, North Dakota, and South Dakota did not offer D-SNPs in 2018 or 2021. Overall, Medicare Advantage enrollment in these states tends to be lower than the national average, partially explaining the relatively low enrollment among dual-eligible individuals in D-SNPs.

Differences in D-SNP enrollment across states may be reflective of several factors, including variation in: state-level policies pertaining to D-SNP enrollment, demographic characteristics of the dual-eligible population in the state, firm strategy, and the ability to establish care networks for dual-eligible individuals.

3. Within states, the share of dual-eligible individuals enrolled in a D-SNP varied across counties in 2021.

Within states, D-SNP enrollment among dual-eligible individuals often varies by county. For example, in Northern California, 20% of dual-eligible individuals in San Francisco County were enrolled in D-SNPs in 2021, compared to only 5% in nearby Santa Clara County. Although there was roughly the same number of dual-eligible individuals in Alameda and Sacramento Counties in 2021, only 16% of dual-eligible individuals in Alameda County were enrolled in D-SNPs, while over one-third (34%) of dual-eligible individuals in Sacramento County were enrolled in D-SNPs. In Florida, the share of dual-eligible individuals in Orlando was 54% (Pasco County) compared to 64% in nearby Tampa (Osceola County). However, in counties in the Florida Panhandle, the share of dual-eligible individuals enrolled in D-SNPs was much lower: 24% in Jackson County and 45% in adjacent Gadsden County. In Puerto Rico (not shown), enrollment of dual-eligible individuals in D-SNPs is nearly universal.

Variations in D-SNP enrollment at the county level may be attributed to several factors, such as rurality and Medicare payment rates.

4. More than 9 in 10 dual-eligible individuals (92%) lived in a county that offered at least one D-SNP in 2022.

More than 9 in 10 (92%) dual-eligible individuals could choose from at least one D-SNP when selecting coverage for 2022, up from 86% in the 2018 plan year. Most of the remaining 8% of dual-eligible individuals without access to a D-SNP lived in the 5 states where D-SNPs were not available in 2022 (Illinois, New Hampshire, South Dakota, Vermont, and Alaska).

5. The average dual-eligible individual could choose from 10 D-SNPs in 2022.

The average dual-eligible individual could choose from 10 D-SNPs in 2022 – more D-SNPs than in any previous year. The number of D-SNPs available to the average dual-eligible individual has more than doubled since 2011. While the average dual-eligible individual could choose from 10 D-SNPs, plan availability varied widely by county in 2022. For example, dual-eligible individuals in most counties composing metropolitan New York City had access to up to 40 D-SNP options, while dual-eligible individuals in neighboring Suffolk County could choose from 17 plans. Dual-eligible individuals in states such as Wyoming and Montana were able to access only one D-SNP. Dual-eligible individuals in Illinois, North Dakota, New Hampshire, Vermont, and Alaska did not have access to a D-SNP in 2022 (and will also not have access to a D-SNP in 2024).

6. About half (52%) of D-SNP enrollees were in UnitedHealthcare or Humana plans in 2023.

UnitedHealthcare and Humana were the dominant providers of D-SNPs in 2023, accounting for 52% of total D-SNP enrollment. Enrollment in UnitedHealthcare plans alone accounted for nearly two-fifths of total D-SNP enrollment (37%) and exceeded combined enrollment of BCBS, Centene, Elevance Non-Blue, CVS Health, Kaiser Permanente, and Cigna (31%) in 2023.

Between 2018 and 2023, the share of enrollees in plans offered by smaller firms has declined from 27% to 17%. Between 2018 and 2023, the share of D-SNP enrollees choosing plans offered by Kaiser Permanente, Cigna, Centene, BCBS, and Elevance Non-Blue declined, while the share of enrollees choosing UnitedHealthcare, CVS Health, and Humana plans increased.

7. Every major insurer increased the number of D-SNP offerings between 2018 and 2024.

In plan year 2024, insurers are offering 851 plans, more than double the number offered in plan year 2018 (401). The growth in offerings between plan years 2018 and 2024 can be attributed to insurers increasing offerings in existing counties and expanding the number counties in which D-SNPs are offered.

Eight major insurers are offering 75% (638) of plans in 2024, up from 68% in plan year 2018. The remaining 25% (213) of plans are offered by smaller insurers. In plan year 2024, CVS will offer 105 plans, an increase of 98 plans since the 2018 plan year. This represents the largest growth in plan offerings among the major insurers. Kaiser Permanente will offer 10 plans, an increase of 5 plans since the 2018 plan year. This represents the smallest growth in plan offerings among major insurers.

One firm, Florida Complete Care, is entering the D-SNP market in 2024 (though it has offered other types of SNPs in prior years), while 5 firms that offered D-SNPs (Ascension, Health Choice Generations Utah, AgeWell New York, Essence Healthcare, and Vantage Health Plan) are exiting the D-SNP market. This differs from prior KFF analysis which examines entries and exits for the overall Medicare Advantage market.

8. D-SNPs are more likely than other Medicare Advantage plans to offer some extra benefits such as over the counter benefits and meal benefits.

The share of D-SNPs offering eye exams and/or eyeglasses (96%), dental care (95%), fitness benefits (94%), or hearing exams and/or aids (92%) was nearly universal for enrollees in D-SNPs as well as individual Medicare Advantage plans in 2024 (Figure 8). However, D-SNPs are more likely than individual Medicare Advantage plans to offer over the counter benefits (96% vs. 85%), meal benefits (86% vs. 72%), bathroom safety devices (32% vs. 22%), and in-home support services (23% vs. 9%). D-SNPs are also more likely to offer transportation services than other Medicare Advantage plans (88% vs. 36%). Many of the additional benefits may be covered by Medicaid as “Medicaid wraparound services.” States are required to cover non-emergency medical transportation (NEMT), all states provide some home-based care (including meals, bathroom safety, or in-home supports), and many states cover some dental and vision services. D-SNPs may expand upon the services offered through Medicaid and it is unknown to what extent the D-SNP offerings supplement the Medicaid services for specific states or plans.

9. In 2021, dual-eligible individuals enrolled in a plan in a D-SNP-only contract were subject to one prior authorization request, on average.

CMS publishes prior authorization data by contract and not by plan type, which makes it impossible to document the total number of prior authorization requests or denials for D-SNP enrollees. This analysis, therefore, relies on data about contracts containing only D-SNPs, which account for about 19% of total D-SNP enrollment. Most D-SNP enrollees (81%) are in plans that are in a contract with other Medicare plan types, which means CMS does not collect or publish prior authorization data for most D-SNP enrollees.

On average, firms with contracts containing only D-SNPs received one prior authorization request per beneficiary in 2021, as compared to previous KFF analysis showing 1.5 prior authorization requests per enrollee for all Medicare Advantage plans, although that estimate included contracts that included both D-SNPs and other plans. This is somewhat surprising, considering dual-eligible individuals generally utilize more health care services than the average Medicare beneficiary.

Unlike other insurers, CVS had more prior authorization requests for enrollees in D-SNP-only contracts than for enrollees in all Medicare Advantage contracts. Prior authorizations ranged across firms from less than one per beneficiary (UnitedHealthcare) to 2.2 per beneficiary (BCBS) for contracts containing only D-SNPs.

10.Despite fewer prior authorizations in D-SNP-only plans compared to all Medicare Advantage plans, the rate of denials was twice as high.

In 2021, firms with contracts containing only D-SNPs received nearly 670,000 prior authorization requests, 12% of which were denied. This is double the denial rate for all Medicare Advantage plans (6%), according to previous KFF analysis. The rate of prior authorization request denials ranged from 5% (Humana) to 15% (CVS Health and Centene) among contracts containing only D-SNPs. Nearly 7% of just over 80,000 denials were appealed, compared to 11% of denials for all Medicare Advantage plans. Just over two-thirds (68%) of those appeals were resolved favorably, in contrast to prior KFF analysis that found over 82% of appeals for all Medicare Advantage contracts were resolved favorably.

Previous KFF analysis of Medicare Advantage prior authorization data revealed an inverse relationship between the insurer’s volume of prior authorization requests and share of requests that were denied. This relationship among D-SNP-only contracts holds for some firms, such as UnitedHealthcare and Humana, but not others. For example, Centene D-SNP-only contracts received on average 1.7 requests per enrollee, higher than the overall rate of one prior authorization request and denied 15% of those requests compared to 12% overall. As with prior authorization data, denial rates are collected and reported at the contract level, which means denial rates in D-SNP-only contracts represent a fraction of individuals enrolled in D-SNPs.

Discussion

In 2021, nearly 3 in 10 (29%) dual-eligible individuals enrolled in D-SNPs, an increase from 20% in 2018. Growth in D-SNP enrollment may be driven by several factors. D-SNPs may be attractive to dual-eligible individuals due to the availability of extra benefits, including benefits that are offered more frequently in D-SNPs than Medicare Advantage plans for general enrollment, such as over the counter benefits and meals. There is also strong interest in D-SNPs among insurers, given relatively high margins. Since 2018, more insurers have offered D-SNPs and larger insurers have offered more plans. Growth in enrollment may be due to some extent to the automatic enrollment of some individuals into D-SNPs. In 2022, more than 9 in 10 dual-eligible individuals (92%) lived in counties where insurers offered at least one D-SNP, with an average of 10 D-SNPs offered per dual-eligible individual, compared to 6 plans in 2018.

While enrollment and plan availability continue to grow, it is not clear how well D-SNPs coordinate with Medicaid to provide the full range of benefits to dual-eligible enrollees. Fewer than one in ten D-SNP enrollees are in fully integrated plans.

Relatedly, little is known about the quality of D-SNPs. Because quality ratings are reported at the contract level, rather than the plan level, it is not possible to assess the quality of D-SNPs that are included in contracts with other plans, affecting most D-SNP enrollees. Most D-SNP enrollees (81%) are in plans that are part of a contract with other plans, which means the quality ratings are at the contract level, not of their specific plan. Further, MedPAC has raised concerns that the current quality measures are not sufficient to adequately assess care delivery in D-SNPs. Early attempts at quality measurements have produced mixed results, with some reporting little variation in quality measurements between plans and others reporting little difference in care quality between D-SNPs and other methods of care for dual-eligible individuals.

Lack of data transparency also contributes to limited understanding of the impact of prior authorization requirements and denials for dual-eligible individuals enrolled in D-SNPs. KFF’s analysis suggests that contracts containing D-SNPs (but no other plans) deny prior authorization requests at a much higher rate than Medicare Advantage plans overall, even though firms with contracts containing only D-SNPs received few authorization requests per enrollee. Again, because prior authorization requests and denials are reported at the contract level, it is not possible to document the number of prior authorization requests for D-SNP enrollees overall or per person, or denial rates.

Further, although most plans offer some extra benefits to D-SNP enrollees, it is not clear how often D-SNP enrollees take advantage of these extra benefits, whether insurers offer adequate networks to access these services, and whether they are of value to a population with such diverse health needs. A recent proposed rule by the Biden Administration would require insurance providers to periodically notify D-SNP enrollees if they are not utilizing supplemental benefits offered by their plan, but there is an absence of comprehensive data pertaining to the use of supplemental benefits among D-SNP and other Medicare Advantage enrollees, according to KFF. In addition, there is some concern among advocates and policymakers about marketing benefits, such as transportation, dental, and vision when dual-eligible individuals may already be eligible for these services through Medicaid. Given the significant needs of the dual-eligible population, and incentives for rapid growth in D-SNP enrollment, including proposed policy changes to institute Special Enrollment Periods to increase D-SNP participation, greater insight into the experiences of D-SNP enrollees would be valuable for beneficiaries and policymakers.

Updated on February 9, 2024

Key Takeaways

• There are efforts underway to put constitutional amendments regarding abortion on the 2024 ballot in as many as 13 states: Arizona, Arkansas, Colorado, Florida, Iowa, Maine, Maryland, Missouri, Montana, Nebraska, Nevada, Pennsylvania, and South Dakota.
• Since Dobbs, 6 states – California, Kansas, Kentucky, Michigan, Vermont, and Ohio – have voted on abortion related constitutional amendments, and the side favoring access to abortion prevailed in every state.
• Citizen-initiated constitutional amendments are a powerful tool that is being utilized to allow voters to garner signatures to place abortion on the ballot without directly involving the legislature or the governor.
• This approach, however, cannot be used in 15 of the states that currently have abortion bans or early gestational limits on the books because they do not have a pathway for citizens to place potential constitutional amendments on the ballot. In those states, the legislature controls the process.

Since the Supreme Court’s Dobbs decision overturning Roe v. Wade, voters in 6 states have weighed in on constitutional amendments regarding abortion. Every time these constitutional amendments have been on the ballot, voters have elected to approve measures that would explicitly protect abortion and deny those that sought to limit abortion. Since then, potential ballot initiatives have captured nationwide attention, and in 2024, 13 states may have abortion measures on their ballot. These measures seek to either explicitly affirm that the state constitution protects the right to abortion or that nothing in the constitution confers such a right. Advocates on both sides are hoping that having abortion on the ballot will motivate voter turnout in this election cycle. This issue brief explains why initiatives have become so popular with advocates on both sides of the issue, reviews the current initiatives that are in progress and may appear in on state ballots the next general election, and outlines the processes states have available to them to use to place initiatives on the ballot.

Interest in Reproductive Rights Ballot Measures Post-Dobbs

State constitutional amendments have drawn much interest since SCOTUS’ decision in Dobbs because they provide more stable protections than laws or state Supreme Courts’ decisions recognizing a right to abortion. Laws are relatively easy to amend or repeal if there is a change in the make-up of the state legislature. For swing states, where neither party has a strong hold on the majority of the legislature, it is difficult to ensure that abortion protections or restrictions will not be repealed if there is a change in party control. Additionally, state Supreme Court rulings recognizing a right to abortion – or stating that there is no such right – in the state constitution may be overruled when there are subsequent changes in the make-up of the court. For example, the South Carolina Supreme Court ruled in January 2023 that the state’s 6-week ban was unconstitutional, but after the legislature appointed a new justice to replace a retiring justice, the Court reversed its ruling in August 2023 and found the restriction is constitutional. Constitutional amendments, on the other hand, would have to be repealed with a ballot measure receiving a majority of the popular vote.

Furthermore, in states where their supreme court has interpreted and recognized a right to abortion based on the state constitution, a constitutional amendment declaring there is no right to abortion would supersede the Court’s decision and allow abortion restrictions and bans to survive judicial review. For example, the goal of the 2022 Kansas ballot measure was to negate the state Supreme Court’s ruling that the Kansas constitution protects the right to abortion. If the ballot measure had passed and the Kansas constitution had been amended to state that nothing in it creates a right to abortion, the state legislature would have been able to ban or severely restrict abortion in Kansas. Conversely, in states where their supreme court has ruled the state constitution does not protect the right to abortion, amending the state constitution to include an explicit right to abortion would supersede supreme court precedent and require the court to block an abortion ban.

Importantly, aside from providing stronger, more stable protections and changing the judicial review of abortion laws, ballot measures seeking to amend state constitutions place issues directly in front of voters. In states where the legislature has sought to limit abortion in ways that are more restrictive than the public favors, a citizen-initiated ballot measure is a way to enact constitutional protections for abortion without directly involving the legislature or Governor. And once the constitution is amended, state policy makers cannot repeal these protections or fully ban or restrict abortion. Citizen-initiated ballot measures provide a direct pathway for the electorate to decide whether or not abortion should be legal in their state, regardless of how their elected representatives have decided to approach abortion policy. According to a recent KFF poll, nationally, the majority of voters say abortion should be legal in all or most cases.

Constitutional Ballot Measures in 2022 and 2023

Due to the advantages presented by ballot measures amending state constitutions, legislatures and citizen groups in several states have moved to add measures enshrining a right to abortion in their constitutions or declaring that there is no such right in the constitution. In 2022, there were four legislatively referred measures on the ballot regarding abortion or reproductive rights more broadly. In 2022, Michigan voters passed the first citizen-initiated constitutional measure protecting the right to abortion and in 2023, Ohio voters also passed a similar citizen-initiated constitutional measure protecting the right to abortion.

Constitutional Ballot Measures That May Be on the 2024 Ballot

Abortion rights groups are seeking to qualify ballot initiatives to uphold abortion rights in three states (Colorado, Montana, and Nevada) where there is no early gestational limit and in six states that ban abortion or have early gestational limits (Arizona, Arkansas, Florida, Missouri, Nebraska, South Dakota) in 2024. There is ongoing litigation challenging the abortion bans in most of these states, but without a constitutional amendment establishing a right to abortion, it is not certain that the respective supreme courts in these states will find the bans unconstitutional, or in the case of Florida, that the Court will uphold its prior decisions recognizing a right to abortion. However, if the initiatives establishing a right to abortion succeed in these states, abortion-rights advocates would have the certainty that these bans will be blocked in court.

Four state legislatures have constructed potential initiatives for constitutional amendments pertaining to abortion. The legislatures in Iowa and Pennsylvania have introduced measures that would amend the state constitution to declare that it does not protect the right to abortion. In Maine and Maryland, state representatives drafted initiatives protecting the right to reproductive freedom or autonomy (which includes the right to abortion). The Maryland initiative passed in the legislature and so far, is the only measure that is certain to appear on the 2024 ballot. The remaining legislatively-referred initiatives must pass in both chambers of the state legislatures, and the citizen-initiated measures are either gathering signatures, or awaiting initial approval or signature certification. See Appendix A for state-by-state details.

Pathways to Getting an Initiative on the Ballot

Legislatively-Referred Initiatives

For legislatively-referred initiatives seeking to amend a state constitution by the electorate, state legislators draft language for the initiative and introduce it to the legislature, in a process that is similar to how they introduce language for bills. Both chambers of the state’s legislature must vote to approve the language before it is placed on the ballot. In every state except for Delaware, initiatives seeking to amend a state’s constitution must be placed on the ballot and cannot be enacted without voter approval.

The process of approving a ballot initiative in the state legislature differs from state to state. While most states require the initiative to pass only once in the legislature, some states require it to pass over two successive legislative sessions. States also have differing requirements for what percentage of the legislature must vote in favor of the initiative for it to move forward, ranging from a simple majority to a two-thirds majority. Once the measure is on the ballot, there are also differing requirements for what share of the votes it must receive to become law.

Citizen-Initiated Measures

Citizens are allowed to propose a constitutional amendment for the ballot in 17 states. The rules for doing so differ slightly from state to state, but generally, citizen groups must submit a draft of the proposed amendment and ballot title to a government official – usually the Secretary of State – for approval. Once the petition has been approved, groups must gather signatures in support of their measure, often with distribution requirements across the states’ congressional districts, and submit their support signatures for validation. If enough signatures from a sample are deemed valid, the measure is cleared to be on the ballot, where it must receive anywhere from a simple majority to 60% of the vote in favor to be approved.

There are different stages of review for ballots in each state. Some states require that ballot measures address only one subject, and the reviewing body or official might reject a petition if they decide it violates this requirement. Petitions may also be rejected if they contain vague or confusing language. However, beyond these considerations, state officials are generally not allowed to reject a petition based on its substance. In the past few years, this has meant that state attorneys general, secretaries of state, or electoral boards that oppose abortion have nevertheless had to approve petitions for ballot measures that would enshrine the right to abortion in the state’s constitution. This has also resulted in state legislatures taking steps to attempt to make it more difficult for measures to pass or even get on the ballot in the first place. For instance, in Ohio, Missouri, Florida, Oklahoma, and Utah, these attempts have included proposed measures to increase the percentage of the vote needed for the ballot initiative to pass. Lawmakers in Missouri and North Dakota have additionally attempted to increase the number of signatures needed to get a measure approved. The North Dakota 2024 constitutional ballot measure additionally seeks to require that constitutional ballot initiatives pass in two separate elections (the primary and general elections) before they can successfully amend the constitution.

Citizen-Initiated Ballot Measures Are Not an Option in All States

Not every state has a pathway for a citizen-initiated constitutional amendment. In twelve states that currently ban abortion or have early gestational limits in effect, there is no process for a citizen-initiated ballot measure. Three additional states (Wyoming, Iowa, and Utah) that have abortion bans currently blocked by courts, also have no process for a citizen-initiated ballot measure. In these fifteen states, unless the state legislatures repeal their bans, the only avenue reproductive rights supporters have to potentially change the legal status of abortion, short of electing pro-choice legislators and policy makers, is to challenge these bans in court. If the State Supreme Court upholds the bans, they will remain in place. This is the current situation in Idaho, Indiana, and South Carolina. Their respective supreme courts ruled their state constitutions do not protect the right to abortion and that the states’ abortion bans are constitutional. Since these rulings, abortion-rights advocates have not had any avenues to change the legal status of abortion in these states. Litigation challenging abortion bans in seven other states (Georgia, Iowa, Kentucky, Louisiana, North Carolina, Utah, Wyoming) with bans and no process for a citizen-initiated ballot measure is making its way through the courts. There are no legal challenges to the bans in Alabama, Mississippi, or West Virginia.

Moreover, three states with abortion bans in place and no process for a citizen-initiated ballot measure (Louisiana, Tennessee, and West Virginia), have state constitutional amendments that explicitly state their constitutions do not protect the right to abortion. These amendments preclude these states’ Supreme Courts from ruling that an abortion ban is unconstitutional because it violates a right to abortion. In other words, the only way for the legal status of abortion to change in these states is for their legislatures to repeal their bans or pass a new constitutional amendment protecting the right to abortion. In two of these states (Louisiana and Tennessee) and in Idaho, Indiana, and Texas, there is ongoing litigation focused only on whether the exceptions to the bans are constitutional, but not challenging the underlying bans

Looking Ahead

When the Supreme Court’s decision in Dobbs gave states the ability to ban or limit abortion, the legal landscape at the state level was activated as never before. In many states, legislators moved quickly to pass new statutes banning, or severely limiting access to abortion, and reproductive rights supporters turned to courts to challenge many of these new and already-existing laws. As the legality of and restrictions of abortion has been debated in state legislatures and courthouses, ballot measures amending state constitutions have emerged as a potential tool for proponents on both sides of the issue. State constitutional amendments explicitly stating whether the state’s constitution protects the right to abortion offer more stable protections for individuals’ right to abortion or the state’s right to ban abortion. These initiatives are being closely watched and it remains to be seen whether they will have an impact on voter turnout, as well as the outcome of candidate races.

Additionally, as the post-Dobbs legal landscape has shifted and begun to settle, many states have abortion laws on the books that do not reflect popular sentiment around abortion. Citizen-initiated constitutional ballot measures allow citizens to put abortion rights directly before the voters, although this is not an option in every state. Still, in many of the states where this is an option, advocacy groups on both sides of the issue are seeking to use this powerful tool to address the legality of abortion in their state.

Appendix

A KFF analysis shows that a new out-of-pocket spending cap in Medicare Part D could translate into savings for well over 1 million beneficiaries when it takes effect next year, including more than 100,000 people each in California, Florida and Texas, based on analyses of drug spending in 2021.

The $2,000 cap, part of the Inflation Reduction Act of 2022, will lead to thousands of dollars in savings for Medicare patients who take high-cost drugs for cancer, rheumatoid arthritis, and other serious conditions. This new limit follows the elimination this year of a longstanding requirement that Part D enrollees pay 5% of their drug costs out-of-pocket after their drug expenditures reach a certain threshold. 

Based on KFF’s review of Part D drug claims data, if the cap been in place in 2021, 1.5 million Medicare beneficiaries would have benefited because their out-of-pocket costs for prescription drugs exceeded $2,000. Of the total 1.5 million, about 200,000 Medicare beneficiaries spent $5,000 or more for their prescriptions that year, while another 300,000 expended between $3,000 and $5,000. The rest spent between $2,000 and $3,000. 

Moreover, the number of people who will see savings from the cap will rise over a longer period of time. A total of 5 million Part D enrollees had out-of-pocket drug costs of $2,000 or more in at least one year during the 10-year period ending in 2021, for instance.

In most states, tens of thousands of Medicare beneficiaries could save money from the new cap next year. In six states — New York, Pennsylvania, Ohio, Illinois, North Carolina, and New Jersey — between 50,000 and 82,000 beneficiaries spent more than $2,000 out-of-pocket for prescription drugs in 2021. The numbers were higher in California, Florida, and Texas, where more than 100,000 Part D enrollees exceeded the threshold that year. 

In 2025, Medicare beneficiaries will pay no more than $2,000 out of pocket for prescription drugs covered under Part D, Medicare’s outpatient drug benefit. This is due to a provision in the Inflation Reduction Act of 2022, which included several changes to the Medicare Part D program designed to lower patient out-of-pocket costs and reduce what Medicare spends on prescription drugs. This new $2,000 cap (indexed annually to the rate of change in Part D costs) comes on top of the elimination of 5% coinsurance in the catastrophic coverage phase of the Part D benefit, in effect for 2024, which translates to a cap of about $3,300 out of pocket for brand-name drugs. These benefit design changes will save thousands of dollars for people who take high-cost drugs for cancer, rheumatoid arthritis, and other serious conditions.

If a $2,000 cap on out-of-pocket drug spending had been in place in 2021, 1.5 million Medicare beneficiaries enrolled in Part D plans would have saved money because they spent $2,000 or more out of pocket on prescription drugs that year. This estimate is based on KFF analysis of Medicare Part D prescription drug claims data for enrollees without Part D low-income subsidies in 2021 (the most recent year available for this analysis). Among these 1.5 million enrollees, most (1.0 million or 68%) spent between $2,000 and $3,000 out of pocket, while 0.3 million (20%) had spending of $3,000 up to $5,000, and 0.2 million (12%) spent $5,000 or more out of pocket.

Over the course of several years, however, far more Part D enrollees will stand to see savings from this new out-of-pocket spending cap than in any single year. A total of 5 million Part D enrollees had out-of-pocket drug costs of $2,000 or more in at least one year during the 10-year period between 2012 and 2021, while 6.8 million Part D enrollees have paid $2,000 or more out of pocket in at least one year since 2007, the first full year of the Part D program (Figure 1).

In most states, tens of thousands, if not hundreds of thousands, of Medicare beneficiaries will feel relief from the new Part D out-of-pocket spending cap (Table 1). In California, Florida, and Texas, more than 100,000 Part D enrollees faced out-of-pocket costs of $2,000 or more in 2021, and in another 6 states (New York, Pennsylvania, Ohio, Illinois, North Carolina, and New Jersey), between 50,000 and 82,000 did so. As at the national level, more Part D enrollees in each state will benefit over time. For example, in Iowa, Louisiana, and Maryland, 73,000 Part D enrollees faced out-of-pocket costs of $2,000 or more in at least one year between 2012 and 2021. In Michigan, New Jersey, and Georgia, 148,000, 158,000, and 159,000 Part D enrollees, respectively, spent $2,000 or more in at least one year over this same 10-year period. In Texas, 364,000 Part D enrollees did so; in Florida and California, around 400,000 enrollees or more.

Capping out-of-pocket spending will help Part D enrollees with relatively high drug costs, which may include only a relatively small number of Part D enrollees in any given year but, as this analysis shows, a larger number over time. People who will be helped include those who have persistently high drug costs over multiple years and others who have high costs in one year but not over time. While a cap on out-of-pocket costs will help millions of Part D enrollees over time, higher plan costs to provide the Part D benefit could also mean higher plan premiums, a dynamic that the Inflation Reduction Act’s premium stabilization provision was designed to mitigate. Although KFF polling shows that a relatively small share of older adults is aware of the Inflation Reduction Act’s $2,000 cap on out-of-pocket drug costs for Part D enrollees that takes effect in 2025, millions of them will benefit from this cap in the years to come.

This post was updated to clarify that less than 10% of the nation’s total health spending is spent on retail prescription drugs and does not include spending on drugs administrated by physicians or in hospitals.Prescription drug costs are a top concern for the American public. While retail prescription drugs represent less than 10% of total U.S. health spending and are not the primary driver of the nation’s high health costs, Americans often pay more for the same prescription drugs than people in other countries spend. Ahead of Thursday’s expected Senate HELP Committee testimony from the CEOs of three major pharmaceutical companies, here are three charts about drug prices in the United States.

1. More than 6 in 10 adults (63%) say that drugs developed over the past 20 years have improved the lives of Americans, though an even larger majority (82%) say drug prices are unreasonable. That may explain why majorities of Republicans, independents and Democrats say there is too little regulation of drug prices and support a wide range of policy actions to rein them in.

2.    List prices for brand-name drugs are often much higher in the U.S. than in other large, wealthy countries. Net prices, which take into account rebates and coupons, also are often higher than in other countries, though it’s hard to say how much due to the lack of transparency in the amount of the rebates. One example: The list prices for the new wave of weight-loss drugs are often more than 4 times higher in the U.S. than in other countries. American paying for these drugs without any help from their health insurance can get Wegovy coupons worth up to $500 per month, but that would still leave them paying more than double the list prices in Germany or the Netherlands.

3.    For the first time, the federal government is negotiating drug prices on behalf of the Medicare program and its beneficiaries as authorized by the Inflation Reduction Act of 2023. Medicare collectively spent about $50 billion in one year on the 10 drugs initially selected for negotiations – drugs that treat diabetes, blood clots, heart failure, psoriasis, rheumatoid arthritis, Crohn’s disease, and blood cancers. The total suggests that there could be significant savings in 2026 and future years as a result of these negotiations. Some affected drug manufacturers and other industry groups are challenging the negotiations process on both constitutional and procedural grounds that, if successful, could end or alter the process.

Learn more:

• Public Opinion on Prescription Drugs and Their Prices
• How do prices of drugs for weight loss in the U.S. compare to peer nations’ prices? 
• FAQs about the Inflation Reduction Act’s Medicare Drug Price Negotiation Program 
• How has U.S. spending on health care changed over time? 
• What to Know About the FDA’s Recent Decision to Allow Florida to Import Prescription Drugs from Canada
• The New Help for Medicare Beneficiaries with High Drug Costs That Few Seem to Know About

About 1 in 5 adolescents report symptoms of anxiety or depression, according to a KFF analysis of a new federal survey of teen health.

While some teens are getting mental health care, a significant share say they are not receiving the therapy they need due to costs, fear of what others will think, and/or not knowing how to get help.

Data from the recently released Teen National Health Interview Survey from July 2021 to December 2022, reveals that 21% of adolescents ages 12-17 report experiencing symptoms of anxiety in the past two weeks, and 17% said they had symptoms of depression. 

Female adolescents were more than twice as likely as their male peers to report feelings of anxiety (31% vs. 12%) and depression (25% vs. 10%) during the survey period. Rates were highest among LGBT+ adolescents, with 43%reporting symptoms of anxiety and 37% saying they had symptoms of depression.

The results of the survey — unique in that they represent direct responses from adolescents themselves, rather than from their parents or guardians – come at a time of heightened concern about the state of American teenagers’ mental health.

The analysis also examines data on the rates of adolescent drug overdoses, suicide and self-harm, by race/ethnicity and sex.  It also examines access to therapy and the share of adolescents who report negative experiences, such as bullying.

While studies have documented rising mental health problems among adolescents in the U.S. for years, the trend was exacerbated by the COVID-19 pandemic. Since 2021, the U.S. Surgeon General has twice issued advisories about the challenges to youth mental health, including the threat posed by excessive social media use. 

Among the other key takeaways from KFF’s analysis:

• Although some adolescents received mental health care, 20% reported not receiving the mental health therapy they needed because of cost, fear of what others would think, and/or because they didn’t know how to get help. This lack of needed therapy or counseling was more pronounced among female (32%) and LGBT+ adolescents (38%).
• Deaths due to drug overdose among adolescents have more than doubled in recent years — fueled by the rise of the synthetic opioid fentanyl — increasing from 253 deaths in 2018 to 722 deaths in 2022. The largest increases in these overdose deaths were among Hispanic and Black adolescents.
• Ninety-two percent of adolescents reported at least two hours of weekday screen time not associated with schoolwork. Research has found that social media use may be associated with poor well-being among youth, with a higher risk of depression for female adolescents. 
• Many adolescents reported enduring negative experiences such as bullying (34%), emotional abuse by a parent (17%), and neighborhood violence (15%) in 2021 and 2022, all of which can influence mental health.

The full analysis, Recent Trends in Mental Health and Substance use Concerns Among Adolescents, is available here.

Also recently released by KFF are two related analyses: One takes stock of the latest efforts in the states to combat the opioid epidemic and another describes the supply and characteristics of substance use and mental health treatment facilities across the U.S.

• As Congress discusses reauthorizing the expired 2018 SUPPORT Act, state Medicaid programs are taking active steps to address the opioid epidemic. State Medicaid efforts include improving access to OUD treatment medications, covering over the counter Narcan, and extending OUD treatment to soon-to-be-released inmates–a population at high risk of overdose death upon release. 
• Low treatment rates for substance use and mental health conditions may be due, in part, to limited treatment options and workforce shortages. In the U.S., there are about 14,700 substance use and 9,500 mental health treatment facilities. Most offer outpatient services, while fewer offer more intensive inpatient or residential care. Facility ownership and insurance participation vary—with private ownership more common in substance use treatment and notable state-by-state variation in many areas. 

Originally published in February 2023, this brief was updated on Feb. 6, 2024, to include more recent information.

In response to the unprecedented nature of the COVID-19 pandemic, the federal government spent billions of dollars in emergency funds between 2020 and 2022 to purchase medical countermeasures – vaccines, including boosters, treatments, and tests – and provided them free of charge to the public. In addition, Congress enacted several bills¹  that included special requirements for their coverage by both public and private insurers, and the administration issued guidance²  and regulations to protect patient access and promote equitable distribution. The effective dates of many, though not all, of these requirements were tied to the public health emergency (PHE) declaration made pursuant to Section 319 of the Public Health Service Act, first declared in January of 2020 and renewed every 90 days through February 9, 2023, effectively ending the PHE on May 11, 2023. The availability of federally-purchased COVID-19 medical countermeasures ensured that they were provided free to all.

Now that the PHE has ended, the federal government has been transitioning these products to the commercial market, where their provision and availability varies by insurance status and other factors. In some cases, access will be curtailed while in others, the federal government is continuing some programs to assist those with limited coverage. This document provides a side-by-side comparison of how vaccines, treatments, and tests were provided during the PHE and with a federally purchased supply compared to the current situation, organized by payer.

Table 1: COVID-19 VaccinesTable 2: COVID-19 TreatmentsTable 3: COVID-19 Tests

(Back to top)

(Back to top)

(Back to top)

1. Families First Coronavirus Response Act (FFCRA), https://www.congress.gov/bill/116th-congress/house-bill/6201 and KFF summary of FFCRA, https://modern.kff.org/coronavirus-covid-19/issue-brief/the-families-first-coronavirus-response-act-summary-of-key-provisions/; The Coronavirus Aid, Relief, and Economic Security (CARES) Act, https://www.congress.gov/bill/116th-congress/house-bill/748 and KFF summary of the CARES Act, https://modern.kff.org/coronavirus-covid-19/issue-brief/the-coronavirus-aid-relief-and-economic-security-act-summary-of-key-health-provisions/; The American Rescue Plan Act of 2021 (ARPA), https://www.congress.gov/bill/117th-congress/house-bill/1319/text and KFF brief on ARPA Medicaid provisions, https://modern.kff.org/medicaid/issue-brief/medicaid-provisions-in-the-american-rescue-plan-act/; The Inflation Reduction Act of 2022(IRA), https://www.congress.gov/bill/117th-congress/house-bill/5376/text and KFF summaries of health provisions in the IRA, https://modern.kff.org/medicare/understanding-the-health-provisions-in-the-senate-reconciliation-legislation/, and Medicaid changes, https://modern.kff.org/policy-watch/medicaid-and-the-inflation-reduction-act-of-2022/. ↩︎
2. See, for example, https://www.medicaid.gov/resources-for-states/coronavirus-disease-2019-covid-19/other-agency-guidance/cms-guidance/index.html and https://www.dol.gov/sites/dolgov/files/EBSA/about-ebsa/our-activities/resource-center/faqs/aca-part-51.pdf. ↩︎