What the Outcome of the Election Could Mean for Medicaid

Published: Sep 6, 2024

While Medicaid policy may not be a central issue candidates are talking about, the outcome of the election could have major implications for Medicaid, the primary program providing comprehensive coverage for over one in five low-income Americans. Medicaid is jointly financed by the federal government and states and administered by states within broad federal guidelines. Medicaid represents nearly one in five dollars spent on health and disproportionately finances safety-net hospitals and clinics, behavioral health care, and long-term services and supports (LTSS) in the community and nursing homes for seniors and people with disabilities. Medicaid has become a major part of our health care system, with two-thirds of Americans having a connection to Medicaid. Public opinion polling shows Medicaid has broad support across political parties.

This brief highlights major policy differences under the Biden-Harris Administration and the Trump Administration on key Medicaid topic areas. The brief examines the records and proposed budgets under President Biden and Vice President Harris and former President Trump as well as major legislative proposals that former President Trump supported. Budget proposals often include policy proposals that would require Congressional action to enact. The 2024 Republican Party platform calls for new affordable health care and prescription drug options with protections for Medicare and seniors but does not include any details about Medicaid policy. To help inform the range of options that could be considered under a Trump presidency, this brief also examines Medicaid policies included in the Republican Study Committee (RSC) FY 2025 budget and the Heritage Foundation’s Project 2025. Former President Trump has not endorsed these proposals, though there would likely be a push for them from conservative groups and Republicans in Congress if Trump wins the election. Policy differences are highlighted across the following key Medicaid topic areas, with more detail about candidate records and specific sources in the Appendix.

The analysis finds that Vice President Harris and former President Trump have vastly different records, views on, and policy proposals for Medicaid (Figure 1). More broadly, as KFF President and CEO Drew Altman recently wrote, “Republicans and Democrats have fundamentally different world views of the program.”

Biden-Harris Administration policy proposals have generally focused on efforts to “protect and strengthen Medicaid and the Affordable Care Act (ACA)” by expanding health coverage and improving continuity of coverage, reducing the rate and number of uninsured, expanding access to care, and reducing health disparities. These policies generally maintain or increase federal Medicaid spending.

Former President Trump has not talked specifically about Medicaid in the current campaign, but his record as president and proposals from Republican groups include plans to repeal or weaken the ACA, cap and reduce Medicaid financing, and restrict Medicaid eligibility, with the overarching goal of reducing federal spending on Medicaid. Estimates of savings range widely, from a $4.5 trillion reduction over 10 years (over a 50% reduction) under the RSC FY 2025 budget proposal, to earlier Congressional Republican proposals that would have reduced Medicaid spending by 25% or one-third, to other proposals with smaller overall reductions but large shifts in spending across states. Reductions in federal spending generally shift risk and costs to states as well as enrollees and would likely increase the number of uninsured people.

Figure 1 is titled "What the Outcome of the Election Could Mean for Medicaid" and shows the diverging stances of Biden-Harris administration and Trump-Republican proposals

Key Takeaways

Vice President Harris inherits the record of the current administration but has also independently focused on reproductive rights and policies aimed at reducing maternal mortality and morbidity. Major Biden-Harris Administration efforts across key Medicaid topic areas include:

  • ACA Medicaid Expansion: The Biden-Harris Administration implemented policies to strengthen the ACA, including enacting legislation with an additional fiscal incentive for states to adopt Medicaid expansion. The administration also proposed a plan to close the coverage gap in states that have not expanded Medicaid, but it did not pass Congress.
  • Financing: The administration built on the existing Medicaid financing framework by expanding federal matching funds for priority areas, such as incentivizing states to expand Medicaid home and community-based services (HCBS).
  • Eligibility, Benefits, and Cost-Sharing: The Biden-Harris Administration enacted legislation to expand access to health insurance coverage by requiring 12-month continuous eligibility for children and creating an option for states to extend postpartum coverage to 12 months as well as cover certain justice-involved youth. During the unwinding of the Medicaid continuous enrollment provision, the administration issued guidance and provided states with additional waiver flexibilities to help eligible enrollees retain coverage.
  • Prescription Drugs: The Inflation Reduction Act (IRA), enacted during the Biden-Harris Administration, included several provisions to lower prescription drug costs and reduce drug spending by the federal government, including allowing the federal government to negotiate prices for some drugs in Medicare and requiring that drug manufacturers pay rebates for price increases above inflation in Medicare. Within Medicaid specifically, the administration has enacted policies (as well as proposed additional policies that did not pass) to expand and strengthen the Medicaid drug rebate program (MDRP).
  • Long-Term Services and Supports: The administration enacted legislation to increase federal funding for Medicaid HCBS and issued regulations to increase access to Medicaid HCBS, promote higher payment rates for home care workers, and establish minimum staffing requirements in nursing facilities. President Biden and Vice President Harris have called for additional, permanent federal funding for HCBS, which did not pass.
  • Managed Care: The Biden-Harris Administration issued updated Medicaid managed care regulations aimed at improving access to care, quality, health outcomes, and health equity as well as took action to improve Medicaid managed care monitoring, oversight, and transparency.

Specific Republican policies, both during and after the Trump Administration, across key Medicaid topic areas include:

  • ACA Medicaid Expansion: The Trump Administration proposed and supported unsuccessful efforts to repeal and replace the ACA, including Medicaid expansion. More recent Republican proposals, the RSC FY 2025 budget and Project 2025, do not specify a complete repeal but seek to reduce the higher match rate for the Medicaid expansion group and eliminate expanded Marketplace subsidies.
  • Financing: The Trump Administration proposed and supported legislative proposals to limit federal Medicaid spending by restructuring Medicaid financing into a block grant or per capita cap. The RSC FY 2025 budget and Project 2025 also suggest a block grant or cap as well as changes in federal matching payments.
  • Eligibility, Benefits, and Cost-Sharing: The Trump Administration approved waivers that included work requirements as a condition of Medicaid eligibility, premiums, and other eligibility restrictions. The RSC FY 2025 budget and Project 2025 also propose restricting Medicaid eligibility, benefits, and continuity of coverage, including actions such as imposing work requirements, eliminating minimum eligibility thresholds, and providing state flexibility to redesign and eliminate certain benefits.
  • Prescription Drugs: The Trump Administration aimed to address high drug costs by allowing the importation of drugs from Canada and encouraging states to submit waivers to create closed formularies. Both the RSC FY 2025 budget and Project 2025 seek to repeal the IRA, citing its implications for drug innovation.
  • Long-Term Services and Supports: The Trump Administration reduced federal oversight over nursing facilities, but more recent Republican proposals give little indication of LTSS policy priorities.
  • Managed Care: The Trump Administration changed Medicaid managed care rules, including relaxing Medicaid managed care plan network adequacy rules and requirements for beneficiary protections. Project 2025 proposes to reform and increase oversight of Medicaid managed care but doesn’t provide specifics.

Details of Medicaid Policies and Proposals

Affordable Care Act (ACA) Medicaid Expansion

Context: The ACA expanded Medicaid coverage to nearly all adults up to 138% of the federal poverty level (FPL) ($20,783 for an individual in 2024), but it is effectively an optional policy for states following a ruling by the Supreme Court. As of summer 2024, 41 states including DC have expanded Medicaid while 10 states have not, leaving 1.5 million uninsured people in the “coverage gap” (adults who have incomes above their state’s eligibility for Medicaid but below poverty, making them ineligible for subsidies in the ACA Marketplaces). The ACA expansion group is financed with a 90% federal match rate or “FMAP”, so states pay 10% of the costs for the expansion group. Data from June 2023 show almost 24 million adults were enrolled in Medicaid through the ACA expansion eligibility pathway. Total spending for this population in fiscal year 2022 was over $155 billion, including $138 billion in federal spending and $17 billion in state spending. The ACA also created health insurance exchange markets, or Marketplaces, which offer subsidized coverage to those who are not eligible for Medicaid and who do not have access to other affordable coverage.

Biden-Harris: The Biden-Harris Administration enacted legislation to strengthen the ACA, including an additional fiscal incentive for states to adopt the ACA Medicaid expansion and temporary enhanced subsidies for Marketplace coverage. The Biden-Harris Administration has also taken action to increase funding for Marketplace outreach and navigators and has supported legislation to close the coverage gap (that never passed). Four new states have adopted Medicaid expansion since Biden and Vice President Harris took office, and the number of people with Marketplace coverage has grown significantly under the administration. The Biden-Harris Administration’s latest budget builds on these actions by proposing to make the enhanced Marketplace subsidies permanent and continuing to call for closing the coverage gap. As a part of a recent campaign speech focused on economic issues in North Carolina, Vice President Harris also highlighted health proposals including extending the enhanced Marketplace subsidies.

Trump and Republican Proposals: Former President Trump modeled proposals included in his 2019 and 2020 budgets on the Graham-Cassidy amendment, which called for repealing and replacing the ACA (the FY 2021 budget was more vague about health reform proposals). Other legislation introduced but not enacted during the Trump presidency — the American Health Care Act (AHCA) and the Better Care Reconciliation Act (BCRA) — also called for repealing and replacing the ACA. More recently released Republican proposals do not specify a complete repeal of the ACA, but the RSC FY 2025 budget proposes to eliminate the expanded ACA subsidies and reduce the FMAP for the Medicaid expansion group. Project 2025 also proposes “a fairer and more rational match rate” for the Medicaid expansion group.

Tradeoffs: The Biden-Harris Administration proposals would support expanded coverage but increase federal spending. Republican proposals would eliminate or severely limit access to affordable health insurance coverage for low-income adults but would reduce federal spending. Rolling back the ACA Medicaid expansion could also have implications for state fiscal conditions and providers. The federal government pays the vast majority of expansion costs, and the literature shows that states expanding Medicaid under the ACA have realized budget savings, revenue gains, overall economic growth as well as observed positive effects on the finances of hospitals and other health care providers. Recent KFF polling finds that more than three in four (78%) enrollees and two-thirds of all adults in a non-expansion state say their state should expand Medicaid to cover more low-income uninsured people. Among the public overall, expanding Medicaid is largely split along partisan lines.

(Back to top)

Financing

Context: Medicaid is an entitlement program, meaning individuals who meet eligibility requirements are guaranteed coverage and states are guaranteed federal matching funds without a cap for qualified services provided to eligible enrollees. The FMAP (federal match rate) for most Medicaid enrollees is determined by a formula in the law that provides a match of at least 50% and provides 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%). Medicaid also provides “disproportionate share hospital” (DSH) payments to hospitals that serve a large number of Medicaid and low-income uninsured patients to offset uncompensated care costs. Medicaid represents almost $1 out of every $5 spent on health care in the U.S. and provides significant financing for hospitals, community health centers, physicians, nursing homes, and community-based LTSS.

For state budgets, Medicaid is a spending item and at the same time the largest source of federal revenues. To help finance the state share of Medicaid, states can use funding from local governments or revenue collected from provider taxes and fees to help finance the state share of Medicaid within certain limits and rules.

Biden-Harris: The Biden-Harris Administration has built on the existing financing framework by expanding federal matching funds for priority areas, such as incentivizing states to expand Medicaid or Medicaid home and community-based services. The administration’s latest budget proposal (along with prior budgets) keeps the Medicaid financing system as is under federal law and also provides funding to improve program integrity and prevent fraud.

Trump and Republican Proposals: The Trump Administration and other Republican proposals examined aim to significantly limit federal Medicaid spending by restructuring Medicaid financing into a block grant or a per capita cap where states would receive a pre-set amount of funding for Medicaid in total or per enrollee. Typically, a base year of Medicaid spending would be established and then the aggregate or per enrollee caps would increase by a specified amount each year, typically tied to inflation or inflation plus some percentage. To generate federal savings, the total amount of federal spending would be less than what is expected under current law. Program costs in excess of the total or per enrollee caps would not be matched by the federal government, leaving states to cover these costs or reduce Medicaid spending. Proposals that cap federal Medicaid funding could also eliminate the entitlement to coverage or federal minimum requirements for eligibility and benefits. The Trump Administration FY 2020 budget proposed to convert federal Medicaid funding to a per capita allotment; this change along with other health care adjustments totaled $1 trillion in spending reductions relative to estimated spending under current law over 10 years. More recent Republican proposals, the RSC FY 2025 budget and Project 2025, call for capped Medicaid spending as well as a match rate of 50% for all eligibility groups and services (RSC FY 2025 budget) or a “blended” match rate (Project 2025).

Proposals that fundamentally change the financing structure and entitlement nature of the program would require legislative change and would have broad implications across all areas of Medicaid; however, even without legislative changes, presidential administrations have options to issue regulations or guidance as well as approve demonstration waivers which could result in incremental program changes. In lieu of major legislative changes to Medicaid financing, former President Trump also issued guidance and encouraged states to apply for Healthy Adult Opportunity waivers that offered states “extensive flexibility” to cover certain groups of enrollees in exchange for limits in federal financing and access to shared savings for spending below the set cap; however, no states were approved for such waivers. In addition to restructuring Medicaid financing, Republican proposals also include provisions to restrict the use of provider taxes, combat Medicaid improper payments, and improve program integrity.

Tradeoffs: Republican proposals would substantially reduce federal Medicaid spending and allow states additional flexibility to administer their programs. KFF analysis has found that Medicaid financing caps would also limit the ability of states to meet changing needs and demands (e.g. changing state demographics, emerging high-cost drugs, etc.), and could lock-in current state spending patterns that reflect historic Medicaid policy choices. States would likely have to make choices to cover fewer people, cut benefits, and/or lower payment rates for providers (though Medicaid is typically already the lowest payer), leaving more people uninsured and reliant on safety net providers or unable to access care. While states may gain additional flexibility to administer their programs, these new options are not likely to make up for significant cuts in federal spending. States would likely face challenges to maintaining coverage, with certain states (like those with higher health care costs or needs) more at risk. Proposed changes to the FMAP floor or the ACA match rate could reduce federal spending but have disparate effects on states. States with higher per capita incomes could be most affected by calls to lower the FMAP floor while states with lower per capita incomes could be most affected by proposals to have a uniform match rate of 50%. Across many Republican proposals, expansion states would fare worse than non-expansion states with proposals to reduce the expansion match.

Recent KFF polling found that nearly 9 in 10 (86%) Medicaid enrollees want Medicaid to largely continue as it is today, while one in ten (14%) support changing Medicaid to cap federal funding and give states greater flexibility in designing their programs. Somewhat fewer adults overall – though notably still a large majority – say the same (86% vs. 71%). KFF polling has also found that majorities of voters are worried about fraud, waste, and abuse in Medicaid, Medicare, and Social Security as well as in private health insurance plans. Both parties support reducing waste, fraud, and abuse and focusing on oversight of managed care plans and providers; however, there was more of a focus on reducing errors in eligibility determinations during the Trump presidency compared to a focus on oversight of providers and managed care plans during the Biden-Harris Administration.

(Back to top)

Eligibility, Benefits, and Cost-Sharing

Context: Medicaid provides financing for a variety of groups and for a wide range of services. To participate in Medicaid and receive federal matching funds, states must meet minimum federal requirements related to who is eligible and what services must be covered. States must provide certain mandatory benefits (e.g., hospital, physician, and nursing home services) to mandatory populations (e.g., children and low-income pregnant people) without waiting lists or enrollment caps. States may also receive federal matching funds to cover “optional” services (e.g., adult dental care and home and community-based services ) or “optional” groups (e.g., children and pregnant people with income above the limits established for mandatory populations). States do have the flexibility to charge premiums and cost-sharing for some enrollees, though federal law limits the extent to which states can impose these measures.

Section 1115 Medicaid demonstration waivers offer states an avenue to test new approaches in Medicaid that differ from what is required by federal statute. While there is great variation in how states have used waivers over time, waivers generally reflect priorities identified by states and the Centers for Medicare & Medicaid Services (CMS). States may obtain “comprehensive” Section 1115 waivers that make broad changes in Medicaid eligibility, benefits, provider payments, and other program rules; other waivers may be narrower and address specific populations or benefits.

Biden-Harris: The Biden-Harris Administration enacted the 2023 Consolidated Appropriations Act, which required states to implement 12-month continuous eligibility for children, made permanent the option for states to extend postpartum coverage to 12 months, and provided states with the option to cover certain justice-involved youth. Vice President Harris has called for reducing maternal mortality and morbidity, including encouraging states to adopt the postpartum Medicaid coverage extension. The Biden-Harris Administration also finalized a rule that aligns and streamlines enrollment and renewal requirements for most individuals in Medicaid; creates timeliness requirements for redeterminations of eligibility; eliminates access barriers for children enrolled in CHIP, such as waiting periods; and facilitates transitions between Medicaid, CHIP, and Marketplace coverage. During the unwinding of the Medicaid continuous enrollment provision, the administration issued guidance and identified a range of strategies and waiver flexibilities states could adopt to limit coverage losses among those who remained eligible.

For the 2025 budget year, the Biden-Harris Administration proposes to require 12 months of postpartum Medicaid coverage, expand state options to provide continuous eligibility for children for multiple years, and eliminate fees and premiums in separate CHIP programs. Further, the Biden-Harris Administration continues to encourage waivers to expand coverage, reduce health disparities, address the social determinants of health, and help individuals transition out of incarceration.

Trump and Republican Proposals: Consistent with the Trump Administration and older legislative proposals, recent Republican proposals would make a number of changes that limit Medicaid eligibility, benefits, and continuity of coverage as well as impose work requirements for enrollees. More specifically, both the RSC FY 2025 budget and Project 2025 propose work requirements, making changes to eligibility thresholds (like eliminating minimum income eligibility for children), establishing flexible accounts (like health savings accounts) to pay for health care, and providing state flexibility to redesign and eliminate certain benefits. Project 2025 also proposes targeted time limits or lifetime caps on benefits, requiring more frequent eligibility redeterminations, and increasing premiums and cost-sharing for some individuals. Capping federal funding, as all Republican proposals examined support, would also have implications for the entitlement (guarantee of coverage for eligible individuals that is part of current law) though none of the proposals clearly note what would become of the entitlement or minimum eligibility thresholds. Section 1115 waiver priorities promoted and approved under former President Trump included eligibility restrictions such as work and premium requirements and other conditions on coverage that did not require legislative changes to advance.

Tradeoffs: The Biden-Harris Administration proposals would expand access to coverage as well as keep more individuals connected to care over time but could increase spending. Continuous eligibility policies, which have been embraced by the Biden-Harris Administration, have been shown to lower rates of “churn,” or the temporary loss of Medicaid coverage in which enrollees disenroll and then re-enroll within a short period of time. When individuals churn on and off coverage, they experience gaps in coverage that can limit access to care and lead to delays in getting needed care. The Biden-Harris Administration proposals would also likely increase enrollment among eligible children by eliminating CHIP premiums and enrollment fees, which can pose barriers to coverage.

In contrast, Republican proposals support more frequent and robust eligibility redetermination processes to ensure that only eligible individuals remain enrolled in the program. However, with more frequent data checks, eligible individuals are at risk for losing coverage if they do not receive or understand notices or forms requesting additional information to verify eligibility or do not respond to requests within required timeframes. Some enrollees, such as working individuals whose monthly income fluctuates, would be at a higher risk of churning on and off Medicaid than others. In addition, most Republican proposals include work requirements. The data show that the majority of Medicaid enrollees are already working, and available implementation data suggests that Medicaid work and reporting requirements are confusing to enrollees and result in substantial coverage loss among eligible individuals who experience challenges with reporting requirements. Republican proposals also provide states additional flexibility by reducing requirements to cover certain benefits and allowing more premiums and cost-sharing. Overall, Republican proposals describe these changes as aligning Medicaid with private insurance, and they would likely reduce federal Medicaid spending. Because people on Medicaid have lower incomes and are sicker than the general population, these changes are also likely to pose barriers to enrollees’ ability to maintain coverage and access needed health care.

(Back to top)

Prescription Drugs

Context: All state Medicaid programs have opted to cover prescription drug benefits, and states have flexibility in how they administer the Medicaid pharmacy benefit within federal guidelines about pricing and rebates. Under the Medicaid Drug Rebate Program (MDRP), a manufacturer must enter into a rebate agreement with the Secretary of Health and Human Services, agreeing to rebate a specified portion of the Medicaid payment for the drug to the states and the federal government. In exchange, Medicaid programs cover nearly all of the manufacturer’s FDA-approved drugs, and the drugs are eligible for federal matching funds.

Biden-Harris: The Biden-Harris Administration enacted the Inflation Reduction Act (IRA) of 2022, a major piece of legislation with a number of prescription drug provisions aimed at reducing federal drug spending, including allowing the federal government to negotiate prices for some drugs in Medicare and requiring that drug manufacturers pay rebates for price increases above inflation in Medicare. The latest proposed budget outlines steps to build on the IRA, including expanding the Medicare drug negotiation program. The administration also approved Florida’s plan to import some prescription drugs from Canada, though implementation is contingent on further action by Florida. Within Medicaid, the Biden-Harris Administration has taken action to strengthen the MDRP by lifting the rebate cap, and their latest budget proposal includes provisions to expand the MDRP and allow the federal government to negotiate for supplemental rebates on behalf of state Medicaid programs, though these provisions have not passed. The Biden-Harris Administration also issued a proposed rule (yet to be finalized) aimed at increasing price transparency and established a voluntary model to increase access to cell and gene therapies for people with Medicaid.

Trump and Republican Proposals: The Trump Administration encouraged states to submit waivers allowing states to create closed formularies, enabling Medicaid programs to negotiate directly with manufacturers for drug coverage (under current law, the MDRP creates an open formulary). The Trump Administration also created a new pathway to allow states to import prescription drugs from Canada, finalized a rule allowing manufacturers to report multiple best prices under value-based purchasing agreements with the goal of increasing value-based arrangements in the MDRP, and proposed lifting the rebate cap (this was later lifted by Biden-Harris Administration legislation). Both the RSC FY 2025 budget and Project 2025 seek to repeal the IRA, citing its impact on innovation and patient access, and Project 2025 proposes to address drug shortages by eliminating manufacturer rebate payments for multi-source generic drugs. Further, broader Republican proposals to cut and cap federal Medicaid spending could have implications for prescription drug coverage, as it is an optional benefit.

Tradeoffs: Recent Republican proposals take aim at the IRA while the Biden-Harris Administration plans to continue strengthening the IRA. KFF polling has found most voters are supportive of the Biden-Harris Administration proposals that build on the IRA. While the IRA is expected to lower prescription drug costs for people with Medicare and reduce drug spending by the federal government, the provisions do interact with the MDRP and could increase overall Medicaid prescription drug spending. Both administrations have taken action on drug importation, but, while the idea of importing prescription drugs has bipartisan support among the general public, there are long-standing concerns in terms of ensuring drug safety. The approach is also opposed by the Pharmaceutical Research and Manufacturers of America (PhRMA) and the Canadian government and resulting savings estimates are unclear.

The Trump and Biden-Harris Administration records on the MDRP have diverged, with the Biden-Harris Administration working to strengthen the MDRP while the Trump Administration proposed allowing states to opt out of the program. While opting out of the MDRP could allow states to have more leverage when negotiating (because they could decide not to cover certain drugs completely), it is not clear that a closed formulary would result in savings, and closed formularies can restrict access to drugs for enrollees. Further, recent Republican proposals to address drug shortages and the late-era Trump Administration rule allowing multiple best price reporting could potentially weaken the MDRP and reduce the rebates states and the federal government receive from manufacturers. According to a Congressional Budget Office (CBO) study, rebates in the MDRP result in lower net drug prices in Medicaid compared with other federal programs.

(Back to top)

Long-Term Services and Supports (LTSS)

Context: LTSS encompass the broad range of paid and unpaid medical and personal care services that assist with the activities of daily living such as eating and bathing and the instrumental activities of daily living such as preparing meals and managing medication. They are provided to people who need such services because of aging, chronic illness, or disability and may be provided in institutional settings such as nursing facilities or in people’s homes and the community. Services provided in non-institutional settings are known as home and community-based services (HCBS), and those settings may include a person’s home, adult day care centers, assisted living settings, and group homes. Medicaid is the primary payer for LTSS, providing services to an estimated 1.4 million enrollees in nursing facilities and 4.5 million enrollees in HCBS settings.

Federal Medicaid statutes require states to cover nursing facility care and home health, but coverage of other LTSS is optional. Most HCBS are provided through “waivers,” which allow states to offer a wide range of benefits and to choose—and limit—the number of people who receive such services. Unlike other Medicaid services, many states use waiting lists when the number of people seeking services exceeds the number of waiver slots available. An estimated 0.7 million people have been on HCBS waiting lists in most years between 2016 and 2023. Medicaid has historically spent more money on LTSS in institutional settings than on LTSS delivered in home and community-based settings, but over the past decade, Medicaid has spent more on HCBS than institutional care due to states’ community integration obligations under the Americans with Disabilities Act and the Supreme Court’s Olmstead decision. In Olmstead, the Supreme Court held that the unjustified institutionalization of people with disabilities is illegal discrimination and violates the Americans with Disabilities Act. A major challenge for Medicaid HCBS is finding enough workers, with employment levels in elderly care and nursing care facilities remaining below their pre-pandemic levels.

Biden-Harris: The Biden-Harris Administration increased federal financing for and oversight over Medicaid LTSS through increased funding, new legislative proposals, and rulemaking. The Administration enacted ARPA during the pandemic, which included a provision that temporarily increased the federal match rate for spending on Medicaid HCBS by 10 percentage points to expand access to HCBS and support direct care workers. The Biden-Harris Administration also issued legislative proposals calling for more permanent and extensive federal investments in Medicaid HCBS. The Biden-Harris Administration’s FY 2025 budget proposes further HCBS investment, requires the reporting of HCBS quality measures, and includes actions to further increase oversight of nursing facilities. The Administration also finalized several regulatory changes that would bolster Medicaid LTSS:

  • Several rules aimed at combatting discrimination on the basis of disability, which may provide people with disabilities greater autonomy and independence while they receive Medicaid LTSS;
  • Requirements for states to provide access to Medicaid HCBS, which includes ensuring payment rates are sufficient to support a workforce, reporting requirements related to waiting lists and other quality measures, and ensuring that 80% of Medicaid payments for certain HCBS are spent on compensation for direct care workers; and
  • Establishment of the first ever requirements for minimum staffing levels in nursing facilities with increased transparency around the percent of Medicaid payments for institutional LTSS that are spent on workforce compensation and private equity ownership of nursing facilities.

Trump and Republican Proposals: The Trump Administration budgets and other Republican proposals do not include specific changes to Medicaid LTSS, but past policies and current proposals hint that services for older adults may be favored over people with disabilities who are under age 65. The Trump Administration’s regulatory actions generally resulted in less federal oversight over LTSS. The Trump Administration took actions to reduce federal oversight of Medicaid LTSS, including issuing a proposed rule that would have reduced federal requirements for nursing facilities (but was never finalized), suspending routine inspections in nursing facilities during the early months of the COVID-19 pandemic, and delaying requirements for states to comply with federal rules surrounding HCBS (the “HCBS Settings rule).” When COVID-19 vaccines became available, the Trump Administration launched the Pharmacy Partnership for Long-Term Care Program to facilitate COVID-19 vaccinations in facilities where residents are ages 65 and older, though these initial federal vaccine programs did not cover settings that primarily serve people with disabilities under ages 65. Similarly, the 2024 Republican Party platform seeks to expand in-home care for older adults, but does not propose similar expansions for those with disabilities under age 65. It describes “shifting resources,” to support new policies to help seniors remain in their homes but does not specific what programs might be cut to free up such resources. Project 2025 includes language that would allow states “flexibility to redesign eligibility, financing, and service delivery of long-term care to serve the most vulnerable and truly needy and eliminate middle-income to upper-income Medicaid recipients,” which could include eliminating some of the optional Medicaid eligibility pathways that offer coverage to higher-income people who LTSS. Proposals that fundamentally change the financing structure of Medicaid such as cutting or capping federal spending would likely result in restrictions to LTSS.

Tradeoffs: The Biden-Harris Administration Medicaid LTSS policies could expand access to services and increase federal oversight of the quality of services provided but could increase spending on LTSS and reduce states’ flexibility in administering the Medicaid LTSS benefits. The Biden-Harris Administration has consistently called for legislative changes to provide additional funding for Medicaid LTSS (HCBS in particular).

In contrast, it is difficult to anticipate what specific proposals would emerge under a Trump Administration. The Trump Administration’s actions coupled with Republican proposals to eliminate the new nursing facility staffing standards and prohibit the requirement that 80% of Medicaid payments for certain HCBS are spent on compensation for direct care workers suggest that there would be less federal oversight of access to and quality of care, but more flexibility for states in providing Medicaid LTSS. Republican proposals support policies focusing on long-term care and helping seniors to remain in their homes but do not mention a role for Medicaid or home care for people under age 65, and nearly two thirds of people who use Medicaid HCBS are under age 65.

(Back to top)

Managed Care

Context: States design and administer their own Medicaid programs within federal rules and determine how they will deliver and pay for care for Medicaid enrollees. Managed care is the dominant delivery system for Medicaid enrollees, with almost three-quarters of Medicaid beneficiaries enrolled in comprehensive managed care organizations (MCOs), accounting for about 52% of total Medicaid spending (or more than $415 billion) in FY 2022. State managed care contracts vary widely, in the populations required to enroll, the services covered (or “carved in”), and the quality and performance incentives and penalties employed. While there are federal and state requirements for managed care plans, plans have flexibility in certain areas, including in setting provider payment rates.

Managed care plans are at financial risk for the services covered under their contracts and receive a per member per month “capitation” payment for these services. To limit the amount that plans can spend on administration and keep as profit, CMS requires states to develop capitation rates for Medicaid to achieve a medical loss ratio (MLR) of at least 85%. The MLR reflects the proportion of total capitation payments received by an MCO spent on clinical services and quality improvement, where the remainder goes to administrative costs and profit. While there is no federal requirement for Medicaid plans to pay remittances to the state if they fail to meet the MLR standard, states have discretion to require remittances.

Biden-Harris: The Biden-Harris Administration released final Medicaid managed care regulations in 2024 aimed at improving access to care, quality, health outcomes, and health equity. The new rules establish national maximum wait time standards for certain appointments and a requirement for states to conduct independent secret shopper surveys to validate MCO compliance with wait time standards. The Biden-Harris Administration also finalized rules related to plan prior authorizations focused on streamlining process, reducing approval wait times, and improving transparency. The Biden-Harris Administration has taken action to improve Medicaid managed care monitoring, oversight, and transparency, releasing reporting templates and toolkits for states and publicly posting state managed care annual reports.

The Biden-Harris Administration enacted the 2024 Consolidated Appropriations Act, which included a financial incentive to encourage certain states to collect remittances from Medicaid MCOs that do not meet minimum MLR requirements. The Biden-Harris Administration’s FY 2024 and 2025 budgets went further proposing to require Medicaid managed care plans to meet an 85% minimum MLR and to require states to collect remittances if plans fail to meet the minimum MLR. The most recent budget also included provisions aimed at strengthening CMS’s Medicaid managed care financial oversight and compliance tools.

Trump and Republican Proposals: During his presidency, Trump took administrative action to change Medicaid managed care rules, including relaxing rules around network adequacy and beneficiary protections. Trump’s FY 2020 budget proposed changes to the Medicaid managed care waiver process (e.g., to lengthen the duration of certain managed care waivers) aimed at reducing administrative burden. Project 2025 notes the intent to reform and increase oversight of Medicaid managed care but doesn’t provide specifics. Proposals that fundamentally change the financing structure of Medicaid would likely also have implications for the delivery of care in Medicaid, impacting managed care plans.

Tradeoffs: Recent Democratic administrations have made sweeping changes to Medicaid managed care rules and regulations to advance program goals related to access to care, quality, fiscal and program integrity, beneficiary protections and support, and monitoring and oversight. However, establishing national standards and other state requirements can add state administrative burdens to come into compliance. The Biden-Harris Administration’s most recent budget proposed requiring Medicaid managed care plans to meet a minimum MLR of 85% (the statutory requirement for Medicare Advantage plans) and requiring states to collect remittances from plans if they fail to meet the required minimum (which would likely result in Medicaid savings)—to encourage plan investment in services and quality improvement and to prevent the retention of excessive profits.

The Trump Administration finalized changes to Medicaid managed care rules with chief goals that differed from Democratic administrations. The Trump Administration’s stated goals included reducing state and federal administrative burden, streamlining regulations, and promoting state flexibility and innovation in care delivery; however, tradeoffs could include fewer beneficiary protections, less oversight involving access to care, and variation in oversight across states.

(Back to top)

Appendix

Comparison of Key Medicaid Policy Proposals and Actions

Medical Debt: The Canary in the Coal Mine for Health Care Affordability

Author: Larry Levitt
Published: Sep 5, 2024

With Vice President Harris promising to address medical debt as part of her economic plan, KFF Executive Vice President for Health Policy Larry Levitt explores why it is a symptom of the broader problem of affordable health care and reviews recent efforts to address it in this JAMA Health Forum post.

A Look at State Efforts to Ban Cellphones in Schools and Implications for Youth Mental Health

Authors: Nirmita Panchal and Sasha Zitter
Published: Sep 5, 2024

Note: Figure 1 was updated on April 30th, 2025, to reflect the recent passage of cell phone ban legislation in Arizona and New York.

Heading into the 2024-2025 school year, a growing number of states are implementing or considering state-wide bans on cellphones in schools. Many leaders in education and policymakers suggest cellphone bans will help mitigate youth mental health concerns and distractions during academic instruction. The resurgence of cellphone bans follows two advisories from the U.S. Surgeon General on the youth mental health crisis and the harmful impacts of social media use and recommendations from UNESCO to limit cellphone use in schools across the world. Unlike many recent political issues, school cellphone ban policies have largely received bipartisan support, and the Biden-Harris administration continues to promote online safety for youth. At the same time, research on the effectiveness of cellphone bans is limited, and although multiple states are adopting these bans, challenges remain with enforcement, accommodating exceptions, and equity.

How widespread are school cellphone bans?

Cellphone bans began decades ago amid concerns about drug deals among students via cellphones or pager devices, and have fluctuated since. In 2009, 91% of public schools prohibited cellphone use, which fell to 66% in 2015 before rising again to 76% in 2021. Cellphone bans are now being considered at the state level in light of growing student academic and mental health concerns that are associated with excessive use of smartphones.

Eleven states have passed state-wide policies that ban or restrict cellphone use in schools as of April 30th, 2025 (Figure 1). These policies vary from state to state. 

  • Arizona’s Governor signed a bill in April of 2025 that instructs schools to limit student cell phone use during the school day, including non-instructional time, with exceptions for educational and medical purposes. The bill also directs schools to restrict internet and social media access for students.
  • Arkansas‘ Governor signed a law in February 2025 requiring each school district to create a cell phone use policy before the 2025-2026 school year that limits students’ phone use during the school day, following its pilot program in 2024. Policies must be submitted to and approved by Arkansas’ Division of Elementary and Secondary Education.
  • California’s Governor recently signed the Phone-Free School Act, which requires school districts and charter schools to develop and adopt a phone policy that either limits or entirely prohibits smartphone use during school by July of 2026. Exceptions will be made for medical necessity, emergencies, educational purposes, or with faculty permission.
  • Florida passed a phone ban for K-12 classrooms that prohibits cellphone use during class time and blocks access to social media for all devices on district Wi-Fi. Additionally, there is a digital literacy component beginning in sixth grade that requires education on the spread of misinformation on social media and digital footprints. The ban went into effect in July 2024.
  • Indiana’s ban prohibits students from using any portable wireless device (including cellphones, gaming devices, laptops, and tablets) during instructional time, with exceptions from teachers and/or administrators, or during emergencies. Each school board in Indiana is then expected to draft and publicly post specific policies for their schools – i.e. whether students can access their devices during lunch or what consequences students may face for using prohibited devices. The ban went into effect in July 2024.
  • Louisiana passed a ban, which will take effect in the 2024-2025 academic year, that prohibits both the use and possession of cellphones throughout the school day. If cellphones are brought onto school property, they must be turned off and stored away. Exceptions can be made for students who require learning accommodations.
  • Minnesota’s bill instructs school districts and charter schools to adopt policies on student cell phone use and possession by March 2025, but it does not specify the nature or extent of these policies.
  • New York’s governor signed a bill in April of 2025 instructing school districts to create and implement plans to restrict all student cellphone use during the school day, including non-instructional time, by the 2025-2026 academic year.
  • Ohio’s ban, similarly to Indiana’s, requires every school district to create and implement official policies regarding cellphone use at school. The bill includes exceptions for those with health conditions that require monitoring or for learning accommodations. The bill will take effect in July 2025.
  • South Carolina’s ban was implemented via the Governor’s Budget Proviso 1.103, which requires public schools seeking State Aid to Classrooms to implement the model policy drafted by the State Board of Education beginning in January 2025. The Board’s model policy was approved in September of 2024, prohibiting students from accessing unauthorized electronic devices unless authorized for educational or health purposes. A special exception is made for students who volunteer for emergency response organizations, who must receive written permission.
  • Virginia’s Governor established Executive Order 33, which ordered state officials to solicit public opinion regarding cellphones in schools to allow them to create definitions of “cellphone-free education” and to publish both model implementation plans and draft policy guidance to inform public school systems’ phone policies. The governor also ordered the state to make $500,000 available to support the implementation of school cellphone policies.
School Cell Phone Bans or Restrictions, by State

Seventeen states have introduced state-wide legislation that bans or restricts cellphone use in schools and education departments in seven states have issued recommended policies or pilot programs that similarly aim to ban or restrict cellphone use in schools (Figure 1). States are taking a variety of measures to mitigate cellphone use during instruction time. In Connecticut and West Virginia, their respective education departments have issued guidance on restricting cellphone use in schools (Figure 1). Pilot programs via the education department in Delaware allocate funds for students to use lockable magnetic phone pouches during school hours. Similarly, while legislation is under consideration in Pennsylvania, the Governor amended the existing School Safety and Mental Health grant program, allowing for the purchase of lockable phone pouches. Georgia has restricted access to social media platforms during school, and New Jersey established a commission to study the effects of social media use at school. Utah introduced a bill that subsequently failed, but draft bills indicate that these states continue to work towards phone-free learning environments.

Implementation and enforcement of cellphone bans may be difficult to navigate. The enforcement of these cellphone bans often becomes an added responsibility for teachers. Exceptions to these bans are also challenging to navigate as many students may need their devices for medical reasons or parents have differing expectations for maintaining contact. Additionally, cellphone bans have brought to light equity concerns – for example, New York’s prior state-wide cellphone ban was lifted in 2015 in part because of stricter enforcement at schools serving students from low-income households compared to schools serving students from high-income households. At the same time, banning cellphones has been linked to positive outcomes, such as improved test scores, especially among students who typically do not perform as well academically.

What is the connection between cellphone use and mental health?

Youth often use cellphones to access social media and social media is linked to poor mental health. In 2023, a survey of adolescents found that 51% reported using social media for at least four hours per day. Adolescent social media use is associated with higher rates of anxiety and depression, exposure to harmful content – the effects of which adolescents are more susceptible to – and body dissatisfaction and eating disorders, especially among girls. Excessive social media use and social media addiction are associated with sleep issues, which may result in negative neurological effects. However, social media use among youth can also be beneficial as it allows for self-expression, finding communities with shared interests, and accessing important resources, including mental health resources.

Approximately nine in ten public schools report occurrences of cyberbullying – a form of bullying through technological devices, including cellphones – among students (Figure 2). Cyberbullying is associated with social and emotional distress, depression, and suicidal ideation among youth and is more often experienced by female and sexual minority youth compared to their peers. In 2023, 16% of high school students reported electronic bullying, and this was heightened among LGBT+ adolescents (25%) and females (21%). Technological devices can also be used to create and spread digitally altered pornographic content without consent – a practice that primarily targets females and may negatively impact their mental health. Further, cellphone ownership among youth is linked to increased experiences of cyberbullying.

9 in 10 Public Schools Report Occurrences of Cyberbullying Among Their Students During the School Year

Excessive cellphone use can distract from in-person socialization and is associated with loneliness among adolescents. Establishing and building relationships with peers is beneficial to youth well-being and can have a protective effect on adolescents experiencing adversity. With the distraction of cellphones, peer relationship-building may be negatively impacted.

Approximately 40% of public schools report moderate to severe negative impacts on student learning and on teacher and staff morale when students use their electronic devices without permission (Figure 3). Many teachers report that students being distracted with their cellphones is a major problem in their classrooms and that enforcing cellphone restrictions is challenging. The presence of smart phones may reduce cognitive capacity, especially for those highly addicted to their phones, and notifications disrupt focus and attention. Further, there is a negative association between time spent on smartphones and academic performance.

4 in 10 Public Schools Report That Student Learning and Teacher Morale are Negatively Impacted by Unpermitted Use of Electronic Devices

What is known on the effectiveness of cellphone bans and other actions to address youth mental health?

While evidence on the outcomes of school cellphone bans is limited, widespread concerns regarding the harms of smartphone use on youth well-being continue to invoke action by policymakers and leaders in education. Emerging research on student outcomes is mixed, with some studies suggesting improvements in student mental health and academic performance and a reduction in bullying, and others showing little to no change. While evidence on school bans is inconsistent, rising concerns regarding the harms of social media and internet use among youth have led to policy and safety measures being introduced at the state and federal level. For instance, policymakers recently introduced bipartisan legislation – the Focus on Learning Act – that calls on the U.S. Department of Education to conduct studies on the impact of cellphone use on students’ academic and mental health outcomes, among other provisions. Additionally, as of December 2024, the U.S. Department of Education called on all states and districts to adopt measures to manage cellphone use in schools and published guidance which includes example policies, considerations to accommodate needs of different populations, and policy evaluation and modification guidelines. This guidance was published in response to the Biden-Harris Administration’s efforts to address youth mental health and online safety. Other multi-prong approaches are also being implemented, such as the Biden-Harris administration’s continued efforts to improve online safety for children. These include creating the Kids Online Health and Safety Task Force, which recently released Best Practices for Families and Guidance for Industry, and a Call to Action to mitigate image-based sexual abuse. Additionally, the Surgeon General recommended that social media platforms include a warning label that states that social media is linked to poor mental health among adolescents.

Potential Impacts of New Requirements in Florida and Texas for Hospitals to Request Patient Immigration Status

Published: Aug 26, 2024

Recent actions in Florida and Texas newly require hospitals to request immigration status from patients, with the aim of assessing the cost of providing care to undocumented immigrants. In May 2023, Florida Governor Ron DeSantis signed Senate Bill (SB) 1718 into law, which, among other actions, requires hospitals that receive Medicaid or Children’s Health Insurance Program (CHIP) funding to collect information on patient immigration status. In August 2024, Texas Governor Greg Abbott issued an executive order that similarly requires hospitals receiving Medicaid or CHIP funding to collect information on patient immigration status effective November 1, 2024. While these actions require hospitals to request this information, they must also inform patients that, as required by federal law, their response will not affect their care. In Florida, the hospitals must also indicate that the response will not result in a report to immigration authorities. This requirement is not specified in the Texas Executive Order. Under federal law, hospitals are required to provide emergency screening and stabilization services to all patients seeking emergency care. This brief examines the potential implications of these requirements for immigrant families and the states’ workforces and economies.

Noncitizen immigrants, including lawfully present and undocumented immigrants, make up about one in ten of residents in Florida and Texas, but larger shares of residents, including many U.S.-born citizen children, live in immigrant families. As of 2022, there were roughly 2 million noncitizen immigrants in Florida, including both lawfully present and undocumented immigrants, who make up 9% of the state’s population, and Texas had about 2.8 million noncitizen immigrants, who make up one in ten of the state’s population (Figure 1). A larger number of residents in both states live in immigrant families, which often include people of mixed immigration statuses, including U.S.-born citizen children. In Florida, 17% or over 740,000 children have at least one noncitizen parent, and in Texas over one in five (22%) or about 1.7 million children are in a family with a noncitizen parent.

1 in 10 People in Florida and Texas is a Noncitizen Immigrant

Overall, KFF and other research shows that despite having a higher uninsured rate, noncitizen immigrants, particularly those who are undocumented, use less health care and have lower health care spending than those born in the U.S. Undocumented immigrants are less likely than citizens to report using health care, including emergency care. Further, immigrants, including undocumented immigrants, have lower per capita health care expenditures than U.S.-born citizens, and data suggest they subsidize health care for U.S.-born citizens by paying more into the system through health insurance premiums and taxes than they utilize. Lower use of health care among immigrants likely reflects a combination of them being younger and healthier than their U.S.-born counterparts as well as them facing increased barriers to care, including lower rates of coverage due to more limited access to private coverage and Medicaid eligibility restrictions for immigrants. Undocumented immigrants are not eligible to enroll in Medicaid or other federally funded health coverage. Medicaid payments for emergency services may be made directly to hospitals for individuals who are otherwise eligible except for immigration status to help cover the costs incurred for providing this care. Federal funds may not be used to provide Medicaid coverage to undocumented immigrants, though some states have used their own funds to provide such coverage.

Initial data from Florida suggest that less than 1% of hospital emergency room visits and admissions were among undocumented immigrants. Florida released a public dashboard and a separate report submitted to the state legislature based on hospital reporting under the new requirement for June through December 2023. These reports show that less than 1% of inpatient admissions and emergency department visits were among patients who identified themselves as not lawfully present. The 7% to 8% of patients who declined to provide responses may include additional undocumented immigrants. The state extrapolates an estimate that the cost of care provided to undocumented immigrants over this period was $556 million. However, as noted by the state in the legislative report and others, it is unclear how much of that care was uncompensated. Costs of care for undocumented immigrants may be covered via self-pay, private coverage, or, in some cases, Emergency Medicaid (if it is emergency care provided to an individual who would otherwise be eligible except for immigration status). The legislative report further specifies that the state did not identify any correlation between the level of uncompensated care and the level of undocumented immigrants presenting at the hospital, and that high levels of uncompensated care were more associated with rural county status than the share of undocumented immigrant patients. There also did not appear to be a correlation between total profitability and the share of undocumented immigrants.

These new requirements, along with other recent restrictive immigration policies in these states, will likely contribute to increased fears among immigrant families, which may negatively impact their daily lives, physical health, and mental well-being. These new hospital requirements sit against a backdrop of other recent restrictive immigration policies enacted by these states. For example, the Florida law also creates penalties for hiring undocumented immigrants, expands employment verification screening requirements, invalidates out-of-state drivers’ licenses for undocumented immigrants, establishes criminal penalties for transporting undocumented immigrants into the state, increases funding to relocate or bus migrants to other parts of the U.S., and expands state authority to carry out immigration enforcement. Texas passed legislation in November 2023 that enables state and local law enforcement agencies to question and arrest any individual they believe to be an undocumented immigrant at ports of entry, although its implementation is currently on hold due to legal challenges. These types of new restrictions, along with the new hospital requirements, will likely increase fears among immigrant families, which may may make them more reluctant to access health care for themselves and their children. Prior KFF analysis of Trump-era restrictive immigration policies found that such fears and impacts extend beyond undocumented immigrants to those who are lawfully present and children in immigrant families, who often are U.S-born citizens. One news report suggests that Florida’s Emergency Medicaid expenditures fell after implementation of the requirements, which may reflect decreased use of care among undocumented immigrants and could have negative health consequences given that this funding goes toward emergent care, including care for labor and delivery. It is also possible that some undocumented immigrants have migrated out of the state. Given these types of concerns, the American Medical Association suggests avoiding explicit documentation of immigration status of patients and their family members in a health record.

These actions also will likely have implications for the states’ economies and workforces. More limited access to health care can potentially negatively affect worker productivity. Moreover, increased fears among immigrant families may impact their work. Following Florida’s passage of SB 1718, it was reported that food service businesses in the state lost not only long-time employees but also customers who now may be afraid of going to public places. Agriculture and construction industries have also taken a hit, with reports of abandoned construction sites in the state following the passage of SB 1718. It is likely that Texas’ economy and workforce will also be impacted due to the chilling effects of and misinformation around SB 4, even though the law is yet to take effect. These impacts may be especially significant in Florida and Texas due to the outsized role immigrants play in the states’ workforces, particularly in certain occupations. Almost three quarters of nonelderly noncitizen immigrants work, similar to the share of their citizen counterparts. Noncitizen immigrants make up 12% of Florida and Texas’ overall nonelderly adult workforce, but they make up higher shares of workers in certain occupations. In Florida, noncitizen immigrants account for one in three (34%) of the state’s construction workers, almost half (47%) of the state’s farming and fishing workers, and one in six (17%) transportation workers. In Texas, noncitizen immigrants make up four in ten (41%) construction workers, one in three (33%) farming and fishing workers, and about one in eight (13%) transportation workers. The impacts of lost workers in these occupations may have larger ripple effects through the states’ economies and beyond.

Noncitizen Immigrants Play Outsized Roles in Certain Occupations in Florida and Texas

How Narrow or Broad Are ACA Marketplace Physician Networks?

Authors: Matthew Rae, Karen Pollitz, Kaye Pestaina, Michelle Long, Justin Lo, and Cynthia Cox
Published: Aug 26, 2024

Report

One way insurers seek to control costs is to limit the size of the physician networks serving their plans. Providers agree to lower fees and other terms with insurers in order to be included in one or more of the networks they offer. Insurers then either limit coverage to services provided by network providers or encourage enrollees to use network providers through lower cost sharing. Reducing the number of providers in-network can effectively reduce plan costs, but it also limits enrollees’ choices, increases wait times, and can complicate the continuity of care for those switching plans. Enrollees receiving care from out-of-network providers often face coverage denials or substantially higher out-of-pocket expenses. These factors highlight how the size and composition of provider networks impact access to care and the financial protection insurance provides enrollees.

The breadth of provider networks in the Affordable Care Act (ACA) Marketplaces has been the subject of significant policy interest. Insurers often compete aggressively to be among the lowest-cost plans, potentially leaving enrollees with poor access. According to the 2023 KFF Survey of Consumer Experiences with Health Insurance, one in five (20%) consumers with Marketplace plans reported that in the past year, a provider they needed was not covered by their insurance, and nearly one in four (23%) said a provider they needed to see that was covered by their insurance did not have appointments available. Enrollees with Marketplace coverage were more likely than those with employer coverage to face these challenges. While the Centers for Medicare and Medicaid Services (CMS) establishes minimum standards for the adequacy of provider networks for Marketplace plans, insurers retain considerable flexibility in how they design networks and how many providers they include. As a result, the breadth of plan networks varies considerably within counties, presenting challenges for consumers who need to select a plan with little information on the network breadth of their options.

This brief examines the share of doctors participating in the provider networks of Qualified Health Plans (QHPs) offered in the individual market in the federal and state Marketplaces in 2021, and how network breadth affected costs for enrollees. The analysis uses data on the physician workforce, from 2021, matching that to provider networks in marketplace plans from the same year. Doctors filing Medicare Part B claims in or near each county are considered to be part of the active workforce available to Marketplace enrollees. Only doctors filing a claim and therefore known to have engaged in patient care in 2021 were included. The share of local physicians participating in a network is a rough measure of how much access enrollees have; depending on the number of providers in the area and the workloads of those physicians, enrollees in plans with similar breadths may face different wait times to book appointments. The share of local physicians participating in-network distinguishes whether enrollees have a broad or narrow choice of local doctors. Those in plans including a small share of doctors have fewer options when trying to find a provider with available appointments. See the Methods section for more details.

Key Findings

  • On average, Marketplace enrollees had access to 40% of the doctors near their home through their plan’s network, with considerable variation around the average. Twenty-three percent of Marketplace enrollees were in a plan with a network that included a quarter or fewer of the doctors in their area, while only 4% were in a plan that included more than three-quarters of the area doctors in their network.
  • Some of the narrowest network plans were found in large metro counties, where enrollees on average had access to 34% of doctors through their plan networks. Marketplace enrollees in Cook County, IL (Chicago) and Lee County, FL (Fort Myers) were enrolled in some of the narrowest networks (with average physician participation rates of 14% and 23%, respectively). Plans in rural counties tended to include a larger share of the doctors in the area, though rural counties had fewer doctors overall relative to the population compared to large metro counties.
  • On average, more than one-quarter (27%) of actively practicing physicians were not included in any Marketplace plan network.
  • On average, Silver plans with higher shares of participating doctors had higher total premiums. Compared to plans where 25% or fewer of doctors participated in-network, those with participation rates between 25% and 50% cost 3% more while those with participation rates of more than 50% cost 8% more. (Silver plans are midlevel plans in terms of patient cost-sharing and are particularly significant because they are the benchmark for federal premium subsidies.)
  • More than 4 million enrollees (37% of all enrollees) lived in a county in which the two lowest-cost Silver plans included fewer than half of the doctors in the area and a broader plan was available. In order for these enrollees to enroll in the cheapest Silver plan that included at least half the doctors, they would have needed to spend an additional $88 per month.

How Broad are Marketplace Plan Physician Networks?

On average, enrollees in the ACA Marketplaces had access to 40% of the doctors near their homes through their plan’s network. This share was similar for pediatric and non-pediatric doctors.

Most Marketplace Enrollees Had Access to Fewer than Half of Local Physicians in their Plan’s Network

A quarter of enrollees were in plans where fewer than 26% of the local doctors participated in their plan’s network, while another quarter were in plans where at least 54% of local doctors participated.

There is no formal definition of what constitutes a narrow network plan. Some researchers have labeled plans covering fewer than a quarter of the physicians in an area as narrow. Under this definition, 23% of Marketplace enrollees were in a narrow network plan. About seven in ten enrollees (70%) were in a plan that included half or fewer of the doctors near their home. Only 4% of enrollees were in a plan that included at least three-quarters of local doctors, and 1% of enrollees were in a plan that included at least 85% of local doctors.

Most Marketplace Enrollees Were in Plans that Included Less than Half of Local Physicians in their Plan’s Network

How Broad Are Plan Networks for Primary Care and Physician Specialties?

Even a plan with a relatively large share of local doctors participating in its network may not have enough doctors in different specialties to meet the needs of plan enrollees. In particular, enrollees with chronic conditions may look for plans that include their doctors across multiple specialties.

Primary Care Physicians: Marketplace enrollees, on average, had plan networks that included 43% of the primary care doctors in their area. A quarter of Marketplace enrollees had plan networks that included fewer than 25% of primary care doctors. More than half a million Marketplace enrollees were in a plan with fewer than 50 in-network primary care doctors near their homes. As is the case for physician networks overall, primary care physician networks tended to be narrower in large metro counties, where the average enrollee had a plan network that included 35% of local primary care doctors. While primary care doctors account for a smaller share of spending than specialists, they play an important role in insurers’ network design either by acting as gatekeepers to specialty care and referring patients to specialists.

Specialists: Marketplace plan networks tended to include a larger share of practicing medical and surgical specialists than primary care physicians. The average Marketplace enrollee had a plan network that included 52% of medical specialists and 53% of surgical specialists in their area; however, one-quarter of Marketplace enrollees had access to fewer than 34% of the medical specialists and 32% of the surgical specialists. On average, Marketplace enrollees had plan networks that included 21% of hospital-based physicians, which may include anesthesiologists, radiologists, pathologists, and emergency physicians.1  Information on additional specialties is available in the appendix.

Psychiatrists: Marketplace networks for psychiatrists were smaller. On average, Marketplace enrollees had access to 37% of the psychiatrists in their area through their plan.2  Twenty-five percent of Marketplace enrollees were in a plan that included 16% or fewer of the psychiatrists near their homes.

The Average Marketplace Enrollee Had Access to About Half of Medical and Surgical Specialists in Their Area, but Fewer than Half of Primary Care Doctors and Other Specialists

How Does Network Breadth Vary by Location?

Network breadth varied based on where plans were offered, with those in urban areas having lower physician participation rates, on average. In 2021, CMS designated county types based on their population and density; there are 78 Large Metro counties and 723 Metro counties. Most Marketplace enrollees lived in one of these urban county designations, including 38% in Large Metro counties and 48% in Metro counties.

Urban Counties: While Large Metro and Metro counties had more doctors, smaller shares of them participated in Marketplace plan networks compared to doctors in more rural areas. Marketplace enrollees in Large Metro counties, on average, had access to 34% of the doctors in their area through their plan networks, with a quarter enrolled in a plan whose network included fewer than 23% of local doctors. Marketplace enrollees in Metro counties, on average, had access to 42% of local doctors through their plan networks, while those in Rural counties, on average, had access to 52% of local doctors.

The Average Enrollee in Large Metro and Metro Counties Had in-Network Access to Less than Half of Local Doctors

The 30 counties with the highest enrollment in the Marketplaces collectively represented 34% of all Marketplace enrollees and 21% of the U.S. population. These counties are typically urban and disproportionately in states that have not expanded Medicaid under the ACA.3 

There was significant variation in network breadth across these 30 counties. Differences in average network breadth across these counties are the result of a combination of factors including the physician workforce, market characteristics, and insurer strategies. With networks with low provider participation rates, most Marketplace enrollees in Cook County, IL (Chicago) had access to fewer than one in six (14%) doctors in their area on average. Similarly, Marketplace enrollees in Lee County, FL (Fort Myers) and Fort Bend County, TX (outside Houston) had in-network access to less than a quarter of local doctors (23% and 24%, respectively). In contrast, some larger US cities had broader networks than those available in Houston and Chicago. For example, enrollees in Middlesex County, MA (outside Boston), Gwinette County, GA (outside Atlanta), and Travis County, TX (Austin) had in-network access to almost half of the doctors in their areas on average (46%, 46%, and 49%, respectively).

In 2021, 14% of Marketplace enrollees (1.6 million people) lived in four counties: Los Angeles, CA; Miami-Dade, FL; Broward, FL (Fort Lauderdale); and Harris, TX (Houston). On average, enrollees in each of these counties had in-network access to less than 4-in-10 local doctors (25%, 36%, 38%, and 25%, respectively).

High physician participation rates may not result in meaningful choice if there are few doctors in the area in the first place. For example, enrollees in Hidalgo County, TX (McAllen), on average, had access to 61% of local doctors through their plan networks, but this may have reflected chronic shortages in the number of practicing doctors in the county.4 

On Average, Marketplace Enrollees in Nearly All Large Counties Were in Plans That Included Fewer Than Half of Local Doctors

Rural Areas: On average, Marketplace enrollees in Rural counties had access to about half (52%) of local doctors through their plan networks, higher than the average in more urban counties. The higher provider participation rates in rural areas, however, need to be considered in the context of the small number of primary care doctors and specialists practicing in these areas. For example, 2.9 million Marketplace enrollees in Rural counties had fewer than 10 dermatologists in their local area, 2.5 million had fewer than 10 gynecologists, and 1.7 million had fewer than 10 cardiologists in their plan networks. In some cases, these providers may already have full panels, and an enrollee’s choice may be even more limited than the number of physicians who accept the plan.

County Demographics: On average, Marketplace enrollees living in counties with a higher share of people of color had narrower networks than counties with a smaller share.5  The quarter of Marketplace enrollees living in the counties with the highest share of people of color had access to 34% of doctors in-network, on average, compared to 42% in counties with a smaller share of people of color. This difference may reflect the higher concentration of people of color in large metro counties, where plans typically had narrower networks.

Counties With Higher Shares of People of Color Had Narrower Physician Networks on Average

How Much Choice Do Consumers Have Over Networks in the County Where They Live?

Provider networks vary within counties, meaning that individuals shopping for a Marketplace plan may have the option to enroll in plans with vastly different network breadths. In 2021, 70% of enrollees (nearly 8 million people) lived in a county where one or more plans covered fewer than a quarter of the doctors in the area. Among these enrollees, nearly 4.3 million (54%) also had the opportunity to enroll in a plan that included more than half the doctors in the area.

In the 30 counties with the most enrollment, enrollees could choose from about 8 distinct plan networks, on average. Even within the same county, enrollees may have access to vastly different shares of physicians in-network. For example, in Lee County, FL (Fort Myers), a quarter of Marketplace enrollees were enrolled in plans with networks that included fewer than 5% of local doctors, while a quarter were enrolled in plans with networks that included more than 45%. Similarly, in Travis County, TX (Austin), a quarter of Marketplace enrollees were enrolled in a plan with a network that included fewer than 36% of local doctors, while a quarter were enrolled in plans that included at least 70%. Consumers in these counties have the opportunity to enroll in plans with vastly different physician networks but often face higher premiums to do so. (See section “How is Network Breadth Related to Plan Premiums?” for details.)

The Vast Majority of Enrollees in Large Counties Were in Plan Networks With Fewer Than Half of the Area Physicians

Access to a “Broad” Network Plan: A large share of Marketplace enrollees (91%) lived in a county in 2021 where they could not choose a plan with a network that included at least 75% of doctors in their areas. Among the 30 counties with the most Marketplace enrollment, only two—Middlesex County, MA (outside Boston) and Hidalgo County, TX (McAllen)—had at least one plan network choice with a physician participation rate of 75% or more. In most cases, the broadest Marketplace plan network offered in these 30 counties was much narrower than this. For example, the physician participation rate for the broadest Marketplace plan network offered was 22% in Cook County, IL (Chicago), 38% in Hillsborough County, FL (Tampa), and 40% in Maricopa County, AZ (Phoenix). In these counties, shoppers were unable to enroll in a plan that covered at least half of the doctors in their community, even if they were willing and able to pay more.

Enrollees in Large Counties Often Had Access to Less Than Half of Doctors

 

Across Large Counties, Physician Networks Varied Considerably, But Few Broad Network Options Were Offered

Doctors Not Participating in Any Marketplace Network: Some doctors did not participate in any Marketplace plan network in 2021. On average, 27% of actively practicing physicians who submitted Medicare claims were not included in any Marketplace plan network offered to enrollees that year. This means that people transitioning to a Marketplace plan from another coverage source may not have been able to find any plan that included their doctor. In some counties, a much higher share of doctors did not participate in any Marketplace network, including Cook County, IL (Chicago), where 60% of doctors did not participate in any Marketplace plan networks, Dallas County, TX (36%), and Lee County, FL (Fort Myers) (41%).

In Many Large Counties, At Least One-Quarter of Physicians Did Not Participate in Any Network

How Visible Are Differences in Network Breadth to Plan Shoppers?

The difficulty of selecting an appropriate plan for a consumer’s health needs is heightened by the tremendous number of choices in many counties. The average Marketplace consumer had a choice of more than 58 plans (including 23 Silver plans) in 2021, a number that has since grown.6 

Plan choices can involve different provider networks. For example, in Harris County, TX (Houston), consumers in 2021 had a choice of 87 plans that used seven different provider networks, with physician participation rates that ranged from 9% to 52%. However, these network differences are largely invisible to consumers. The lack of consumer tools to evaluate and measure plan networks can make it more challenging to choose a plan. Other than in a limited pilot operating in two states (Tennessee and Texas), the only tool available for HealthCare.gov consumers to evaluate a plan’s network is to search for individual providers, one by one, in directories, which may not always be up to date.

Further complicating the challenges of selecting plans, the marketing names of plans offered by the same insurer using different provider networks do not clearly indicate network differences. For example, AmeriHealth of New Jersey offers multiple Silver plans in Camden County, NJ. The narrow plan was marketed as “IHC Silver EPO AmeriHealth Advantage” (with a physician participation rate of 40%), while the broader network Silver plan was marketed as “IHC Silver EPO Regional Preferred” (with a physician participation rate of 74%). Based on these names, shoppers may not be able to discern that these plans had different networks with very different participation rates.

Shoppers can also search by plan type. The vast majority of Marketplace enrollees (84%) were in HMO or EPO plans in 2021, which have closed networks that generally do not cover non-emergency services provided outside of their provider network. A smaller share of Marketplace enrollees were in PPO plans (13%) and POS plans (4%), which provide some coverage for out-of-network care. The cost for such care can be quite expensive because out-of-network providers can sometimes balance bill and cost sharing for their services is typically higher and not subject to the annual out-of-pocket maximum.

Marketplace consumers seeking access to a broader choice of physicians and who have the choice of a PPO plan might assume such plan networks are analogous to the broad PPO networks offered to many in the employer market. On average in 2021, Marketplace enrollees who signed up for PPO plans had access to 53% of local doctors through their plan networks, compared to 37% for those enrolled in HMOs and 38% for those enrolled in EPO plans. However, plan type is not necessarily reflective of network breadth. In almost half (46%) of counties with both a PPO and either an HMO or EPO Marketplace plan, at least one HMO or EPO plan had a broader network than a PPO plan. Many Marketplace enrollees also did not have the option to choose a PPO plan: 60% of enrollees lived in a county in which only closed-network (HMO and/or EPO) plans were available.

Marketplace plans are categorized into metal levels based on the overall level of cost sharing required by the plans (deductibles, copays, etc.). In 2021, enrollees in Bronze, Silver, and Gold plans had access to similar shares of physicians in their areas (41%, 39%, and 44%, respectively). This is the result of issuers utilizing the same networks across metal levels within a county. In only 1% of counties did an insurer’s broadest Silver plan use a different network than its broadest Bronze plan.

HealthCare.gov has not yet widely released a consumer assistance tool to aid shoppers in filtering options by network breadth. Since 2017, CMS has operated a limited pilot with information on network breadth for consumers in Tennessee and Texas.7  Under this network transparency pilot, CMS provides measures of plan network breadth for hospitals, primary care providers, and pediatricians as an aid to Marketplace shoppers in those states. CMS calculates a participation rate by determining the share of providers participating in any Marketplace networks in the area. CMS then categorizes plan networks as “Basic” (0%-29%), “Standard” (30%-69%), or “Broad” (70%+), based on how many physicians participate in at least one QHP network. Whereas the denominator used throughout this analysis is physicians who submitted claims to Medicare, the CMS tool only considers providers that participate in Marketplace plans. Therefore, even plans with narrow networks in areas where most doctors do not participate in Marketplace plans could be labeled “standard” or “broad” using this method. For example, whereas 90% of physicians in Travis County, TX (Austin) who take Medicare participated in at least one Marketplace plan in 2021, only 64% of doctors in Dallas County, TX did. Therefore, a plan covering a quarter of all the available doctors in both counties would be considered a “basic” plan in Travis County, TX but a “standard” plan in Dallas County.

Generally, the method used in the CMS “network transparency” tool does not seem to facilitate comparing plan networks across counties and may exaggerate the breadth of plan networks, potentially leading some consumers to believe that their plan includes a larger share of local providers than it actually does. Under the CMS pilot method, only 16% of Marketplace enrollees in 2021 were enrolled in a plan that would be considered “basic”; this compares to 33% of Marketplace enrollees would be considered to be in a basic plan if the definition of local doctors used in this paper were applied.

Nearly Twice As Many Enrollees Were in a Plan That Would be Considered 'Basic' if CMS Pilot Study Measure Used A Broader Definition of Practicing Physicians

Network Breadth by Plan Insurer

Marketplace shoppers may consider who the insurer is when making inferences about plan networks.

Blue Cross and/or Blue Shield (BCBS) plans are sponsored by a mixture of for-profit and tax-exempt insurers. While these companies are run independently, they are affiliated through an association, and many share a common heritage. In many states, the BCBS affiliates are the largest insurers participating in the Marketplace and may in some cases also be the largest insurers or administrators for employer-sponsored coverage as well. On average, enrollees in BCBS Marketplace plans in 2021 had access to 49% of doctors in their areas through their plan networks, a larger share than enrollees in plans offered by other insurers (35%).8  Even so, BCBS Marketplace plan networks, on average, excluded about half of the doctors available to those in traditional Medicare. Further, there was considerable variation in participation rates by doctors among plans sponsored by BCBS insurers, sometimes even within the same county. For example, in Wayne County, MI (Detroit), the Blue Care Network and Blue Cross/BlueShield plan network participation rates ranged from 20% to 59% across plan options. Similarly, in Camden County, NJ, Independence Blue Cross offered two networks, with physician participation rates of 40% and 74%. Florida Blue in Miami-Dade County, FL offered multiple plan networks with participation rates ranging from 25% to 51%.

Network Breadth Varies Across Insurers

Insurers Also Participating in Medicaid Managed Care: Insurers with a large presence in the Medicaid managed care organization (MCO) market also have a solid footprint in the Marketplaces. Overall, the breadth of Marketplace plan networks sponsored by MCO insurers was similar to that of insurers overall (41% vs. 40%, respectively).9  One of the largest MCOs that expanded into the Marketplaces is Centene Corporation, which sponsors plans under Ambetter, Health Net, and other brand names. The average participation rate for doctors in plan networks offered by Centene was lower than the overall Marketplace average (33% vs. 40%). Molina, another major MCO insurer offering Marketplace plans, had an average physician participation rate of 35% in its plan networks.

Integrated Delivery Systems: Integrated delivery systems, such as Kaiser Permanente, Geisinger Health Plan, and the Chinese Community Health Plan, institute a different approach to network design. Under these plans, health care financing and delivery are conducted by the same organization. Providers are typically employees of the plan or an affiliated medical group, and these plans generally do not cover non-emergency care provided by doctors outside of the network. Although enrollees in these plans may not have a wide choice of physicians in the area, these integrated models strive to improve access through care coordination and may be less complex for patients to navigate which providers are in and out of their networks. Enrollees in Kaiser plans, by far the largest integrated delivery system, on average, had access to about one in five (19%) doctors in their area. Of note, the breadth of Kaiser physician networks does not lower the overall Marketplace average substantially because only 7% of Marketplace enrollees nationally were enrolled in Kaiser plans.

Non-profit Insurers: On average, Marketplace enrollees covered by plans sponsored by non-profit insurers in 2021 had in-network access to 43% of the doctors in their areas, compared to 38% for those covered by for-profit insurers. Excluding enrollees in Kaiser health plans, enrollees covered by non-profit insurers had access to 47% of local doctors on an in-network basis on average.

How is Network Breadth Related to Plan Premiums?

On average, Silver plans with higher shares of participating doctors had higher total premiums. When compared to plans where fewer than 25% of doctors participated in-network, those with participation rates between 25% and 50% cost 3% more while those with participation rates of more than 50% cost 8% more. While other factors also contribute to plan premiums, including the breadth of hospital networks and the plan design, narrow physician networks were associated with meaningfully lower total costs. The average total premium for a 40-year-old enrolled in a Silver Marketplace plan in 2021 was $466 a month. For these enrollees to sign up for a Silver plan that included more than 50% of area physicians, their premiums would have increased $37 per month. The statistical model used to estimate these premium differences is described in the methods.

Silver Plans With Broader Physician Networks Had Higher Premiums Than Those with Narrower Networks

Enrollee Cost to Purchase a Broader Plan

Consumers with private health insurance generally consider the breadth of provider networks very important when choosing a plan, yet many remain price-sensitive when selecting plans with higher costs. A 2019 KFF/LA Times survey found that 36% of adults with employer coverage said the cost of the plan (premiums and cost sharing) was the main reason they chose their plan, while 20% cited the choice of providers.

One way to illustrate how the cost of broader plans is passed on to consumers is to consider the counties where enrollees face higher premiums for a broader plan. Most (90%) of Marketplace enrollees receive a tax credit to offset all or part of the cost of the monthly premium. The size of the premium tax credit available to enrollees is based on both household income and the cost of the benchmark plan, defined as the second-lowest-cost Silver plan. ACA enrollees are responsible for paying the entire amount between the cost of the benchmark plan and a higher-cost plan. Enrollees in counties where the benchmark plans have relatively low physician participation rates may need to pay a significant amount to enroll in a broad network plan.

Among Marketplace enrollees, 74% percent, or 8.5 million enrollees, were in a county where the two lowest-cost Silver plans had fewer than 50% of physicians participating in their networks. Of these, about half, or 4.3 million enrollees, did not have a Silver plan available to them that included at least half of the local physicians in its network; 4.2 million enrollees did have at least one such plan available to them. For those 4.2 million people, the average additional cost to enroll in a Silver plan with at least half the local doctors participating was $88 (for a 40-year-old).

One in five Marketplace enrollees (19%, or 2 million enrollees) lived in a county where the two lowest-cost Silver plans included fewer than 25% of local physicians in-network. Fifty percent of these enrollees, or 1 million enrollees, lived in a county where at least one plan included at least half the doctors. Among these enrollees, the cost to enroll in a plan with at least half the local doctors would have cost $95 more than the benchmark plan each month.

Implications for Consumers and Potential Federal Efforts to Increase Access to Care

Having a plan with a narrow network increases the chances that an enrollee receives care out-of-network, either inadvertently (e.g., receiving care from an out-of-network provider they did not choose at an in-network facility), or because they are unable to find an in-network physician at the time and place they need. It can also have consequences for enrollees’ ability to seek care in a timely fashion and their health. The 2023 KFF Survey of Consumer Experiences with Health Insurance found that 20% of adults with Marketplace coverage said that in the past year, a particular doctor or hospital they needed was not covered by their insurance. Among Marketplace enrollees who experienced this problem, 34% said that needed care was delayed, 34% said they were unable to get needed care, and 25% experienced a decline in health status.

Additionally, going out-of-network can be costly for enrollees. Enrollees using out-of-network providers may face higher cost sharing and balance billing if the services provided are not regulated by the No Surprises Act. Among those who indicated experiencing a network adequacy problem in the consumer survey, almost half (47%) said they ended up paying more out of pocket for care than expected, including 22% who said the additional cost was $500 or more.

Some have suggested that the design of the Marketplace encourages insurers to offer narrower networks compared to those included in employer plans in order to keep premiums down. Employers use health benefits to attract and retain workers and have an incentive to create broader networks that appeal to their workforce. One analysis found that primary care networks for large group plans were 25% larger than those found on the Marketplaces.10  The higher prevalence of narrow network plans corresponds to a greater share of enrollees facing challenges finding in-network providers. The 2023 KFF Survey of Consumer Experiences with Health Insurance found that adults with Marketplace coverage were more likely than those with employer-sponsored health insurance to report that a particular doctor or hospital they needed was not covered by their insurance (20% vs. 13%) (Figure 14). Additionally, 34% of Marketplace enrollees in fair or poor health reported that a particular doctor or hospital they needed was not covered by their plan, nearly two times more than those with an employer plan (16%). Similarly, a forthcoming KFF analysis of the 2022 National Health Interview Survey found that challenges finding doctors led some adults to delay or skip care (Appendix Figure 7). Those with non-group coverage, such as Marketplace plans, were twice as likely as those with employer plans to indicate that they had delayed or skipped care in the past year because they couldn’t find a doctor who accepted their plan (7% vs. 3%). Among those who visited a hospital or emergency room during the past year, 11% of non-group enrollees reported skipping or delaying care, compared to 5% of those with employer coverage.

Marketplace Enrollees Are More Likely Than Those with Employer Plans To Need a Provider Who Was Not In Their Network

Even still, network breadth is only one component of access to care and may not always gauge how well enrollees are served. There are many aspects consumers consider when selecting a plan. This analysis examines network breadth but does not address other standards that health plans, physician networks, and physicians are required to meet. Enrollees in plans with broad networks may still face challenges scheduling appointments and considerable wait times. For some specialties, such as psychiatry, workforce shortages make it hard for enrollees to find providers even in plans that include a broad swath of physicians. Workforce shortages in many rural areas mean that even if a plan has a broad provider network, there still may be an insufficient number of providers to meet the needs of that community. Furthermore, many enrollees face additional challenges using their plan, including stringent prior authorization requirements.

Similarly, a plan with a narrow network—measured as the share of physicians in the area participating—may still provide adequate access to care, just not necessarily with a broad choice of providers. States use a range of network adequacy rules, with many requiring the inclusion of different types of providers, but only ten evaluate wait times to determine if a network meets minimum standards. The ACA requires that Marketplace plans maintain networks sufficient in number and types of providers for the purpose of ensuring that all services will be accessible without unreasonable delay. Currently, federal network adequacy standards require that plans provide access to at least one in-network provider for 90% of plan enrollees living within certain time/distance thresholds (for example, in large metro areas, no more than 10 minutes or 5 miles from a primary care provider, or no more than 30 minutes or 10 miles from an oncologist.) Although these standards measure geographic proximity to in-network care, they do not measure network breadth. Additionally, starting in 2025, federal Marketplace plans will be required to meet maximum appointment wait-time standards (e.g., no more than a 15-calendar day wait for routine primary care appointments or 30 days for non-urgent specialty care appointments).

A central challenge in analyzing network breadth is the quality of available data. The inclusion of so-called “phantom providers”—physicians listed in the network but who are not actually available to plan enrollees at the location or in the specialty they are listed—may increase the apparent breadth of plan networks without actually increasing access to care. Federal laws and regulations require Marketplace plans to publish online an up-to-date and complete provider directory. However, CMS has found high rates of incomplete and inaccurate information in these directories. Additionally, the No Surprises Act Improvements in plan directory data would facilitate regulation and decrease the burden on consumers comparing and using the plan. In 2022, CMS solicited public comment on establishing a national provider directory that private plans could use as a database for their own plan directories. Further action on this proposal is still pending, but this could improve available information about the landscape of available providers, allowing for the development of improved consumer information about provider ratios that show the share of practicing area providers (overall and by specialty) included in the provider network of each QHP.

This work was supported in part by a grant from the Robert Wood Johnson Foundation. KFF maintains full editorial control over all of its policy analysis, polling, and journalism activities.

Methods

Methods and Scope

Plan Enrollment: This analysis estimates the share of physicians included in individual Marketplace plans in 2021. In total, 11.4 million enrollees selected a plan on either HealthCare.gov (7.9 million) or a state-based Marketplace (3.6 million). There were 5,120 plans offered, 4,619 of which had at least ten enrollees. Plan enrollment at the county level was estimated by distributing HIOS level enrollment in each county to the associated plans offered in the county, proportional to their total enrollment. An insurer may use the same provider network for several plans, either in different markets or within the same service area. In some areas, insurers may use multiple provider networks across the plan choices they offer.

Provider Network Data: Information on plan provider directories was compiled by Ideon through an API with insurers as well as other data including the National Plan and Provider Enumeration System (NPPES). These files are widely cited as some of the most reliable snapshots of provider directory data currently available (Zhu et al., Oh et al., Politzer et al., Marr et al.). Data for carriers not participating in the Ideon API were supplemented with carriers’ public filings. The 2021 Ideon researcher file contained incomplete information for a limited number of plans. To account for networks with truncated data, we applied several trimming rules. First, networks with fewer than 223 physicians (the 2nd percentile) or without a hospital were excluded. Second, following Zhu et al., networks with fewer than 2.5% of the physicians within the market were excluded. In total, plans with about 2.5% of Marketplace enrollment were excluded from the analysis. The Ideon researcher file was made available with the support of the Robert Wood Johnson Foundation.

Local Markets: Local physicians are defined as those who practice within the same county as an enrollee or are within the distance thresholds specified as part of CMS’s network adequacy standards for HealthCare.gov plans. Thus, a wider area is used for enrollees living in more rural counties, or for specialist physicians. This approach accounts for differences in county geographical sizes and acknowledges that physician markets do not conform to county boundaries. For example, in a Large Metro county, primary care physicians are included in the denominator if they are within the county or 5 miles of its center, compared to 30 miles from the county center in a Rural county. A specialist such as a cardiologist is included in Large Metro counties if they are within the county or 10 miles of its center, or within 60 miles of a Rural county center. While the mileage standards in network adequacy regulations are based on the proximity to plan enrollees, this analysis measures the distance from the population-weighted center of the county.

 

Mileage Thresholds for Defining Local Areas

In total, CMS designates 78 “Large Metro” counties based on their population and population density and 723 “Metro” counties. Most ACA enrollees live in one of these urban county designations, including 38% in Large Metro counties and 48% in Metro counties. For example, Large Metro areas are classified as counties with at least one million people and a population density of at least 1,000 people per square mile, or counties with 500,000-999,999 people and a population density of at least 1,500 people per square mile, or counties with a population density of at least 5,000 people. The county classifications follow the definitions used in the Medicare Advantage network adequacy rules (Table 3-1) and are available in the 2022 Health Service Delivery file published by CMS.

The Vast Majority of Marketplace Enrollees Lived in Large Metro and Metro Counties

Active Physician Workforce: Private health plan network directories often include many physicians who may have retired or are otherwise no longer working in patient care. To estimate the total number of physicians who are in active practice, we relied on Medicare Data on Provider Practice and Specialty (MD-PPAS), a federal database of physicians who submitted at least one Medicare Part B claim in 2021 and therefore saw at least one Medicare patient in the year. Medicare Part B is the largest payer of physician services, covering some 58 million people ages 65 and older and younger adults with disabilities. Virtually all non-pediatric physicians participate in the program, with about 1% formally opting out altogether. In total, 680,000 physicians, including 181,000 primary care doctors, were included in MD-PPAS in 2021.

Estimates of the number of physicians vary. The Health Resources and Services Administration (HRSA) and the Association of American Medical Colleges (AAMC) use the American Medical Association (AMA) Physician Masterfile, which contains 947,000 physicians. The Bureau of Labor Statistics estimates there were 762,000 jobs for physicians and surgeons in 2021. Many of these physicians may be engaged in jobs that do not require them to regularly see patients, including research, administration, or management roles. As a proxy for the number of physicians working in patient care, this analysis is limited to physicians who submitted at least one Medicare Part B claim in the year. This may include at least some physicians who are in roles where they only occasionally see patients.

Physician Specialties and Addresses: Specialties are defined in MD-PPAS based on physician submissions to the Provider Enrollment, Chain, and Ownership System (PECOS) as well as which claims were submitted to Medicare. Throughout, MD-PPAS’s discrete classification of five specialties (psychiatry, primary care, medical specialist, surgical specialist, and hospitalist) is used. A provider is considered a hospitalist if they self-report being a hospitalist in PECOS, or if more than 90% of their claim lines were delivered on an inpatient basis. In total, 29% of physicians in the MD-PPAS file are defined as hospitalists. Sub-specialties, including those listed in the appendix, are defined using a provider’s primary National Uniform Claim Committee (NUCC) taxonomy code in NPPES. Specialty groupings are defined using the definitions specified in CMS’s network adequacy template.

Within and across network directories, physicians may have multiple address listings, in some cases because they have several practice locations, and in other cases, plan directories may have incorrect or obsolete address information. Physicians were assigned to a single address based on the location in which they were listed most frequently across directories for plans with at least 25 enrollees.

Plan Premiums: The percentage change in premiums associated with physician network breadth is calculated by modeling the logged-transformed premium for a Silver plan for a 40-year-old enrollee across all 38,000 Marketplace plan/county combinations. The model holds constant plan type and local market characteristics through fixed effects within the rating area. Plans with physician penetration rates between 25% and 50% and more than 50% have significantly higher premiums than plans narrower than these thresholds (p <.0001 and p <.0001, respectively). There is considerable variation in the premiums. Factors other than physician networks, including market characteristics and hospital networks, are also related to plan premiums.

Limitations: A central challenge in analyzing provider networks is determining the size of the physician workforce. While the vast majority of physicians engaged in active practice accept Medicare, some physicians may be inadvertently missed, including those in closed-network HMOs serving exclusively commercial populations or those specializing in services not typically used by Medicare enrollees. Telehealth providers whose addresses are not within the local market are also excluded. Further, this analysis only considers individual-level physicians enumerated in the plan directory. In some cases, plans may include group health practices in their networks and not individually list providers.

Conversely, this analysis may exaggerate the breadth of provider networks. “Phantom providers,” or physicians who are listed in the plan directory but no longer accept the plan, may artificially increase the breadth of some plans. One secret shopper study documented that 10% of physicians in California Marketplace plans’ provider directories had never accepted the plan. Additionally, this analysis includes all providers whose NPIs are in the directory; some providers may be included in the network for some services or at some locations but not others.

Inherently, this analysis only measures whether providers are included in a plan’s network, not their availability for enrollees wishing to schedule appointments. Some relatively small networks may have physicians with smaller panels and be able to better accommodate enrollees’ health needs than relatively broad networks in which providers are included in a large number of plans.

Appendix

Most Enrollees Had In-Network Access to Fewer than Half of the Primary Care Doctors in Their Area
Marketplace Enrollees in Large Metro Counties Had In-Network Access to a Smaller Share of Specialists Than Less Urban Counties
The Average Marketplace Enrollee Had In-Network Access to Fewer Than Half of Psychiatrists in Their Area
The Average Breadth of Marketplace Physician Networks Varied Across States
The Average Breadth of Marketplace Physician Networks Varied Across Counties
Neither Metal Level Nor Plan Type Were Reliable Indications of Network Breadth
Percent of Adults Who Delayed or Did Not Get Medical Care Because They Couldn't Find a Doctor Who Accepts Their Insurance, 2022

Endnotes

  1. Hospitalists are specialists whom consumers are less likely to have the opportunity to choose, and who are more likely to charge higher out-of-network fees. However, the No Surprises Act protects Marketplace consumers, in some circumstances, from so-called surprise out-of-network bills from these physicians. It is yet to be seen whether this law will encourage more hospital-based physicians to join provider networks. ↩︎
  2. Compared to other specialties, a higher share of psychiatrists have opted out of Medicare (7.7%). In addition, psychiatrists only billing private payers are not included in the MD-PPAS data and as a result, are not included in this analysis. This analysis includes only physicians and so does not measure network inclusion of other non-physician mental health professionals such as psychologists, clinical social workers, or counselors. A higher share of psychiatrists are not accepting new patients covered by Medicare than other specialties. ↩︎
  3. In addition, other state policy decisions impact Marketplace enrollment. For example, while six counties in New York State rank in the top 30 by population, none rank in the top 30 counties by Marketplace enrollment because some New Yorkers are eligible for the Basic Health Plan. ↩︎
  4. In total, 921 doctors were practicing in Hidalgo County, TX, a quarter of the workforce of similar size counties. Among the 30 counties with the largest Marketplace enrollment, four (Clark County, NV; Hidalgo County, TX; Riverside County, CA; and Osceola County, FL) have census population to primary care physician ratios that place them in the top 25% of most underserved counties in the U.S. HHS Area Resource File identified 406 primary care physicians (M.D. and D.O.) in Hidalgo, fewer than the rest of the counties with high enrollment except Osceola County, FL. Health Resources and Services Administration (HRSA) identifies part of all 30 counties as primary care shortage areas, with the exception of Collin County, TX (Plano). ↩︎
  5. People of color include people of any race who identify as Hispanic or those that identify as Black/African American, AIAN, or NHOPI, or two or more races. County classifications are from the CDC/ATSDR Social Vulnerability Index. ↩︎
  6. Studies have illustrated that the number of plans available to enrollees continues to increase. Assistant Secretary for Planning and Evaluation (ASPE) found that the average number of plans in HealthCare.gov states increased from approximately 26 in 2019 to 61 in 2021 and 108 in 2022. This average is weighted to county-level Marketplace enrollment. When considering only plans with enrollment, the average Marketplace consumer had access to 56 plans in 2021. ↩︎
  7. Maine Marketplace plans were part of this pilot until opening its own state-based Marketplace website. ↩︎
  8. Companies and related subsidiaries were grouped by their parent or group affiliation using data obtained from HHS Medical Loss Ratio public use files, Mark Farrah Associates Health Coverage Portal TM, and supplemented with additional research, including insurer press releases and healthinsurance.org. Mark Farrah Associates collects data from plan regulatory filings from the National Association of Insurance Commissioners (NAIC). The level of coordination between subsidiaries sponsored by a single parent varies. This approach follows the method used here and here. BCBS member organizations are grouped together based on the categorization by Mark Farrah. For the purposes of this report, Elevance (formerly Anthem) is not included with other BCBS plans. ↩︎
  9. In total, 58% of Marketplace enrollees were covered by a plan whose parent company had at least some Medicaid enrollment in their state. Insurers that also had Medicaid enrollment participated in 74% of counties nationally. Insurers in states without comprehensive, risk-based MCOs (AL, AK, CT, ID, ME, MT, OK, SD, VT, WY) were not classified as MCOs. Insurers were classified as sponsoring an MCO if they had Medicaid enrollment in Mark Farrah Associates Health Coverage PortalTM. ↩︎
  10. Graves et al., similarly used Ideon 2019 data and found that large group plans cover 57.3% of primary care physicians compared to 45.7% in Marketplaces. This analysis considers a different definition of local markets, and does not weight network breadth by enrollment.   ↩︎
News Release

KFF Analysis Finds Physician Networks in ACA Marketplace Plans Vary Widely, and Enrollees Typically Pay More in Premiums to Access Broader Networks

A Quarter of Practicing Physicians Did Not Participate in Any Marketplace Plans in 2021

Published: Aug 26, 2024

A KFF analysis of physician networks in the Affordable Care Act’s Marketplace plans finds wide variations in the share of local practicing physicians who participate, with the least costly plans generally having a smaller share of physicians than more expensive plans.

The analysis examines the breadth of physician networks listed in Marketplace plan directories in 2021 in nearly every county nationally in relation to the number of actively practicing physicians locally. 

On average, Marketplace enrollees had access to 40% of practicing physicians in their area in 2021, with wide variations across plans. For instance, 23% of enrollees were in plans with no more than a quarter of local doctors, while only 4% were in plans with at least three-quarters of local doctors.

Plans with networks that include a larger share of local physicians typically have higher premiums. For example, Marketplace silver plans with at least half of local participating doctors in their networks on average cost 8% more than plans with less than a quarter of participating doctors on average.This can lead to savings for consumers comfortable with a narrower doctor network, but higher costs for enrollees who want access to plans with a broader network. For example, in counties where the two lowest cost plans have narrower networks (up to a quarter of doctors) but one with a broader network is offered (at least half of doctors), it would cost $95 extra per month to enroll in the broader network plan.

The breadth of a plan’s network can be a factor in enrollees’ ability to access care.  A 2023 KFF survey found that one in five consumers enrolled in Marketplace plans said that a provider they needed was not covered by their insurance, more than the share with employer coverage who say so.

Other findings include:

  • Marketplace consumers living in large metropolitan areas on average are in networks that include only about a third of local physicians, while those living in rural counties on average are in networks with about half of local physicians. Those differences reflect the smaller pool of providers in many rural areas.
  • Even with higher average participation rates, enrollees in rural counties often have relatively few in-network doctors, particularly specialists. For instance, at least 2.5 million enrollees in rural counties have fewer than 10 dermatologists and fewer than 10 gynecologists in their local area.
  • Marketplace enrollees on average have in-network access to about half of active OB-GYN physicians (55%) and surgical specialists (53%) in their markets, but a smaller share of primary care doctors (43%) and psychiatrists (37%).
  • On average, 27% of active practicing physicians were not in any Marketplace plan networks, with much higher shares in some counties such as Cook County/Chicago in Illinois (60%), Dallas County in Texas (36%), and Lee County/Fort Myers in Florida (41%). Marketplace enrollees who want to see those doctors would not be able to find any plan that includes them in its network.

For consumers, assessing the breadth of Marketplace plans’ networks when choosing a plan can be extremely difficult. In 2021, for example, enrollees on average had a choice of 58 different plans in their county, including variations offered by the same insurer with different provider networks, with no overall way to measure the breadth of each option’s network. In addition, not all providers listed in a plan’s network will be open to taking new patients, further narrowing options for some consumers.

The analysis also examines how much choice consumers have in the breadth of the networks in their county where they live, variations within and across geographic areas, and differences across plan insurers. 

KFF Tracker: U.S. Global Health Programs by Country and Region

Published: Aug 23, 2024

This fact sheet does not reflect recent changes that have been implemented by the Trump administration, including a foreign aid review and restructuring. For more information, see KFF’s Overview of President Trump’s Executive Actions on Global Health.

The U.S. supports global health programs in almost 80 countries, with additional countries reached through its regional efforts and contributions to multilateral organizations.1  In each partner country, U.S. programs often operate in multiple program or health areas, which may include: the President’s Emergency Plan for AIDS Relief (PEPFAR), Tuberculosis (TB), the President’s Malaria Initiative (PMI), Neglected Tropical Diseases (NTDs), Family Planning and Reproductive Health (FP/RH), Maternal and Child Health (MCH), Nutrition, and Global Health Security. This tracker provides an overview of U.S. bilateral global health programs by country and region through:

  • a map of U.S. bilateral global health programs by country (Figure 1);
  • a list of countries where the U.S. operates bilateral global health programs by program area (Table 1); and
  • a summary of the number of countries reached by region and program area (Table 2).

The tracker currently reflects FY 2023 data2  and will be updated annually. See also the KFF tracker for U.S. global health country-level funding.

Map of U.S. Global Health Programs by Country, FY 2023
U.S. Bilateral Global Health Programs by Program Area and Country, FY 2023
Number of Countries Where the U.S. Operates Global Health Programs by Program Area and Region, FY 2023
  1. Number of countries represents countries that received funding directly from the U.S. government as reported on ForeignAssistance.gov; additional countries may be reached through “regional” and “worldwide” programming. ↩︎
  2. Reflects U.S. global health programs by country and region as identified in FY 2023 planned funding data from ForeignAssistance.gov for all global health programs, with the exception of Neglected Tropical Diseases (NTDs), whose countries were identified through the FY 2023 USAID NTD fact sheet and personal communication with USAID. See also KFF’s U.S. Global Health Budget Tracker for more details on planned funding by country. ↩︎

Beyond Cost, What Barriers to Health Care do Consumers Face?

Published: Aug 22, 2024

High cost-sharing and expenses not covered by insurance leave some people with expensive medical bills. But costs are not the only barriers to health care access.

According to KFF’s new analysis, many adults can face logistical barriers to care, like work schedules or finding a doctor in network or appointment. In 2022, about 1 in 5 adults under age 65 experienced at least one barrier to accessing care aside from cost.

The full analysis and other data on health costs are available on the Peterson-KFF Health System Tracker, an online information hub dedicated to monitoring and assessing the performance of the U.S. health system.

VOLUME 5

AI Chatbots as Health Information Sources

This is Irving Washington and Hagere Yilma. We direct KFF’s Health Misinformation and Trust Initiative and on behalf of all of our colleagues across KFF who work on misinformation and trust we are pleased to bring you this edition of our bi-weekly Monitor.


Summary

In this issue, we take a closer look at the reliability of artificial intelligence (AI) chatbots as a source of health information. We explore public opinion on chatbot accuracy based on KFF surveys and highlight recent examples of AI-generated election misinformation in the news. In addition, we share our firsthand experience querying AI chatbots on health topics and discuss research on gaps in safeguards.


Stylized quote card reads: “While most of the attention around AI in health is focused on how it can transform medical practice and create new business opportunities, consumers are also using it, and the jury is still out on whether it will empower or confuse them,” KFF President and CEO Drew Altman said. “At KFF, our focus will be on how AI and other information technologies affect people.”

Latest KFF Health Misinformation Tracking Poll Highlights Public Uncertainty Over AI’s role in Delivering Accurate Health Information

With the growing public interest in artificial intelligence and with many companies integrating AI into their consumer-facing platforms, the latest KFF Health Misinformation Tracking Poll finds that about two-thirds of adults say they have used or interacted with artificial intelligence. While AI may serve as a beneficial tool in efforts to dispel misinformation, it may also increase the spread of false or misleading claims if misused. Notably, when it comes to information provided by AI chatbots, most adults (56%) – including half of AI users – are not confident that they can tell the difference between what is true and what is false (Figure 1).

Most Adults Are Not Confident They Can Tell Whether Information From AI Chatbots Is True or False 

The KFF Health Misinformation Tracking Poll also finds that most adults are not confident that health information provided by AI chatbots is accurate. While about half of the public say they trust AI chatbots, such as ChatGPT, Microsoft CoPilot, or Google Gemini, to provide reliable information on practical tasks like cooking and home maintenance and on technology, fewer say they trust chatbots to provide reliable health (29%) or political information (19%; Figure 2).

Half of Adults Trust AI Chatbots to Provide Reliable Information About Practical Tasks, Technology; Fewer Trust Their Reliability for Health  and Political Information 

At this early stage in the development of consumer-facing, generative AI models, most of the public (55%) are uncertain if these technologies are having a positive or negative impact on those seeking health information online. About one in five adults (23%) say AI is doing more to hurt those seeking accurate health information while a similar share (21%) say it is doing more to help those efforts.


Deeper Dive: How AI Chatbots Are Changing in Handling Health Misinformation

Screenshots of three different chatbot models responding to a question about Ivermectin as an effective COVID-19 treatment.

While some research suggests AI chatbots are just as accurate medical professionals in answering health queries, concerns about biased or inaccurate information persist. To enhance accuracy and reliability, AI chatbots are regularly updated to improve the chatbot’s ability to identify and correct misinformation. Over the past year, developers have trained AI models on larger and more diverse data sets of information, improving AI’s ability to cross-reference information from multiple reliable sources to verify claims and detect inconsistencies.

While some platforms focus on user experience and management tools, the general trend is to use advanced AI techniques to better understand context, protect data accuracy, and provide more reliable information. Both Google and Microsoft have recently renamed their AI chatbots to reflect these improvements: Google’s Bard is now called Gemini, and Microsoft’s Bing Chat has been renamed Copilot. OpenAI has also upgraded ChatGPT, including a new real-time voice interactions, which Axios notes could make people more comfortable using the AI chatbot for health information.

To understand how three well-known AI chatbots – ChatGPT, Google Gemini (formerly Google Bard), and Microsoft CoPilot (formerly Bing Chat) – have changed in how they handle health-related questions, KFF’s Hagere Yilma asked each of the chatbots in November 2023, March 2024, and again in August 2024 whether the 11 false claims examined in the KFF Health Misinformation Tracking Poll were true or false. Below is a summary of her observations (full responses from AI chatbots can be found here). Her observations shared here provide a glimpse into the accuracy and reliability of these chatbots, but only reflect the experience of a single chatbot user and are not generalizable scientific research. Chatbots may give different answers depending on the individual user, the questions asked, and updates to the AI models.

Chatbots Differ in Directness When Addressing False Claims, Often Highlighting Complexity

For the most part, each chatbot pointed out false claims, but sometimes they explained that the statement’s accuracy was more complicated instead of just saying it was false. When we first tested the chatbots, both Google Gemini and Microsoft CoPilot directly refuted false claims, while ChatGPT tended to approach these claims with more caution. Rather than definitively labeling some claims as false, ChatGPT noted the complexity of the issue and the need for further research. For example, when asked if the claim that ivermectin as an effective COVID-19 treatment is true, ChatGPT said that there is still some debate about ivermectin’s effectiveness and suggested that more research is needed, without directly calling the statement false. When we revisited these chatbots in March and August 2024, ChatGPT became more assertive, labeling more claims as false, but still labeled two of the statements about firearms as “not entirely accurate” or “complex” rather than outright refuting it. In March 2024, CoPilot also labeled the same two statements about firearms as “not entirely accurate” or “lacks conclusive evidence.”

Challenges in Citing Sources

The chatbots had different approaches to sharing scientific evidence when supporting their responses. In November 2023 and March 2024, ChatGPT usually mentioned that there is scientific evidence refuting the tested claims but didn’t cite specific studies. For example, when asked if COVID-19 vaccines have caused thousands of deaths in otherwise healthy people, ChatGPT said “The overwhelming evidence from clinical trials and real-world data indicates that the benefits of COVID-19 vaccination in reducing the risk of severe illness, hospitalization, and death far outweigh any potential risks” but did not offer any details about the trials or data it was referring to. On the other hand, Gemini and CoPilot cited specific studies as evidence, but Gemini typically did not provide links and sometimes provided inaccurate details about the studies. CoPilot provided links, but these sometimes led to third-party summaries instead of the actual research, which could make it difficult for users to verify the information for themselves.

Chatbots’ Use of Public Health References Evolves Over Time

Over time, the chatbots showed notable changes in how they reference public health institutions to support their answers. In 2023, ChatGPT took a cautious approach, only citing specific agencies like the CDC or FDA for COVID or vaccine-related questions. For most other health claims, it would generally suggest consulting trusted sources without naming them. For example, when asked if the Affordable Care Act established a government panel to make decisions about end-of-life care for people on Medicare, ChatGPT mentioned “It’s important to rely on accurate and credible sources when evaluating claims about healthcare policies and to avoid misinformation…” but didn’t cite any credible sources. Google Gemini and Microsoft CoPilot, on the other hand, initially referenced specific institutions as trusted sources for most questions in 2023.

By 2024, we observed a shift: ChatGPT began referencing specific institutions across a broader range of health topics, while Gemini shifted to providing general resource links and only for some questions. However, CoPilot maintained consistency throughout the entire period, referencing statistics and recommendations from public health organizations while also including links to a broader range of sources, such as news articles, fact-checking resources, research studies, and practice guidelines.

The Bottom Line

While our observations reflect our own limited test and are not generalizable, there are still a few takeaways to consider. AI chatbots can be a convenient starting point for quick health info, thanks to their speed and ease of use. But they’re not perfect or always reliable. Sometimes these tools give misleading information, misrepresent sources, or leave out important context. To be on the safe side, it’s a good idea to double-check chatbot answers by looking at multiple sources. You should also stay informed about system updates, as chatbot responses may change with each update.


Recent Developments

AI Chatbots Can Also Spread Election Misinformation

Hill Street Studios / Getty Images

The World Economic Forum’s Global Risks Report (2024) identified misinformation and disinformation fueled by generative AI as the leading short-term threat to global stability and democratic processes. Ahead of the 2024 U.S. election, a New York Times (NYT) article demonstrated how easily AI chatbots can be manipulated to spread misinformation. NYT staffers customized chatbots by feeding them millions of social media posts from platforms like Reddit and Parler, allowing the bots to develop both liberal and conservative viewpoints. When asked about the election and other contentious issues, the chatbots generated extreme, biased, and often misleading responses, demonstrating how AI could flood social media with disinformation.

Five Secretaries of State wrote an open letter to Elon Musk, calling for immediate changes to the AI chatbot Grok after it spread false information about Kamala Harris’s eligibility for the 2024 presidential ballot. The letter, led by the Minnesota Secretary of State, highlighted the chatbot’s inaccuracies, such as incorrectly stating that ballot deadlines had passed in several states, which could have misled voters. The secretaries stressed the importance of providing accurate election information and suggested that Grok should direct users to trusted resources such as CanIVote.org.

AI-Fueled Russian Disinformation Campaign Targets Paris Olympics and Boxer Imane Khelif

Maja Hitij (Staff) / Getty Images

AI played a role in a Russian disinformation campaign targeting the 2024 Paris Olympics. According to the Associated Press, AI was used to generate fake images, audio, and video, including a viral video that portrayed Paris as a decaying, crime-ridden city. The video, which featured an AI-enhanced singer mocking the Games, was quickly translated into 13 languages by AI. This disinformation campaign also amplified false gender claims about Algerian boxer Imane Khelif, stemming from her controversial disqualification by a Russian-influenced boxing association. These claims then gained traction online, especially after public figures like presidential candidate Donald Trump weighed in. The campaign to undermine the Olympics demonstrated how AI tools are being used to spread false stories on a global scale.

AlonzoDesign / Getty Images

Research Updates

Inconsistent Safeguards in AI Chatbots Can Lead to Health Disinformation

A study published earlier this year in BMJ evaluated how well large language models (LLMs) could prevent users from prompting chatbots to create health disinformation. It found that while some AI chatbots consistently avoided creating false information, other models frequently created false health claims, especially when prompted with ambiguous or complex health scenarios. In addition, the study found that the safeguards were inconsistent – some models provided accurate information in one instance but not in others under similar conditions. The researchers criticized the lack of transparency from AI developers, who often did not disclose the specific measures they had taken to mitigate these challenges.

Source: Menz, B. D., Kuderer, N. M., Bacchi, S., Modi, N. D., Chin-Yee, B., Hu, T., … & Hopkins, A. M. (2024). Current safeguards, risk mitigation, and transparency measures of large language models against the generation of health disinformation: repeated cross-sectional analysis. BMJ, 384.

ChatGPT Updates Provide More Accurate Answers to Vaccine Myths

A 2023 study published in Vaccines evaluated how well ChatGPT, including both the free GPT-3.5 and the paid GPT-4.0 versions, answered questions about vaccination myths dispelled by the World Health Organization. The study found that GPT-4.0 provided more accurate and clearer answers compared to GPT-3.5, achieving 85.4% overall accuracy. However, both versions occasionally provided misleading information and were not entirely error-free, suggesting that while AI tools can assist in healthcare information, they should be used with caution and supplemented by expert advice.

Source: Deiana, G., Dettori, M., Arghittu, A., Azara, A., Gabutti, G., & Castiglia, P. (2023). Artificial intelligence and public health: evaluating ChatGPT responses to vaccination myths and misconceptions. Vaccines, 11(7), 1217.

About The Health Information and Trust Initiative: the Health Information and Trust Initiative is a KFF program aimed at tracking health misinformation in the U.S., analyzing its impact on the American people, and mobilizing media to address the problem. Our goal is to be of service to everyone working on health misinformation, strengthen efforts to counter misinformation, and build trust. 


View all KFF Monitors

The Monitor is a report from KFF’s Health Information and Trust initiative that focuses on recent developments in health information. It’s free and published twice a month.

Sign up to receive KFF Monitor
email updates


Support for the Health Information and Trust initiative is provided by the Robert Wood Johnson Foundation (RWJF). The views expressed do not necessarily reflect the views of RWJF and KFF maintains full editorial control over all of its policy analysis, polling, and journalism activities. The Public Good Projects (PGP) provides media monitoring data KFF uses in producing the Monitor.


How Did Medicaid Renewal Outcomes Change During the Unwinding?

Authors: Bradley Corallo and Jennifer Tolbert
Published: Aug 21, 2024

After a three-year pause in Medicaid disenrollments, the continuous enrollment provision ended on March 31, 2023, kicking off a long process in which states were required to complete renewals verifying the eligibility of all enrollees in the program. States were permitted to resume disenrollments in April 2023, but some states delayed the start of their unwinding periods until June or July 2023. By August 2024, all but three states have completed their unwinding periods (one additional state, New York, has not set an end date for unwinding). States also had considerable flexibility in how they implemented their unwinding plans, and these decisions likely affected renewal outcomes. For example, 17 states opted to prioritize “likely ineligible” enrollees early in their unwinding period, while some states chose to de-prioritize certain vulnerable populations for later in the unwinding period. States also implemented a substantial number of policy and procedural changes leading up to and during the unwinding, such as adopting federal flexibilities, updating eligibility systems and processes, and experimenting with new outreach strategies and partnerships.

During the unwinding period, states were required to report monthly data on renewal outcomes, providing for the first time, a mechanism for monitoring the renewal process across states. To date, roughly 24 million have been disenrolled and 54 million have had their coverage renewed during the unwinding. As a result, national Medicaid/CHIP enrollment has declined by more than 13 million (13.9%) from its peak at the start of unwinding and stands at 82 million people enrolled as of April 2024. Because enrollment data include new people entering the program as well as people reenrolling after losing Medicaid, the net change in enrollment is smaller than the total disenrolled during the unwinding. April 2024 enrollment is still roughly 10.4 million (14.6%) higher than pre-pandemic enrollment levels of about 71 million, though most states had not yet finished their unwinding periods as of April 2024.

This policy watch uses unwinding data collected through KFF’s Medicaid Enrollment and Unwinding Tracker to examine how national-level renewal outcomes changed over the course of unwinding, including changes in the share of people who had their coverage renewed or were disenrolled from Medicaid each month. States’ unwinding periods have been aligned and are reported as months into the unwinding period (e.g., Month 1, Month 2, Month 3…). For example, Month 1 for Arizona is April 2023 but June 2023 for California. The data reflect updated renewal reports, where available, which follow-up on outcomes for cases initially reported as pending. Due to the lag in reporting updated data, some states’ updated renewal reports are not available for later months.

Over the course of the unwinding period, the share of people whose coverage was renewed increased while the share who were disenrolled dropped. From the start of each state’s unwinding period in Month 1 through Month 12, the share of people who retained Medicaid increased from 58% to 66% (Figure 1). Over the same period, the share of disenrollments declined from 38% to 23%. A sizable drop in the number of people disenrolled each month, from 2.8 million people in the first month to 1.6 million people in the twelfth month, appears to be driving the trends, although the number of people whose coverage was renewed each month increased slightly over the period from 4.1 million to 4.6 million. The decrease in the number of people disenrolled partially reflects states working through their “likely ineligible” populations in the early months of unwinding, as well as states’ policy and procedural changes in response to early unwinding data.

Monthly Medicaid Renewal Outcomes During Unwinding

Among those who retained coverage, the share of people renewed on an ex parte basis increased from 51% in Month 1 to 70% in Month 12 (Figure 2). Ex parte renewals, also known as auto-renewals, require states to verify an enrollee’s eligibility based on data available to the state without requiring the enrollee to submit documentation. The Centers for Medicare and Medicaid Services (CMS) encouraged states to increase ex parte renewal rates – and provided new flexibilities during the unwinding period for them to do so – as a strategy to help eligible people avoid losing Medicaid coverage during the renewal process. Notably, there was a large increase in ex parte renewals in Month 7, jumping from 57% to 68%. The timing of this jump in ex parte renewals reflects many factors, including policy and procedural changes, as well as CMS taking enforcement action in 29 states to address noncompliance with federal ex parte requirements.

The Share of Medicaid Enrollees Renewed on an Ex Parte Bases vs. Renewed via Renewal Form, Among Those Maintaining Medicaid Coverage

Although the total number of disenrollments declined during the unwinding period, among people disenrolled from Medicaid, the share of people disenrolled for procedural reasons remained high. Procedural disenrollments occur when someone’s eligibility (or ineligibility) could not be verified, typically because the renewal process was not completed. During the unwinding, procedural disenrollments as a share of total disenrollments decreased from 74% in Month 1 to 66% in Month 12 (Figure 3). However, while states took steps to reduce procedural disenrollments through increased outreach to enrollees and other policy and procedural changes, roughly two-thirds of people who were disenrolled from Medicaid in Month 12 of unwinding lost coverage for procedural or paperwork reasons, reflecting ongoing issues with the renewal process.

The Share of Medicaid Enrollees Disenrolled for Procedural Reasons vs. the Share Determined Ineligible, Among Those Disenrolled From Medicaid

The national data present a high-level picture of renewal outcomes during the unwinding but mask differences across states. For example, while most states experienced gains in ex parte renewal rates during the unwinding, the magnitude of the changes varied. Similarly, in most states, the procedural disenrollment rate declined, but in some states the rate increased, possibly reflecting pauses in procedural disenrollments in the early months of unwinding or other procedural issues. Differences in state policies, procedures, enrollee populations, and system capabilities are likely the biggest factors driving the variation in renewal outcomes across states. However, more analysis is needed to understand which factors had the largest impact on renewal outcomes.

Ex Parte Renewal Rates Among Those Retaining Medicaid Coverage, States' First Three Months vs. Last Three Months of Unwinding