The Business of Health with Chip Kahn

AI: Rewiring the Machine

July 7, 2026

Video

Audio

Listen On:

Apple

Spotify

YouTube Music

About this Episode

Episode 11, AI Series: In a conversation focused on the technology underlying AI in health care, Chip is joined by Seema Verma, former administrator of the Centers for Medicare & Medicaid Services (CMS), and now Executive Vice President and General Manager at Oracle Health and Life Sciences, the second largest electronic health records (EHR) platform in the U.S. Seema shares her insights on the evolution of EHRs and why it’s necessary to redesign these systems to better integrate AI capabilities and improve the quality of care.

The Host

Headshot photo of Chip Kahn wearing a navy blue suit with a red tie, red pendant on lapel, and glasses.

Chip Kahn

Sr. Visiting Fellow

Charles N. Kahn III is a senior visiting fellow at KFF. He is also a visiting senior fellow at the American Enterprise Institute and a nonresident senior scholar at the University of Southern California’s Schaeffer Center for Health Policy & Economics. He serves as co-chair of the international Future of Health collaborative.

Guest

Seema Verma

Executive Vice President and General Manager, Oracle Health and Life Sciences

Seema Verma is Executive Vice President and General Manager of Oracle Health and Life Sciences, leading global strategy to modernize healthcare through data, connectivity, and AI. Previously, she served as Administrator of the Centers for Medicare & Medicaid Services.  

A recognized industry leader, she serves on multiple healthcare boards and has been named among Modern Healthcare’s Most Influential People and Becker’s Healthcare’s Great Leaders in 2026. She has a bachelor’s degree in life sciences from the University of Maryland and a master’s degree in public health from John’s Hopkins University. 

PREVIOUS EPISODES

Guardrails for AI in Health Care — How High?

The AI Arms Race in Administrative Health Care

AI: Show Me the Outcomes

SERIES

The Business of Health with Chip Kahn

This weekly podcast features insightful conversations between host Chip Kahn and his guests, who discuss the business of health care, connecting the dots between the health care business, policy, and patients.

The podcast’s first series on AI in health care illuminates how AI is changing health care, and features guests who are deploying this technology, managing its consequences, and designing policy around it.

More On
Other Health Artificial Intelligence Care Coordination Delivery System Health I.T. Health System Performance Health Workforce Hospitals Medical Technology Physicians Treatment

Medicare Advantage Insurers Deny Prior Authorization Requests for Post Acute Care at Substantially Higher Rates Than the Overall Denial Rate

Published: Jul 6, 2026

Prior authorization practices by health insurers have come under scrutiny in recent years, in part spurred by the public sentiment that delays and denials of care are a problem. According to KFF polling, about seven in ten insured adults say prior authorization is a burden. New evidence from the Office of Inspector General (OIG) within the Department of Health and Human Services documents the high rate of denials of prior authorization requests for certain post-acute care services in Medicare Advantage plans, which now enroll more than half of all Medicare beneficiaries.

The OIG recently published two reports finding that Medicare Advantage insurers deny more than half of all prior authorization requests for the most expensive types of post-acute care, including 65% of requests for stays in long-term care hospitals (LTCHs) and 54% of requests for stays in inpatient rehabilitation facilities (IRFs), as well as 12% of requests for stays in skilled nursing facilities (SNFs). These denial rates are higher, and in the case of LTCHs and IRFs substantially higher, than the overall Medicare Advantage prior authorization denial rate found by KFF in previous analysis of less than 8% for all services (Figure 1). The OIG also found substantial variation across insurers, highlighting the heterogeneous experience Medicare Advantage enrollees could face depending on the private insurer that administers their Medicare benefits.

Medicare Advantage Insurers Deny Prior Authorization Requests for Post-Acute Care at Substantially Higher Rates Than the Overall Denial Rate (Bar Chart)

Medicare Advantage Insurers Deny Prior Authorization Requests for Post-Acute Care at Substantially Higher Rates Than the Overall Denial Rate

Share of prior authorization requests that were initially denied by Medicare Advantage insurers, 2024

Values for long-term care hospitals, inpatient rehabilitation facilities, and skilled nursing facilities reflects June 2024 data reported to the HHS OIG by Medicare Advantage insurers with at least 200,000 enrollees. Value for all services reflects full year data reported to CMS as part of the Part C and Part D reporting requirements.

KFF analysis of Medicare Limited Data Set, Contract Year 2024 Part C and Part D Reporting Requirements, the Department of Health and Human Services Office of the Inspector General.

Insurers use prior authorization to reduce the use of unnecessary or low-value care and to restrain costs. KFF analysis shows that virtually all Medicare Advantage enrollees are in a plan that requires prior authorization for at least some services – most often, high-cost services. For example, in 2026, 95% of Medicare Advantage enrollees are in a plan that requires prior authorization for skilled nursing facility stays. According to the Medicare Payment Advisory Commission (MedPAC), the average Medicare payment in 2023 for traditional Medicare beneficiaries was $43,000 per LTCH stay, $24,000 per IRF stay, and $16,000 per SNF stay.

In 2024, insurers made nearly 53 million prior authorization determinations for Medicare Advantage enrollees. In contrast, prior authorization is rarely used in traditional Medicare (notwithstanding a new Innovation Center model testing the use of AI tools to conduct prior authorization for a limited set of services in traditional Medicare). The new OIG findings suggest the burden of delays and denials from the use of prior authorization is greater for Medicare Advantage enrollees with higher health needs and in more fragile condition. LTCHs generally treat patients with multiple serious conditions, providing services such as respiratory therapy, head trauma treatment, and pain management over the course of hospital stays that extend more than 25 days, on average. IRFs provide intensive rehabilitation services, including for people recovering from strokes or brain injuries. The initial denial of the prior authorization request meant that the requested post-acute care was delayed between 5 and 6 days, on average. In addition to having potential health implications for enrollees seeking post-acute care, the delay could mean higher out-of-pocket spending for the associated hospital stay, because many Medicare Advantage enrollees face daily cost-sharing requirements for hospital stays.

Additionally, the OIG found that when denials were appealed – which happened for 36% of LTCH denials, 31% of IRF denials, and 18% of SNF denials – the requested service was ultimately approved much of the time for LTCHs (36%) and IRFs (43%), and virtually all of the time for SNFs (95%). The extremely high rate of overturning the initial decision upon appeal for SNFs raises questions about whether this care is being routinely inappropriately denied. At the same time, if insurers anticipate that only a relatively small number of initial denials will be appealed, the high overturn rate could reflect a determination by insurers that reversing an initial denial is preferable to having the appeal continue to the next stage. At that point, an independent review entity (IRE) would hear the case, and if the IRE disagrees with the Medicare Advantage insurer’s initial determination to deny a service, that would have a negative impact on a plan’s star ratings.

The findings in the OIG reports are consistent with a previous Senate investigation that found the largest Medicare Advantage insurers denied prior authorization requests for post-acute care at substantially higher rates than other services between 2019 and 2022. Together, these reports underscore the value of having service level data on the use of prior authorization in Medicare Advantage. However, detailed data on the use of prior authorization and denial rates by type of service in Medicare Advantage are not yet required to be reported and therefore not routinely available. The lack of detailed data on prior authorization requests, denials, and appeals has made it difficult to understand the impact on people seeking care and to assess whether initiatives, such as the pledge taken by several private insurers last summer to improve the prior authorization process, are leading to meaningful change. CMS introduced a pilot program to collect more detailed data at the plan and service level this year and anticipates requiring this information beginning in 2027. Nevertheless, it will be several years before those data are available.

More On
Medicare Health Care Utilization Medicare Advantage Prior Authorization

Related Content

Medicare Advantage Insurers Made Nearly 53 Million Prior Authorization Determinations in 2024

Prior Authorization Metrics Provide New Insights into Insurer Practices, but Gaps Remain

Medicare Advantage in 2026: Premiums, Out-of-Pocket Limits, Supplemental Benefits, and Prior Authorization

Medical Frailty and Medicaid Work Requirements: Challenges for People with HIV

Published: Jul 1, 2026

On June 1, 2026, the Centers for Medicare and Medicaid Services (CMS) issued an interim final rule providing states with guidance for implementing Medicaid “community engagement” or Medicaid work requirements as part of the 2025 federal budget reconciliation law. The law requires states to condition Medicaid eligibility for enrollees with coverage through the Affordable Care Act (ACA) expansion or under certain waivers on meeting these requirements or qualifying for an exclusion, including one related to being “medically frail or otherwise” having “a special medical need.” In defining medical frailty, the rule introduces a two-part test, requiring an enrollee to both have a qualifying condition and demonstrate that the condition impairs their ability to fulfill the community engagement requirement, differing from stakeholders’ expectations. Early on, and as with Nebraska’s early implementation, states believed they would be able to exclude people based on presence of a condition alone and several states planned to exclude all people with HIV.

On June 29, 2026, twenty-four (24) states and two (2) state governors sued CMS in Massachusetts District Court challenging aspects of the regulation, including its requirement that to qualify for the medical frailty exclusion an enrollee’s condition must significantly impair their ability to comply with community engagement requirements. Among other arguments, the plaintiff states claim this additional requirement is contrary to the reconciliation law (H.R.1) and that “H.R. 1’s broad statutory exclusions exist for good reason. People with disabilities, patients in the middle of cancer treatment, or those struggling with another serious or complex health condition, shouldn’t be at risk of losing the care that helps maintain their health.” Whether the court grants their request to enjoin and vacate the challenged provisions, including the medical frailty two-part test, is yet to be seen.

In the meantime, because Medicaid is the primary source of insurance coverage for people with HIV, this new requirement and state implementation decisions, will have a significant impact on this population’s access to Medicaid going forward and could affect the nation’s efforts to address HIV.

This analysis reviews the implications of the rule’s definition of medical frailty for people with HIV. (For a broad overview of medical frailty, definition of terms, and the impact of the regulation beyond HIV, see this KFF analysis.)

Medicaid expansion is the most common pathway for Medicaid coverage for people with HIV, so many with HIV will be subject to new work / community engagement requirements. Medicaid is the largest source of insurance coverage for people with HIV and plays a larger role in covering adults with HIV than adults without HIV. Nationwide, nearly half (46%) of people with HIV had coverage through the Medicaid program in 2023. In states that have expanded their Medicaid programs, Medicaid expansion is the primary pathway to coverage for people with HIV. In 2023, 60% of adults under age 65 with HIV in expansion states had coverage through the expansion pathway and would be subject to work requirements (see Figure 1). Before states expanded Medicaid programs under the ACA, many people with HIV did not have access to affordable coverage until they had an advanced condition to qualify through a permanent disability pathway and many were uninsured, despite coverage and access to care having the potential to stave disability off in the first place.

In Medicaid Expansion States, Six in Ten (60%) Medicaid Enrollees with HIV Have Coverage Through the Expansion Pathway (Stacked column chart)

In Medicaid Expansion States, Six in Ten (60%) Medicaid Enrollees with HIV Have Coverage Through the Expansion Pathway 

Distribution of adult Medicaid enrollees with and without HIV, by eligibility group and state Medicaid expansion status, 2023

This analysis reflects enrollees ages 19 to 64 with at least one month of Medicaid enrollment. In 2023, there were 332,000 enrollees with HIV and 57.5 million enrollees without HIV. As of December 2023, there were 41 expansion states, including DC. Enrollees with HIV were flagged if they had a relevant ICD 9 or ICD 10 code, any drug claims for a single tablet regiment (STR), or two or more other non-STR ARVs.

KFF analysis of the T-MSIS Research Identifiable Files, 2023 (Preliminary)

The approach to determining medical frailty specified in the rule will make it more difficult for individuals with HIV to obtain a medical frailty exclusion from work requirements. The rule imposes a two-part test that defines as medically frail an individual who is blind or disabled; has a substance use disorder; has a “disabling” mental disorder; has a physical, intellectual, or developmental disability that limits the ability to perform one or more activities of daily living (ADL); or has a “serious or complex” medical condition and whose condition impairs their ability to fulfill the community engagement requirements (including but not limited to work). While the rule includes HIV/AIDS as one of 19 example conditions that would be reasonable for states to consider as a serious or complex medical condition, it further specifies, using HIV as an example, that, “Individuals with HIV/AIDS are medically frail if they are determined to have a serious or complex medical condition that significantly impairs the individual's ability to comply with the community engagement requirement, which is less likely to be the case if the acuity of their condition is not severe.” Access to antiretroviral medication, including through Medicaid, is necessary to manage HIV and prevent immune system dysfunction, illness, and ultimately death. To the extent people with HIV lose access to Medicaid due to work requirements, including failure to navigate reporting rules, they may develop more severe conditions. 

People with HIV whose condition is well managed may qualify as medically frail if they have another medical condition that limits their ability to work. Nearly three-quarters (73%) of people with HIV enrolled in Medicaid have chronic conditions (other than HIV), compared with four in ten (42%) of those without HIV. Alternatively, people with HIV may qualify as medically frail under one of the other categories. Notably, people with HIV are more likely than other Medicaid enrollees to have an SUD or mental health condition. Nearly, one-quarter (23%) of people with HIV had an SUD diagnosis compared to 8% of Medicaid enrollees without HIV and over one-third (36%) of Medicaid enrollees with HIV had a mental health condition diagnosis, some of which may be considered “disabling,” compared to 16% of Medicaid enrollees without HIV. People with HIV also experience high rates of disability--half (50%) of likely Medicaid expansion enrollees with HIV have a disability, including a functional disability (e.g. difficulty climbing stairs, dressing oneself, etc.) or an AIDS (stage III HIV) diagnosis.

The rule requires states to use claims and encounter data to identify individuals who may be medically frail, but does not provide guidance on how to assess whether a condition limits an individual’s ability to meet the community engagement requirements. States will be required to maintain an auditable list of medically frail conditions (in the form of diagnosis codes) that could include HIV. States will have discretion over the creation of this list and may include only some codes for each condition. For example, Nebraska’s list of ICD-10 codes released before the state implemented work requirements on May 1, 2026 only included one of several codes for HIV which would not capture all enrollees with HIV. The rule also makes clear that diagnoses alone cannot be used to determine medical frailty because of the need to assess whether the condition impairs the ability to work or engage in community service. This additional requirement will limit the ability to verify medical frailty on an automated, or ex parte, basis and will require states to use other verification methods.

The regulation offers examples of the types of providers that states could use to verify medical frailty including a range of clinicians. CMS’s inclusion of “clinical social workers” on this list could be especially meaningful for people with HIV given that many get care through clinics with integrated social and support services whose staff help with insurance navigation. However, the administrative burden on treating providers is likely to be significant

For people with HIV, the reliance on data sharing, confirmation from treating providers, and health screeners and self-attestation to verify medical frailty exclusion status may raise unique privacy issues and barriers due to stigma: 

  • Data sharing: Some states are exploring using a data-sharing process between the state Medicaid and state public health/HIV office which could help them identify enrollees without HIV related claims histories, including those new to Medicaid. Some states already have a data sharing agreement in place. However, this public health data is highly sensitive, and some have raised concerns about data privacy and security related to HIV status. 
  • Provider documentation. As noted, the rule permits states to accept documentation of qualifying conditions and medical frailty from providers. However, the requirement to assess and report the severity of patients’ conditions and the impact on their ability to meet the work requirements may raise ethical concerns for these providers, particularly given the emphasis in HIV care on care engagement for both the patient’s and public health.
  • Health screeners and self-attestation. The rule encourages states to use health screeners at application and renewal to identify individuals who may be medically frail, which could include people with HIV. Separately, though its use will be more limited starting in January 2028, most states will also likely allow self-attestation when existing data sources are insufficient to document a qualifying condition and the inability to work. However, the stigma associated with HIV may discourage individuals from disclosing their condition and how it impacts their life.

Coverage loss for people with HIV could negatively impact individual health, public health, and place an increased burden on already stretched HIV programs. Given the new requirements in the regulation, a blanket exclusion for people with HIV will not be possible which will mean a greater staff burden (at the state Medicaid agency and in clinics), higher costs, and potentially wide scale churn, disenrollment, or coverage rejections for those with HIV. While earlier KFF research found that one-third (33%) of likely expansion enrollees with HIV were working at least 20 hours per week and another 4% had dependents at home, the need to document work compliance or medical frailty status, could challenge coverage retention for people with HIV which could lead to disruptions in care and treatment and subsequently increased risk of morbidity, mortality, and HIV transmission. Such a scenario also runs counter to federal goals in the Administration’s Ending the HIV Epidemic Initiative and the Ryan White Program Moving Forward (formerly Ryan White Program 2030) vision. Indeed, four in ten new HIV transmissions are associated with someone who is aware of their HIV status but not in care. Treatment interruptions can also lead to antiretroviral resistance, making future treatment and care more complex. Additionally, if people with HIV lose Medicaid coverage some may turn to the federal Ryan White Program. This comes at a time when state Ryan White Programs across the country are facing budget crises due to a range of factors and coverage losses due to work requirements represent an additional challenge for programs to weather.

More On
HIV/AIDS Medicaid Access to Care AIDS Drug Assistance Program (ADAP) Chronic Diseases Eligibility HIV/AIDS in U.S. Low Income Public Health

Related Content

The Medical Frailty Exemption from Medicaid Work Requirements: Key Takeaways from the CMS Interim Final Rule

The Medical Frailty Exemption from Medicaid Work Requirements: Key Issues to Watch for in Upcoming CMS Guidance

Domestic HIV Funding in the White House FY2027 Budget Request

KFF Tracker: America First MOU Bilateral Global Health Agreements

Published: Jul 1, 2026

Editorial Note: Originally published on January 13, 2026, this resource will be updated as needed, most recently on July 1, 2026, to reflect additional developments.

On September 18, 2025, the U.S. government (USG) released its new America First Global Health Strategy, which details how the U.S. will engage in global health efforts moving forward. As part of this new strategy, the U.S. has announced that it will be establishing bilateral health cooperation agreements with countries that receive U.S. global health assistance. These agreements, or Memorandums of Understanding (MOUs), between the U.S. and partner countries represent five-year plans (for the period 2026-2030) outlining U.S. engagement in each country’s health efforts with the goal of “helping countries move toward more resilient and durable health systems.” Central to these plans is transitioning country programs from U.S. assistance to long-term country ownership, with a pledge from each partner country to increase its domestic health spending, or co-investment in health, over the next five years as the U.S. decreases its health assistance. The U.S. began signing these agreements in late 2025 and this process is ongoing. Implementation is slated for later this year.

This tracker provides an overview of the MOUs signed to date. Data are based on press releases issued by the State Department, U.S. embassies, and partner country Ministries of Health, as well as MOU documents (if publicly available). See Methods for more information. This tracker will be updated as agreements are signed and more data become available.

USG Global Health MOUs by Country (Table)

USG Global Health MOUs by Country

Data as of July 1, 2026.

*Global health MOU agreements for Bolivia, Botswana, Panama, and South Sudan are for a three-year period.<br>**Exact agreement date unknown for Madagascar.<br>***The Philippines did not sign an official MOU but rather a "Strategic Objective Agreement." While an overall funding amount is available, co-financing details are not yet publicly available.

KFF analysis of U.S. Department of State, U.S. embassy, and Ministries of Health press releases, and MOU documents. MOU source documents from Emily Bass's Substack, Public Citizen, and Think Global Health. • <a href="https://files.kff.org/Global-Health-Policy/America-First-MOU-Bilateral-Global-Health-Agreements-Tracker/Table-1.xlsx&quot; download="Table-1.xlsx" target="_blank" rel="nofollow noopener">Get the data</a>

Signed USG Global Health MOUs by Country (Choropleth map)

Signed USG Global Health MOUs by Country

Number of countries with signed agreements: 34 countries Data as of July 1, 2026.

Signed USG global health MOU agreements are for the 5-year period, 2026-2030, with the exception of Bolivia, Botswana, Panama, and South Sudan which have a signed USG global health MOU agreement for a 3-year period. The Philippines did not sign an official MOU but rather a "Strategic Objective Agreement."

KFF analysis of U.S. Department of State, U.S. embassy, and Ministries of Health press releases, and MOU documents.

Global Health MOU Funding by Country (Bar Chart)

Global Health MOU Funding by Country

Number of countries with signed agreements: 34 countries Total funding announced for the 5-year period (2026-2030): $24.3 billion Data as of July 1, 2026.

Totals represent U.S. and country co-investment funding. Global health MOU agreements for Bolivia, Botswana, Panama, and South Sudan are for a three-year period. The Philippines did not sign an official MOU but rather a "Strategic Objective Agreement."

KFF analysis of U.S. Department of State, U.S. embassy, and Ministries of Health press releases, and MOU documents.

USG Global Health MOU Co-Financing Share by Country (Stacked Bars)

USG Global Health MOU Co-Financing Share by Country

Number of countries with signed agreements: 34 countries Overall country co-investment share for the 5-year period (2026-2030): 39% Click on the buttons below to see data for different charts: Share &nbsp  Amount &nbsp  Data as of July 1, 2026.

Global health MOU agreements for Bolivia, Botswana, Panama, and South Sudan are for a three-year period. *The Philippines did not sign an official MOU but rather a "Strategic Objective Agreement." While an overall funding amount is available, co-financing details are not yet publicly available.

KFF analysis of U.S. Department of State, U.S. embassy, and Ministries of Health press releases, and MOU documents.

USG Global Health MOU Co-Financing Amount by Country

Number of countries with signed agreements: 34 countries Click on the buttons below to see data for different charts: Share &nbsp  Amount &nbsp  Data as of July 1, 2026.

Global health MOU agreements for Bolivia, Botswana, Panama, and South Sudan are for a three-year period. *The Philippines did not sign an official MOU but rather a "Strategic Objective Agreement." While an overall funding amount is available, co-financing details are not yet publicly available.

KFF analysis of U.S. Department of State, U.S. embassy, and Ministries of Health press releases, and MOU documents.

USG Global Health MOU Program Areas by Country (Table)

USG Global Health MOU Program Areas by Country

Data as of July 1, 2026.

Global health program area information based on publicly available information but may not reflect the full scope of areas to be addressed. Global health MOU agreements for Bolivia, Botswana, Panama, and South Sudan are for a three-year period. *The Philippines did not sign an official MOU but rather a "Strategic Objective Agreement."

KFF analysis of U.S. Department of State, U.S. embassy, and Ministries of Health press releases, and MOU documents.

Historical vs. Proposed 5-Year USG Global Health MOU Funding by Country (Grouped Bars)

Historical vs. Proposed 5-Year USG Global Health MOU Funding by Country

Number of countries with signed agreements: 34 countries Total change (and percent change) in 5-year global health funding among countries with agreements: -$7.3 billion (-34%) Click on the buttons below to see data for different charts: Funding Amount &nbsp  Funding Change &nbsp  Percent Change &nbsp  Data as of July 1, 2026.

Historical funding totals represent appropriated and planned amounts for FY21-23, and requested amounts for FY24-25 for the Global Health Programs (GHP) account. Represents U.S. funding only; does not include country co-investment funding.<br>*Global health MOU agreements for Bolivia, Botswana, Panama, and South Sudan are for a three-year period. For Botswana and South Sudan, prior funding represents planned global health funding for FY23-25. Due to limited data, Bolivia and Panama comparisons cannot be provided.<br>**The Philippines did not sign an official MOU but rather a "Strategic Objective Agreement." While an overall funding amount is available, co-financing details are not yet publicly available.

KFF analysis of U.S. Department of State, U.S. embassy, and Ministries of Health press releases, and MOU documents.

Historical vs. Proposed 5-Year USG Global Health MOU Funding, Funding Change by Country

Number of countries with signed agreements: 34 countries Total change (and percent change) in 5-year global health funding among countries with agreements: -$7.3 billion (-34%) Click on the buttons below to see data for different charts: Funding Amount &nbsp  Funding Change &nbsp  Percent Change &nbsp  Data as of July 1, 2026.

Historical funding totals represent appropriated and planned amounts for FY21-23, and requested amounts for FY24-25 for the Global Health Programs (GHP) account. Represents U.S. funding only; does not include country co-investment funding. <br>*Global health MOU agreements for Bolivia, Botswana, Panama, and South Sudan are for a three-year period. For Botswana and South Sudan, prior funding represents planned global health funding for FY23-25. Due to limited data, Bolivia and Panama comparisons cannot be provided.<br>**The Philippines did not sign an official MOU but rather a "Strategic Objective Agreement." While an overall funding amount is available, co-financing details are not yet publicly available.

KFF analysis of U.S. Department of State, U.S. embassy, and Ministries of Health press releases, and MOU documents.

Historical vs. Proposed 5-Year USG Global Health MOU Funding, Percent Change by Country

Number of countries with signed agreements: 34 countries Total change (and percent change) in 5-year global health funding among countries with agreements: -$7.3 billion (-34%) Click on the buttons below to see data for different charts: Funding Amount &nbsp  Funding Change &nbsp  Percent Change &nbsp  Data as of July 1, 2026.

Historical funding totals represent appropriated and planned amounts for FY21-23, and requested amounts for FY24-25 for the Global Health Programs (GHP) account. Represents U.S. funding only; does not include country co-investment funding. <br>*Global health MOU agreements for Bolivia, Botswana, Panama, and South Sudan are for a three-year period. For Botswana and South Sudan, prior funding represents planned global health funding for FY23-25. Due to limited data, Bolivia and Panama comparisons cannot be provided.<br>**The Philippines did not sign an official MOU but rather a "Strategic Objective Agreement." While an overall funding amount is available, co-financing details are not yet publicly available.

KFF analysis of U.S. Department of State, U.S. embassy, and Ministries of Health press releases, and MOU documents.

Methods

This tracker provides information on U.S. MOU bilateral global health agreements to date. Information is sourced from publicly available U.S. Department of State, U.S. embassies, and partner country Ministries of Health press release statements and MOU texts, and will be updated as more information becomes available and when additional agreements are signed. Currently, MOU text, which contains the most detailed information of these sources, is publicly available for only a limited number of countries; for these countries, data were sourced directly from these MOU documents. For countries with available MOU documents, overall totals are based on the sum of annual amounts presented in the text. 

Program areas are captured using keyword searches; for global health security (GHS) specifically, country agreements were categorized as targeting GHS if they specifically mentioned GHS, or if they included descriptions of outbreak preparedness and response activities and containing health threats. Due to the limited nature of press release statements, this tracker may not comprehensively capture the global health program areas targeted in each country’s agreement.

More On
Global Health Policy Financing Global Health Aid Cuts

Related Content

Analysis of U.S. and Global Fund Funding Reductions in MOU Countries

U.S. Global Health Country-Level Funding Tracker

The Trump Administration’s Foreign Aid Review: Reorganization of U.S. Global Health Programs

How Has Projected Medicaid Spending and Enrollment Changed Since Passage of the 2025 Reconciliation Law?

Published: Jul 1, 2026

The Congressional Budget Office (CBO), known as Congress’s “scorekeeper,” projects federal spending and revenues over the next decade and cost estimates of proposed legislation are measured against those projections. Those projections include spending on major federal programs, such as Medicaid. CBO also typically releases a detailed baseline for federal spending on Medicaid that includes estimates of enrollment by eligibility group and spending by service category. The 2025 reconciliation law, signed into law by President Trump on July 4, 2025, made major changes to federal revenues and spending, with CBO estimating the new law would reduce federal spending on Medicaid by $911 billion over the 2025-2034 period, relative to its January 2025 baseline projections of Medicaid spending under the law and regulations at the time.

CBO’s latest projections of Medicaid spending and enrollment from February 2026 show how enrollment and spending are expected to change over the next decade, accounting for the historic policy changes and their expected reductions in future federal Medicaid spending as well as other economic and technical changes. This policy watch compares CBO’s February 2026 projections of Medicaid spending and enrollment to earlier CBO projections. Projections of spending are compared to those from January 2025 (the baseline used to score the 2025 reconciliation law), but the most recent prior Medicaid enrollment projections are from June 2024. CBO’s newest projections show that enrollment is estimated to be 13% lower and spending 8% lower at the end of the budget windows, highlighting a significant shift in baselines stemming from Medicaid cuts in the 2025 reconciliation law. However, those changes understate the true effects of the 2025 reconciliation law because other factors caused Medicaid baseline spending to increase. 

CBO’s most recent Medicaid projections highlight the effects of the 2025 reconciliation law in reducing future Medicaid spending. The most recently released detailed CBO baseline shows that, following passage of the Medicaid changes in the reconciliation law, Medicaid spending is now expected to grow more slowly over time relative to earlier projections. As a result, 2035 spending is projected to be 8% lower than it was in the January 2025 baseline ($941 billion instead of $1.03 trillion, Figure 1). Over the entire 2025-2035 period, federal Medicaid spending in CBO’s latest baseline is projected to be $503 billion lower than estimated in the January 2025 baseline, before the passage of the 2025 reconciliation law. The reconciliation law included major changes to Medicaid eligibility, including the implementation of new Medicaid work requirements, and substantial changes to Medicaid financing, which together contribute to Medicaid’s lower baseline compared with prior years. CBO projects that federal Medicaid spending will still grow but more slowly because of the reductions in the reconciliation law. As a result, fewer people will be covered, and aggregate federal Medicaid spending will likely not keep pace with the increase in health care costs.

CBO’s Most Recent Medicaid Projections Highlight the Effects of the 2025 Reconciliation Law in Reducing Future Medicaid Spending (Line chart)

CBO’s Most Recent Medicaid Projections Highlight the Effects of the 2025 Reconciliation Law in Reducing Future Medicaid Spending

Federal Medicaid outlays

2025-2035 is the period for which comparable spending projections are available.

KFF analysis of Congressional Budget Office (CBO) Medicaid baseline projections

CBO’s latest spending projections also account for economic and technical changes that increased Medicaid spending relative to the January 2025 baseline, so comparing baselines may understate the effects of Medicaid cuts in the 2025 reconciliation law. One of the biggest reasons for increased spending was higher-than-expected per enrollee spending in 2025. CBO reports that costs per enrollee grew by 16% in that year, primarily because of declining health status after the COVID-19 continuous enrollment period ended. Those higher 2025 costs per enrollee compound over time due to inflation and rising health care costs. If the most recent baseline projections did not account for those 2025 cost increases, the differences between January 2025 and February 2026 Medicaid spending projections, driven by the Medicaid policy changes in the reconciliation law, would be larger.

CBO’s latest Medicaid projections also show the impact of the 2025 reconciliation law on reducing future Medicaid enrollment. The latest baseline shows that total average annual Medicaid enrollment is expected to decline, falling to 74 million enrollees by 2034 (a 13% reduction) compared with 85 million projected in the detailed baseline released before the new law’s passage. Many individuals who lose Medicaid coverage do not have another source of affordable health coverage and will become uninsured. CBO’s earlier estimates of the Medicaid policy changes in the reconciliation law found the new law will reduce Medicaid enrollment by more than 11 million and increase the number of people without insurance by 7.5 million in 2034, though these estimates do not account for recently released rules related to work requirements that could affect enrollment projections. CBO may release updated coverage estimates in the coming months. Data show that being uninsured has implications for access to care, financial stability, and health outcomes.

CBO's Latest Medicaid Projections Also Show the Impact of the 2025 Reconciliation Law on Reducing Future Medicaid Enrollment (Line chart)

CBO's Latest Medicaid Projections Also Show the Impact of the 2025 Reconciliation Law on Reducing Future Medicaid Enrollment

Total average annual Medicaid enrollment

CBO's January 2025 baseline did not include Medicaid enrollment estimates; enrollment projections from the June 2024 baseline are shown here. 2025-2034 is the period for which comparable enrollment projections are available.

KFF analysis of Congressional Budget Office (CBO) Medicaid baseline projections

Reductions in future Medicaid enrollment shown in CBO’s most recent projections are concentrated among ACA expansion adults, other adults, and children. The eligibility changes in the 2025 reconciliation law primarily affect adults in the ACA Medicaid expansion group (including new work requirements and more frequent eligibility determinations). Comparing CBO’s projections with those prior to passage of the reconciliation law shows the largest change in enrollment among the ACA Medicaid expansion group (5 million fewer expansion enrollees in 2034). The latest detailed baseline also shows 3 million fewer children and 2 million fewer other adult enrollees than the previous detailed baseline, likely due to provisions that affect groups beyond the expansion group and research showing that coverage loss among parents may reduce enrollment among children.

Reductions in Future Medicaid Enrollment Shown in CBO’s Most Recent Projections Are Concentrated Among ACA Expansion Adults, Other Adults, and Children (Grouped column chart)

Reductions in Future Medicaid Enrollment Shown in CBO’s Most Recent Projections Are Concentrated Among ACA Expansion Adults, Other Adults, and Children

Difference in 2034 average annual Medicaid enrollment by eligiblity group projected in the February 2026 CBO baseline compared with the June 2024 baseline

CBO's January 2025 baseline did not include Medicaid enrollment estimates; enrollment projections from the June 2024 baseline are use here. 2034 is the latest year for which comparable enrollment projections are available. Reductions in 2034 Medicaid enrollment by eligibility group do not sum to total 2034 Medicaid enrollment reductions due to rounding.

KFF analysis of Congressional Budget Office (CBO) Medicaid baseline projections

More On
Medicaid Congressional Budget Office (CBO) Enrollment Federal Budget Financing

Related Content

Health Provisions in the 2025 Federal Budget Reconciliation Law

Allocating CBO’s Estimates of Federal Medicaid Spending Reductions Across the States: Enacted Reconciliation Package

Tracking Implementation of the 2025 Reconciliation Law: Medicaid Work Requirements

Medicaid Financing: The Basics

Decoding Medicare Advantage Coding Intensity

Published: Jul 1, 2026

In recent years, federal payments to Medicare Advantage plans, and how they are adjusted for enrollee health status, have come under increased scrutiny. Medicare Advantage plans receive a capitated amount for each enrollee, and these payments are “risk adjusted” based on the diagnosis codes reported by the insurer to the Centers for Medicare & Medicaid Services (CMS) for each enrollee. Plans receive higher payments for enrollees who are sicker and expected to have higher health care spending, and lower payments for enrollees who are healthier and expected to have lower health care spending. The purpose of this risk adjustment is to ensure plans receive adequate payments to treat sicker, higher-cost patients and reduce incentives to enroll primarily healthier, lower cost, beneficiaries. However, since the approach to risk adjusting payments relies heavily on the diagnosis codes recorded for Medicare Advantage enrollees, it provides a strong financial incentive for private insurers to capture as many diagnosis codes for each enrollee as possible, which increases payments and contributes to higher Medicare spending.  

In contrast, payments under traditional Medicare only require the diagnosis codes necessary to support the services delivered. This means physicians and other health care providers do not have the same incentive to maximize the number of health care conditions documented through diagnosis codes. Differences in coding practices between traditional Medicare and Medicare Advantage (also referred to as coding intensity) mean that Medicare Advantage enrollees appear to be in worse health than they would if they received their Medicare benefits through traditional Medicare. Since the Medicare Advantage risk adjustment model is calibrated on traditional Medicare beneficiaries, the payments to Medicare Advantage plans are higher than necessary to cover expected costs, on average. According to the Medicare Payment Advisory Commission (MedPAC), in 2026, total payments to Medicare Advantage plans are $76 billion higher than traditional Medicare would spend for the same beneficiaries, of which $28 billion is attributed to coding intensity.

CMS has expressed a commitment to improving the accuracy of payments to Medicare Advantage and reducing the role coding practices play in determining the amount private plans receive from the federal government. Toward this end, the 2027 rate notice finalized a policy changing how certain diagnoses are considered when adjusting federal payments to Medicare Advantage plans for an enrollee’s health status. As policymakers and administration officials consider issues related to Medicare Advantage payments, this issue brief answers key questions about coding intensity, recent steps taken by CMS to address the impact of coding on payment, the effects on Medicare beneficiaries, and other proposals to improve Medicare Advantage payment accuracy.

What is coding?

Doctors and other health care providers include diagnosis codes on claims they submit to payers (either Medicare Administrative Contractors (MACs) for traditional Medicare or private insurers for Medicare Advantage) indicating a patient’s health conditions that support the health care services they delivered. The diagnosis codes for traditional Medicare beneficiaries are also used by CMS, along with other information, to develop a risk adjustment model estimating the relationship between a person’s health status (expressed as a “risk score”) and their projected health care spending. Medicare Advantage insurers submit the diagnosis codes documented by health care providers serving their enrollees to CMS for use in adjusting the payments the plans receive from the federal government using this risk adjustment model.

While the diagnosis codes used to develop the risk adjustment model only come from the claims providers submit for services rendered to traditional Medicare beneficiaries, those used to adjust payments to Medicare Advantage plans can be supplemented in two ways. First, Medicare Advantage plans may conduct health risk assessments (HRAs) and include the diagnosis codes for any conditions identified during this questionnaire in what is submitted to CMS – even when there are no related services delivered during the year to treat those conditions. KFF analysis finds that insurers often use rewards and incentives to encourage enrollees to complete HRAs. Second, Medicare Advantage plans may conduct chart reviews, which examine a person’s medical records, sometimes using AI tools, to determine if they are consistent with the information submitted by health care providers to the insurer. KFF analysis finds that chart reviews are used to add diagnosis codes that do not otherwise appear on a record for an encounter with a physician, increasing payments from CMS to Medicare Advantage insurers for one in six Medicare Advantage enrollees.

What is coding intensity?

Coding intensity is the degree to which a person’s health care conditions are documented through diagnosis codes. Differences in coding patterns across groups of beneficiaries, such as Medicare Advantage enrollees and traditional Medicare beneficiaries, or those in Medicare Advantage plans sponsored by different insurers, are described as differences in coding intensity. Higher coding intensity is not necessarily fraudulent, but fraud can contribute to higher coding intensity.

Because Medicare Advantage payments are generally higher for enrollees with more diagnosis codes (and therefore higher risk scores), private insurers have an incentive to document more health conditions, but there is no similar incentive in traditional Medicare. Recognizing this incentive, lawmakers have required CMS to reduce Medicare Advantage risk scores by at least 5.9% across the board before adjusting payments to private plans. However, that adjustment does not fully account for difference in coding patterns, and a number of studies have documented that risk scores are still higher in Medicare Advantage after applying the coding intensity adjustment than they would be if enrollees received their Medicare benefits under traditional Medicare. The magnitude of the uncorrected coding intensity after the adjustment has varied over time, ranging from as low as 2% in 2016 to 10% in 2023, and is estimated to be approximately 4% in 2026, according to MedPAC. The magnitude also varies by insurer and is larger for insurers comprising a larger share of enrollment. See Box 1 for an illustrative example of higher coding intensity and the effect on Medicare Advantage payments.

Box 1. Illustrative Example of How Higher Coding Intensity in Medicare Advantage Increases Payments to Private Insurers.

The risk adjustment model, which is used to assign a risk score to all Medicare Advantage enrollees, specifies “coefficients” for each factor that contributes to a person’s risk score. Each coefficient reflects the average marginal impact, or how much higher traditional Medicare spending is expected to be, due to that factor. To illustrate how coding intensity increases Medicare Advantage payments, consider Mr. Smith, who is 73-years old, living in the community, and received health care services to treat type 2 diabetes and heart failure last year. He is enrolled in a Medicare Advantage plan that receives $12,000 per year for an average Medicare beneficiary (risk score = 1).

If Mr. Smith’s Medicare Advantage plan codes consistent with traditional Medicare, the coefficients from the 2026 Risk Adjustment Model for each of the factors contributing to Mr. Smith’s risk score would be: Male 70-74 years – 0.396, Diabetes with Chronic Complications (HCC37) – 0.166, Heart Failure (HC226) – 0.336, and an interaction for having both diabetes and heart failure – 0.112. Mr. Smith’s risk score would be equal to the sum of these coefficients, 1.034, or 0.912 after applying the 1.067 normalization factor for the 2026 plan payment year (which is used to ensure the average risk score is equal to 1 in years beyond the initial estimation year) and the 5.9% coding adjustment (which applies to all plans regardless of whether they code consistent with traditional Medicare or have higher coding intensity). The plan would receive payments totaling $10,943 for Mr. Smith if he is enrolled the entire year ($12,000 * 0.912).

If instead the Medicare Advantage plan has higher coding intensity, it is possible that an additional diagnosis (or diagnoses) could be added to Mr. Smith’s record. For example, if the plan does a chart review and uncovers that Mr. Smith also meets the definition for morbid obesity, a condition that is documented more often in Medicare Advantage than traditional Medicare, his unadjusted risk score would increase by 0.186, bringing it up to 1.220. After applying the 1.067 normalization factor and 5.9% coding adjustment, his risk score would be 1.076. The plan would receive payments totaling $12,911 if Mr. Smith is enrolled the entire year ($12,000 * 1.076).

As a result of higher coding intensity, the plan receives nearly $2,000 more over the year for Mr. Smith – 18% more – than if it coded consistent with traditional Medicare (Figure 1).

Illustrative Example of Impact of Higher Coding Intensity on Total Medicare Advantage Payments for the Year (Stacked column chart)

Illustrative Example of Impact of Higher Coding Intensity on Total Medicare Advantage Payments for the Year

Total payments are $1,968 higher in the example of higher coding intensity compared with coding behavior consistent with traditional Medicare

Illustrative payment for 73-year old male who has health care provider encounter records with diagnosis codes for heart failure and diabetes. In the example, higher coding intensity includes the addition of a morbid obesity diagnosis added as part of a chart review. Applies coefficients from V28 risk adjustment model to a plan that receives a payment of $1,000 per month ($12,000 per year) for an average Medicare beneficiary (risk score = 1, before across-the-board coding adjustment). Reflects the 1.067 normalization factor and impact of the across-the-board coding adjustment of 5.9%.

What has CMS done to reduce coding intensity?

Risk Model Revisions. CMS periodically revises the risk adjustment model. Most recently, CMS updated the data used to calibrate the model and changed how certain conditions that were coded more frequently in Medicare Advantage than traditional Medicare were incorporated (or not). The move to the new model (referred to as V28 because it is the 28th version of the model) was phased in between 2024 and 2026.

Following full implementation of the new risk adjustment model, MedPAC estimated that the impact of coding intensity on Medicare Advantage payments has declined from increasing payments by 10% in 2022 to 4% in 2026 (the first year the V28 model is fully in effect). A recent analysis from CMS staff approached the analysis from a different angle – examining what the impact of using V28 would have been in 2022 if it had been in effect. That analysis finds uncorrected coding intensity (after applying the 5.9% adjustment) would have been between 1.5% and 2.0%, compared to 10% under the previous risk adjustment model (V24) that was in effect in 2022. This is consistent with the findings of other researchers, but the estimate for 2022 is not directly comparable to the MedPAC analysis because the CMS analysis modeled the impact of V28 in an earlier year in which it was not in effect, while MedPAC looks at the uncorrected coding intensity using the risk model in effect in the current payment year (2026).

Analyses of Medicare Advantage risk score trends have consistently found that coding intensity grows over time. Thus, the CMS staff estimate that uncorrected coding intensity would have been between 1.5% and 2.0% if the V28 model had been fully implemented in 2022 is consistent with MedPAC’s higher estimate of uncorrected coding intensity in 2026 of 4%, which incorporates growth in coding intensity between 2022 and 2026.

Coding Intensity. CMS also routinely makes other changes to the risk adjustment process separate from moving to a new model. For example, in the 2027 rate notice, CMS finalized a policy to exclude diagnosis codes added for enrollees based on chart review records that are not linked to an encounter with a health care provider (referred to as “unlinked” chart reviews). CMS estimates the new policy will reduce average payments to Medicare Advantage plans by 1.5% compared to what they would have been otherwise. While that estimate is similar in magnitude to the CMS staff estimate of uncorrected coding intensity in 2022 if the V28 model had been in effect, the two are not directly comparable because the impact of removing unlinked chart reviews applies to the 2027 plan year payment.

The use of chart reviews has come under scrutiny because analysis of Medicare Advantage insurers’ coding practices consistently finds that chart reviews are the primary contributor to higher coding intensity in Medicare Advantage. However, chart reviews are likely to continue to contribute to higher coding intensity in Medicare Advantage even after excluding diagnoses from unlinked chart reviews. Based on KFF analysis of Medicare Advantage encounter data for 2022, diagnoses from unlinked chart reviews comprised one-third of all diagnoses added through the chart review process in 2022, meaning that diagnoses from chart reviews that were linked to an encounter account for most of the diagnosis codes added during the chart review process.

Additionally, it is likely that an even smaller share of all diagnoses added on chart reviews will be impacted by the new policy to exclude diagnosis codes from unlinked chart reviews because the condition categories for which specific diagnoses were most commonly added in an unlinked chart review in 2022 were substantially impacted by the move to the V28 risk adjustment model. For example, vascular disease was among the most common conditions added on an unlinked chart review in 2022 that increased payment, but this condition category was substantially narrowed as part of the shift to V28 – meaning that many of these diagnoses codes would no longer count towards payment under the risk adjustment model regardless of the policy change related to unlinked chart reviews. Finally, the impact could be less than CMS estimates if insurers put more effort into linking chart reviews to encounters so that included diagnoses can be considered for risk adjustment purposes.

How are Medicare beneficiaries impacted by coding intensity and changes to the risk adjustment model?

Since higher risk scores increase Medicare payments to plans, higher coding intensity provides plans with the option to offer more extra benefits to enrollees, such as dental, vision, and hearing coverage, as well as reduced cost sharing. In addition, this additional funding from the federal government can be used for other purposes, such as increasing plan margins or paying for more advertising, as long as the Medicare Advantage insurer meets the minimum medical loss ratio required under law. (The medical loss ratio is the share of premium revenues going to pay for claims versus administrative overhead and profit.) 

Industry representatives have raised concerns that payment changes, including efforts to address coding intensity, could result in plans offering fewer extra benefits or raising costs for Medicare Advantage enrollees. While there have been some changes to plan benefits and costs following the implementation of V28, including modest increases in out-of-pocket limits and decreases in some extra benefits, such as the availability of funds provided to pay for over-the-counter drugs and supplies, private insurers have generally absorbed a large portion of payment changes. An analysis of the first two years of the phase in of the new risk score model finds that insurers reduced benefits or raised costs by between 17% and 24% of the anticipated reduction in plan payments. That is a smaller effect than previous analysis of the impact of changes to Medicare Advantage payments, which found that private insurers passed through about half of the payment change in the form of fewer benefits and/or higher costs.

What additional steps can be taken to improve the accuracy of Medicare Advantage payments?

Policy proposals to address the remaining uncorrected coding intensity in Medicare Advantage include expanding the sources of diagnoses that are ineligible for risk adjustment to include all chart reviews and HRAs, increasing the 5.9% across-the-board adjustment to risk scores, or applying a tiered adjustment to risk scores based on historical coding intensity (so plans with higher coding intensity in previous years would have larger adjustments to their risk scores). Additionally, updates to the risk score model, such as those proposed but not finalized by CMS for 2027, may also better align the adjustments for health status to the expected impact on spending by incorporating more recent data to better reflect current treatment patterns and costs.

Beyond coding intensity, favorable selection into Medicare Advantage also increases payments above what traditional Medicare would spend for the same beneficiaries. Favorable selection occurs when the people who enroll in Medicare Advantage have lower actual health care use and spending, on average, than what is predicted by the risk score model. For example, previous KFF analysis found that Medicare beneficiaries who enroll in Medicare Advantage have lower spending than those who remain in traditional Medicare, after adjusting for health risk using the risk adjustment model. MedPAC estimates that the largest component of higher payments to Medicare Advantage plans relative to traditional Medicare is favorable selection into Medicare Advantage. The impact of favorable selection has been relatively stable over time, ranging between 9% and 11%, according to MedPAC; in other words, resulting in payments to Medicare Advantage plans for enrollees that are 9% to 11% higher than costs would be in traditional Medicare. In 2026, favorable selection is estimated to add $57 billion to Medicare spending.

To address the impact of favorable selection on Medicare Advantage payments, policymakers could make changes to the maximum amount the federal government is willing to pay Medicare Advantage plans, also known as benchmarks. One approach is an across-the-board reduction in benchmarks, such as the “discount rate” proposed by MedPAC. This would account for Medicare Advantage enrollees having lower expected health care spending, before any effects of Medicare Advantage plan design, than traditional Medicare beneficiaries with similar risk profiles, on which current benchmarks are based. Other options include expanding the sources of data used in the risk adjustment model to predict a Medicare Advantage enrollee’s costs, such as prescription drug claims, clinical data from electronic health records, or certain measures currently included in the Consumer Assessment of Healthcare Providers and Systems (CAHPS) (see for example, a recent proof-of-concept study). These measures could be applied on equal footing between Medicare Advantage and traditional Medicare, and among different Medicare Advantage plans. In addition, two-sided reinsurance, which would provide additional payments to insurers with enrollees who have extremely high and unexpected costs and require insurers with enrollees with substantially lower than predicted spending to make payments into the program, could be added to the Medicare Advantage payment system. That would reduce the financial rewards for attracting enrollees who use substantially fewer health care services and protect insurers against the financial costs of enrollees who use substantially more health care services.

This work was supported in part by Arnold Ventures. KFF maintains full editorial control over all of its policy analysis, polling, and journalism activities.

More On
Medicare Federal Budget Health Status Medicare Advantage

Related Content

How Medicare Pays Medicare Advantage Plans: Issues and Policy Options

Chart Reviews Increase Payments to Medicare Advantage Insurers for 1 in 6 Enrollees

Medicare Will Spend More Than $13 Billion on the Medicare Advantage Quality Bonus Program in 2026

Recent Research on How Experiencing Racial Discrimination Impacts Health

Published: Jun 30, 2026
Section:
0 / 0

Introduction

Copy link to Introduction

Racial discrimination is an underlying driver of health disparities that affects experiences across many aspects of everyday life as well as in health care settings. Repeated and ongoing exposure to racial discrimination can negatively affect individuals’ health and well-being, increasing risks of poor outcomes across multiple domains. Understanding how exposure to racial discrimination affects health can inform efforts to reduce health disparities. Racial and ethnic health and health care disparities result in higher rates of illness and death across a wide range of health conditions and are costly to the health care system, resulting in excess medical care costs and lost productivity, as well as additional economic losses due to premature deaths each year. Amid current federal efforts to reduce resources and initiatives focused on addressing disparities, identifying and understanding the continued evidence base about the role of racial discrimination in contributing to negative health outcomes remains important.

This brief provides an overview of the relationship between racial discrimination and health and highlights research published since 2015 examining mechanisms underlying health outcomes linked to self-reported experiences of racial discrimination, including biological changes, chronic stress, mental health, substance use, pregnancy-related outcomes, and sleep. Other research has also identified how structural racism negatively impacts health but is beyond the scope of this brief.

A large body of research conducted over several decades prior to 2015 has linked experiences of racial discrimination to negative health outcomes. While not exhaustive, this brief builds on past analyses by capturing more recent literature, including large-scale longitudinal studies and those based on methodological advances that study genetic, protein, and brain imaging biomarkers to better understand biological changes linking experiences of racial discrimination to health outcomes. Criteria for inclusion included studies conducted among U.S. populations that used validated measures of racial discrimination experiences and examined associations with biomarkers or health outcomes by race. Key takeaways include the following:

  • Recent research builds upon earlier evidence that racial discrimination is associated with worse health across multiple domains, which may contribute to health disparities. Research linked self-reported experiences of racial discrimination to a greater risk of chronic disease, mental health disorders, substance use, adverse pregnancy outcomes, and sleep problems among people of color compared to White people. For example, Black and Hispanic people who reported experiencing racial discrimination had elevated risk for cardiovascular disease compared to those who did not report discrimination. Exposure to racial discrimination also is associated with higher rates of preterm births and low birth weight among infants born to Black women compared to White women.
  • Emerging studies suggest racial discrimination may impact health outcomes through biological mechanisms linked to stress, inflammation, and changes in the brain. Recent findings associated experiences of racial discrimination with elevated stress and inflammation, shortened telomere length, and changes in brain structure and activity that can increase the risk of chronic disease, poor mental health outcomes, and shortened lifespans.
  • Some research gaps and limitations remain. Most studies relied on self-reported experiences of discrimination, which researchers identified as challenging to measure. Moreover, most research focused primarily on Black populations and therefore gaps remain in understanding impacts for other groups who experience ongoing discrimination. Some studies controlled for a more robust set of potential confounding factors, such as age, gender, income, and education, than others, largely due to limitations in sample size. Additionally, studies had mixed findings on the protective social and coping factors that mitigate the negative effects of racial discrimination.

Future research into how experiences of racial discrimination impact health may be limited due to actions by Trump administration, including executive orders eliminating federal diversity, equity, inclusion, and accessibility (DEIA) programs and related initiatives. A major impact of these efforts has been a reduction in federal support for health disparities research, which may limit the information available to track disparities and better understand the underlying factors affecting health outcomes.

Background

Copy link to Background

Racism and discrimination at all levels contribute to differences in experiences across many aspects of everyday life which can negatively impact people’s health and well-being. It contributes to underlying inequities in social and economic factors that reflect historical and contemporary policies and drive racial and ethnic disparities in health, including access to housing, food, and economic and educational opportunities. However, racial health disparities persist even when controlling for differences in socioeconomic status. Many people of color continue to report experiences with daily discrimination. KFF survey data from 2023 found that at least half of American Indian or Alaska Native (AIAN) (58%), Black (54%), Hispanic adults (50%), and about 4 in 10 Asian adults (42%) say they experienced at least one type of interpersonal discrimination in daily life in the past year. These experiences included receiving poorer service than others at restaurants or stores; people acting as if they are afraid of them or as if they aren’t smart; being threatened or harassed; or being criticized for speaking a language other than English. 

A large body of research conducted over several decades has consistently documented strong associations between self-reported experiences of racial discrimination and negative health outcomes. These outcomes include poor mental health, such as depression, anxiety, and psychological distress. In addition, numerous studies have linked racial discrimination to physical health outcomes, such as hypertension, cardiovascular disease, obesity, asthma, and breast cancer, underscoring its broad impact on both physical and psychological well-being. Two conceptual frameworks help explain the mechanisms through which these associations may arise. The weathering hypothesis describes how chronic exposure to social and economic adversity, including racism and socioeconomic disadvantages, can accelerate health deterioration and contribute to racial health disparities. Allostatic load theory similarly focuses on cumulative physiologic “wear and tear” from repeated or chronic stress activation, which is associated with poorer health outcomes. Allostatic load is typically measured using various indicators, including blood pressure, cardiometabolic indicators, and other biomarkers. Research has found elevated allostatic load among adults who experienced various types of discrimination, including childhood racial discrimination.

Recent Research on Racial Discrimination and Health

Copy link to Recent Research on Racial Discrimination and Health

Building on existing research, studies since 2015 have sought to replicate previous studies to confirm and extend findings by applying existing theories to more subgroups and outcomes and using new tools. The research often relies on associating health outcomes and biological indicators with self-reported experiences of racial discrimination, most commonly captured using validated survey instruments which measure individuals’ exposure to racial discrimination, such as the Experiences of Discrimination Scale. As this is an evolving body of research, several research limitations exist. Nearly all research studies utilize self-reported experiences of racial discrimination, which may underestimate actual exposure due to social desirability bias, recall errors, or confounds with other intersectional social factors. Some survey instruments that assess perception of racial discrimination may not perform equivalently across different racial and socioeconomic groups, which could affect cross-group comparisons. Most studies focus on Black people, while studies including populations such as AIAN and Native Hawaiian and Pacific Islander (NHPI) people are limited, reducing the generalizability of findings to these and other groups, such as Hispanic and Asian people, who experience ongoing racism. While most studies also analyzed interactions and effects due to variables other than exposure to racial discrimination, such as age, gender, income, and education, some used a less robust set of potential confounding factors due to limitations in sample size.

Impacts on Chronic Disease, Biological Changes, and Stress

Multiple studies find that racial discrimination is associated with a higher risk of chronic diseases and other conditions that may increase mortality risk. Research suggests that racial discrimination is associated with higher risk for cardiovascular and metabolic diseases, including high blood pressure, obesity, diabetes, chronic kidney disease, and other health conditions. A study found that people who reported experiencing racial discrimination had a 5% elevated risk for cardiovascular disease compared to those who did not report discrimination, with the strongest association between racial discrimination and cardiovascular disease risk seen among Asian and Latino people and among women compared to men. Additionally, exposure to experiences of racial discrimination during childhood among Black adults was associated with poorer cardiovascular health outcomes in adulthood compared to those who did not report discrimination. Other studies have highlighted a connection between racial discrimination and higher rates of obesity among women, as well as other health outcomes, including lupus and organ damage among Black women. One study found that experiencing discrimination was associated with an increase in mortality risk due to cardiovascular disease among Black people regardless of health behaviors, clinical risk factors, or social factors such as gender or racial and ethnic residential segregation. Another study found that experiencing racial discrimination was associated with higher risk of mortality due to any cause among Black adults ages 50 and older, even when controlling for health, behavioral, and economic factors.

Experiencing racial discrimination is associated with biological changes that increase inflammation and stress, which may increase the risk of developing chronic conditions and shorten lifespan. Inflammation is a natural response to injury and illness, but chronic inflammation that occurs in the absence of injury or illness can lead to various health issues, including cardiovascular disease, diabetes, and immune function. A longitudinal study found that elevated inflammation and higher levels of cumulative lifespan stress, which included experiences of discrimination, partly accounted for the shorter lifespans seen among Black participants compared to those who were White. Recent studies showed that individuals who experienced discrimination, including racial discrimination, exhibited higher levels of inflammation biomarkers compared to those who did not report discrimination. For example, one study found that Black people experienced more stress than White people across various measures, including due to racial discrimination, and that stress exposure was strongly associated with higher levels of a protein associated with inflammation. Research also found a similar pattern among pregnant Black women who reported experiencing racial discrimination compared to those who had not. Further, research at the genetic level, including the mechanisms that control inflammation levels, found that Black study participants had higher inflammatory signaling than White participants, and that racial discrimination explained over half of the race-related differences in expression of genes that promote inflammation. Another study of Black and White adults found that, among participants reporting high perceived discrimination, Black adults had a higher expression of a different set of genes linked to immune function and inflammation compared with White participants, suggesting a unique gene expression linked to experiences of racial discrimination.

Racial discrimination is also associated with shortened telomere length, which contributes to accelerated biological aging. Telomeres, which protect the ends of chromosomes, naturally shorten over time and serve as an indicator of aging. Chronic stress can accelerate telomere shortening and is associated with earlier onset of age-related disease, such as heart disease and cancer. Research found that, among Black people, those who reported racial discrimination had faster telomere shortening over a ten-year period than those who did not, though a separate study observed this effect only among Black adults ages 50 and older due to everyday discrimination rather than racial discrimination specifically. Another study found that experiences of racial discrimination were associated with shorter telomere length among Black women, Black people with high socioeconomic status, Black adults under age 40, and White men under age 40, illustrating complex interactions between experiences of discrimination and other sociodemographic factors. One hypothesis for the association observed among younger White men is that younger White males feeling “targeted” due to perceptions of race-related disadvantages as a result of increases in diversity efforts. Such perceptions may be especially pronounced among younger adults navigating educational and career advancement, although additional research is needed to better understand these relationships.

Research also has identified mitigating social and coping factors that may limit the impact of racial discrimination on stress and allostatic load, but some findings on protective buffers are mixed. Research among Black youth ages 16–18 found that higher parental and peer emotional support was associated with lower allostatic loads. A study found that, among Black women who reported experiences of racial discrimination, those with higher education levels and lower poverty status had lower allostatic load regardless of how much racial discrimination they reported experiencing compared to those with lower education levels. However, a study comparing allostatic load between Black and White adults found that experiencing higher levels of racial discrimination was associated with higher allostatic load regardless of education level, income, or wealth. Another study found that “John Henryism” among Black people, a high-effort, active coping style in response to racism and sociodemographic challenges, was associated with fewer depressive episodes but higher allostatic load, suggesting that some social coping strategies may come at the expense of health. Similarly, research conducted on the “superwoman schema” among Black women, where resilience, self-reliance, and other social processes are central to coping with discrimination, found mixed results where some coping strategies limit stress while others exacerbate it.

Pregnancy and Birth Outcomes

Recent research expands on a large body of research that shows racial discrimination is associated with adverse pregnancy outcomes. Research has documented that racism and chronic stress contribute to poor maternal and infant health outcomes, including higher rates of pregnancy-related depression and preterm birth among Black women and higher rates of mortality among Black infants. Racial discrimination during pregnancy may contribute to disparities in maternal and infant health outcomes as research has found that women of color experienced greater stress from experiencing racial discrimination than White women. Pregnancy-specific stress and lifetime exposure to racial discrimination disproportionately affect Black women and other women of color, and racial discrimination is associated with increased risk of psychological distress and reduced social support during pregnancy. Exposure to racial discrimination is associated with higher rates of preterm births, small for gestational age, and low weight births among infants born to Black women compared to White women. Additionally, experiencing racial discrimination during pregnancy is associated with elevated stress-related inflammatory markers and poor sleep among Black women, which may negatively affect maternal and perinatal health outcomes. Research among Black women also links maternal experiences of racial discrimination to poor sleep health among their children.

Mental Health and Substance Use

Recent evidence builds upon prior literature linking racism and discrimination to negative mental health outcomes, including post-traumatic stress disorder (PTSD), depression, and anxiety. Racial discrimination is associated with depression, anxiety, post-traumatic stress symptoms, and suicidal ideation and attempts among Black people. Recent studies indicated that exposure to discrimination exacerbated PTSD symptoms following traumatic injuries, as the added stress of racial bias may compound the psychological impact of the original trauma. Additionally, research among Black adults found that higher experiences of racial discrimination were associated with non-remitting PTSD, or PTSD that does not improve over time. Among Black youth between ages 9–14, higher levels of racial discrimination were associated with a greater risk of developing depressive symptoms over time. A study of school-age youth between 6th and 12th grade found that experiences of racial discrimination were associated with serious psychological distress and suicidality, most prominently among Black, Asian, and multiracial students. Research among Black youth between ages 11–19 found associations between online racial discrimination, PTSD symptoms, and suicidal ideation.

Emerging research also suggests that racial discrimination may contribute to changes to brain structure and activity that may increase risk of brain disorders and poor mental health outcomes. Brain structure plays a role in determining cognitive function and emotional regulation, with certain changes in brain volume, white matter integrity, and connectivity between different parts of the brain potentially increasing vulnerability to brain disorders and mental health conditions such as PTSD, depression, and anxiety. Research utilizing brain imaging methods has shown that experiencing racial discrimination is linked to reduced white matter integrity among Black adults 55 and older, which may increase the risk of stroke, dementia, and cognitive decline, and lower overall brain volume, which may be associated with depression. Among Black women, experiencing racial discrimination was associated with further reduced white matter integrity even when accounting for changes associated with trauma and PTSD. Research among Black youth found that coping with racial discrimination was associated with changes in brain activity that increased anxiety, depression, aggression, and rule-breaking symptoms. Research among trauma-exposed Black women also found that experiencing racial discrimination was associated with changes in connectivity between certain brain regions, including heightened activation in brain regions associated with threat vigilance and response, a state of chronic heightened stress. Another study among those who experienced a traumatic brain injury found that exposure to racial discrimination was associated with heightened connections in brain areas responsible for threat arousal, which is a state of heightened alertness that typically activates a stress response due to danger.

Experiencing racial discrimination is associated with an increased risk of substance use and alcohol use disorders. A review of studies found significant links between experiences of racial discrimination and both substance use and negative mental health outcomes. Experiences of racial discrimination were found to be associated with an increased risk for alcohol use disorder among AIAN, Black, Hispanic, and NHPI adults. Research also found that, among Hispanic college students, racial discrimination was a significant risk factor for the development of maladaptive alcohol use. A study of Black adults ages 18–24 found that experiences of racial discrimination were associated with past-year drug use and with frequent drug use, with a stronger association seen among those with a higher socioeconomic status compared to adults with lower socioeconomic status.

Sleep Disruption

Data suggest that racial discrimination is associated with sleep disruption, which may contribute to a range of negative health outcomes. Poor sleep is associated with a range of negative health outcomes, including increased inflammation, heightened risk for diabetes and obesity, and mental health issues such as depression and anxiety. Among youth ages 13–15, experiences of racial discrimination were linked to shorter sleep duration, more frequent disturbances, increased depressive symptoms, and lower levels of self-esteem. Additionally, research among college students found that experiences of racial discrimination contributed to greater increases in sleep problems among Black students compared to White students. Other research among people diagnosed with insomnia disorder found that experiences of racial discrimination were a significant factor in the link between race and insomnia severity for Black, Asian, and multiracial individuals.

References

Copy link to References

Agbonlahor, O., DeJarnett, N., Hart, J. L., Bhatnagar, A., McLeish, A. C., & Walker, K. L. (2024). Racial/Ethnic discrimination and cardiometabolic diseases: A systematic review. Journal of Racial and Ethnic Health Disparities, 11(2), 783–807. https://doi.org/10.1007/s40615-023-01561-1

Allen, A. M., Thomas, M. D., Michaels, E. K., Reeves, A. N., Okoye, U., Price, M. M., Hasson, R. E., Syme, S. L., & Chae, D. H. (2019). Racial discrimination, educational attainment, and biological dysregulation among midlife African American women. Psychoneuroendocrinology, 99, 225–235. https://doi.org/10.1016/j.psyneuen.2018.09.001

Allen, A. M., Wang, Y., Chae, D. H., Price, M. M., Powell, W., Steed, T. C., Rose Black, A., Dhabhar, F. S., Marquez‐Magaña, L., & Woods‐Giscombe, C. L. (2019). Racial discrimination, the superwoman schema, and allostatic load: Exploring an integrative stress‐coping model among African American women. Annals of the New York Academy of Sciences, 1457(1), 104–127. https://doi.org/10.1111/nyas.14188

Artiga, S., Hamel, L., Gonzalez-Barrera, A., Montero, A., Hill, L., Presiado, M., Kirzinger, A., & Lopes, L. (2023, December 5). Survey on Racism, Discrimination and Health. KFF. https://www.kff.org/racial-equity-and-health-policy/poll-finding/survey-on-racism-discrimination-and-health/

Bastos, J. L., & Harnois, C. E. (2020). Does the Everyday Discrimination Scale generate meaningful cross-group estimates? A psychometric evaluation. Social Science & Medicine, 265(113321). https://doi.org/10.1016/j.socscimed.2020.113321

Beldon, M. A., Clay, S. L., Uhr, S. D., Woolfolk, C. L., & Canton, I. J. (2024). Exposure to racism and adverse pregnancy outcomes for Black women: A systematic review and meta-analysis. Journal of Immigrant and Minority Health, 27. https://doi.org/10.1007/s10903-024-01641-2

Berger, M., & Sarnyai, Z. (2014). “More than skin deep”: Stress neurobiology and mental health consequences of racial discrimination. Stress, 18(1), 1–10. https://doi.org/10.3109/10253890.2014.989204

Bernardo, C. de O., Bastos, J. L., González-Chica, D. A., Peres, M. A., & Paradies, Y. C. (2017). Interpersonal discrimination and markers of adiposity in longitudinal studies: A systematic review. Obesity Reviews, 18(9), 1040–1049. https://doi.org/10.1111/obr.12564

Bird, C. M., Webb, E. K., Schramm, A. T., Torres, L., Larson, C., & deRoon ‐Cassini, T. A. (2021). Racial Discrimination is associated with acute posttraumatic stress symptoms and predicts future posttraumatic stress disorder symptom severity in trauma‐exposed Black adults in the United States. Journal of Traumatic Stress, 34(5). https://doi.org/10.1002/jts.22670

Boen, C. (2019). Death by a thousand cuts: Stress exposure and Black–White disparities in physiological functioning in late life. The Journals of Gerontology: Series B, 75(9), 1937–1950. https://doi.org/10.1093/geronb/gbz068

Boyd, D. T., Quinn, C. R., Durkee, M. I., Williams, E.-D. G., Constant, A., Washington, D., Butler-Barnes, S. T., & Ewing, A. P. (2024). Perceived discrimination, mental health help-seeking attitudes, and suicide ideation, planning, and attempts among black young adults. BMC Public Health, 24(1). https://doi.org/10.1186/s12889-024-19519-1

Brody, G. H., Lei, M.-K., Chae, D. H., Yu, T., Kogan, S. M., & Beach, S. R. H. (2014). Perceived discrimination among African American adolescents and allostatic load: A longitudinal analysis with buffering effects. Child Development, 85(3), 989–1002. https://doi.org/10.1111/cdev.12213

Cahill, M., Illback, R., & Peiper, N. (2024). Perceived racial discrimination, psychological distress, and suicidal behavior in adolescence: Secondary analysis of cross-sectional data from a statewide youth survey. Healthcare, 12(10), 1011. https://doi.org/10.3390/healthcare12101011

Carliner, H., Delker, E., Fink, D. S., Keyes, K. M., & Hasin, D. S. (2016). Racial discrimination, socioeconomic position, and illicit drug use among US Blacks. Social Psychiatry and Psychiatric Epidemiology, 51(4), 551–560. https://doi.org/10.1007/s00127-016-1174-y

Cave, L., Cooper, M. N., Zubrick, S. R., & Shepherd, C. C. J. (2020). Racial discrimination and child and adolescent health in longitudinal studies: a systematic review. Social Science & Medicine, 250(112864), 112864. https://doi.org/10.1016/j.socscimed.2020.112864

Chae, D. H., Martz, C. D., Fuller-Rowell, T. E., Spears, E. C., Smith, T. T. G., Hunter, E. A., Drenkard, C., & Lim, S. S. (2019). Racial discrimination, disease activity, and organ damage: The Black women’s experiences living with lupus (BeWELL) study. American Journal of Epidemiology, 188(8). https://doi.org/10.1093/aje/kwz105

Chae, D. H., Wang, Y., Martz, C. D., Slopen, N., Yip, T., Adler, N. E., Fuller-Rowell, T. E., Lin, J., Matthews, K. A., Brody, G. H., Spears, E. C., Puterman, E., & Epel, E. S. (2020). Racial discrimination and telomere shortening among African Americans: The coronary artery risk development in young adults (CARDIA) study. Health Psychology, 39(3), 209–219. https://doi.org/10.1037/hea0000832

Chen, S., & Mallory, A. B. (2021). The effect of racial discrimination on mental and physical health: A propensity score weighting approach. Social Science & Medicine, 285(114308). https://doi.org/10.1016/j.socscimed.2021.114308

Cheng, H.-L., & Mallinckrodt, B. (2015). Racial/ethnic discrimination, posttraumatic stress symptoms, and alcohol problems in a longitudinal study of Hispanic/Latino college students. Journal of Counseling Psychology, 62(1), 38–49. https://doi.org/10.1037/cou0000052

Cheng, P., Cuellar, R., Johnson, D. A., Kalmbach, D. A., Joseph, C. L., Cuamatzi Castelan, A., Sagong, C., Casement, M. D., & Drake, C. L. (2020). Racial discrimination as a mediator of racial disparities in insomnia disorder. Sleep Health, 6(5), 543–549. https://doi.org/10.1016/j.sleh.2020.07.007

Cleveland Clinic. (2024, March 22). Inflammation: What is it, causes, symptoms & treatment. Cleveland Clinic. https://my.clevelandclinic.org/health/symptoms/21660-inflammation

Cobb, R. J., Sheehan, C., Louie, P., & Erving, C. L. (2021). Multiple reasons for perceived everyday discrimination and all-cause mortality risk among older Black adults. The Journals of Gerontology, 77(2), 310–314. https://doi.org/10.1093/gerona/glab281

Cohen, M. F., Corwin, E. J., Johnson, D. A., Amore, A. D., Brown, A. L., Barbee, N. R., Brennan, P. A., & Dunlop, A. L. (2022). Discrimination is associated with poor sleep quality in pregnant Black American women. Sleep Medicine, 100, 39–48. https://doi.org/10.1016/j.sleep.2022.07.015

Cohen, M. F., Dunlop, A. L., Johnson, D. A., Dunn Amore, A., Corwin, E. J., & Brennan, P. A. (2022). Intergenerational effects of discrimination on Black American children’s sleep health. International Journal of Environmental Research and Public Health, 19(7), 4021. https://doi.org/10.3390/ijerph19074021

Colten, H. R., Altevogt, B. M., & Institute of Medicine (US) Committee on Sleep Medicine and Research. (2006). Extent and health consequences of chronic sleep loss and sleep disorders. NIH.gov; National Academies Press (US). https://www.ncbi.nlm.nih.gov/books/NBK19961/

Cuevas, A. G., Ho, T., Rodgers, J., DeNufrio, D., Alley, L., Allen, J., & Williams, D. R. (2021). Developmental timing of initial racial discrimination exposure is associated with cardiovascular health conditions in adulthood. Ethnicity & Health, 26(7), 949–962. https://doi.org/10.1080/13557858.2019.1613517

Cuevas, A. G., McSorley, A.-M., Adiammi Lyngdoh, Fatoumata Kaba-Diakité, Harris, A., Brennan Rhodes-Bratton, & Rouhani, S. (2024). Education, income, wealth, and discrimination in Black-White allostatic load disparities. American Journal of Preventive Medicine, 67(1), 97–104. https://doi.org/10.1016/j.amepre.2024.02.021

Cuevas, A. G., Ong, A. D., Carvalho, K., Ho, T., Chan, S. W. (Celine), Allen, J. D., Chen, R., Rodgers, J., Biba, U., & Williams, D. R. (2020). Discrimination and systemic inflammation: A critical review and synthesis. Brain, Behavior, and Immunity, 89, 465–479. https://doi.org/10.1016/j.bbi.2020.07.017

Fani, N., Carter, S. E., Harnett, N. G., Ressler, K. J., & Bradley, B. (2021). Association of racial discrimination with neural response to threat in Black women in the US exposed to trauma. JAMA Psychiatry, 78(9). https://doi.org/10.1001/jamapsychiatry.2021.1480

Fani, N., Harnett, N. G., Bradley, B., Mekawi, Y., Powers, A., Stevens, J. S., Ressler, K. J., & Carter, S. (2021). Racial discrimination and white matter microstructure in trauma-exposed Black women. Biological Psychiatry, 91(3). https://doi.org/10.1016/j.biopsych.2021.08.011

Forde, A. T., Crookes, D. M., Suglia, S. F., & Demmer, R. T. (2019). The weathering hypothesis as an explanation for racial disparities in health: A systematic review. Annals of Epidemiology, 33, 1–18. https://doi.org/10.1016/j.annepidem.2019.02.011

Fuller-Rowell, T. E., Curtis, D. S., El-Sheikh, M., Duke, A. M., Ryff, C. D., & Zgierska, A. E. (2017). Racial discrimination mediates race differences in sleep problems: a longitudinal analysis. Cultural Diversity and Ethnic Minority Psychology, 23(2), 165–173. https://doi.org/10.1037/cdp0000104

Giurgescu, C., Engeland, C. G., Templin, T. N., Zenk, S. N., Koenig, M. D., & Garfield, L. (2016). Racial discrimination predicts greater systemic inflammation in pregnant African American women. Applied Nursing Research: ANR, 32, 98–103. https://doi.org/10.1016/j.apnr.2016.06.008

Giurgescu, C., Zenk, S. N., Engeland, C. G., Garfield, L., & Templin, T. N. (2017). Racial discrimination and psychological wellbeing of pregnant women. MCN, the American Journal of Maternal/Child Nursing, 42(1), 8–13. https://doi.org/10.1097/nmc.0000000000000297

Glass, J. E., Williams, E. C., & Oh, H. (2020). Racial/ethnic discrimination and alcohol use disorder severity among United States adults. Drug and Alcohol Dependence, 216(108203), 108203. https://doi.org/10.1016/j.drugalcdep.2020.108203

Guidi, J., Lucente, M., Sonino, N., & Fava, Giovanni A. (2020). Allostatic load and its impact on health: A systematic review. Psychotherapy and Psychosomatics, 90(1), 1–17. https://doi.org/10.1159/000510696

Hill, L., & Artiga, S. (2026, June 4). Elimination of federal diversity initiatives: Updates and current status. KFF. https://www.kff.org/racial-equity-and-health-policy/elimination-of-federal-diversity-initiatives-updates-and-current-status/

Hill, L., Artiga, S., Pillai, A., & Rao, A. (2025, March 21). Elimination of federal diversity initiatives: Implications for racial health equity. KFF. https://www.kff.org/racial-equity-and-health-policy/elimination-of-federal-diversity-initiatives-implications-for-racial-health-equity/

Hill, L., Rao, A., Artiga, S., & Ranji, U. (2025, December 3). Racial disparities in maternal and infant health: Current status and key issues. KFF. https://www.kff.org/racial-equity-and-health-policy/racial-disparities-in-maternal-and-infant-health-current-status-and-key-issues/

Johnson, A., Dobbs, P. D., Coleman, L., & Maness, S. (2023). Pregnancy-specific stress and racial discrimination among U.S. women. Maternal and Child Health Journal, 27(2), 328–334. https://doi.org/10.1007/s10995-022-03567-3

Jones, C. P. (2000). Levels of racism: A theoretic framework and a gardener’s tale. American Journal of Public Health, 90(8), 1212–1215. https://doi.org/10.2105/ajph.90.8.1212

KFF. (2023). How history has shaped racial and ethnic health disparities: A timeline of policies and events. KFF. https://www.kff.org/how-history-has-shaped-racial-and-ethnic-health-disparities-a-timeline-of-policies-and-events/

Krieger, N., Smith, K., Naishadham, D., Hartman, C., & Barbeau, E. M. (2005). Experiences of discrimination: Validity and reliability of a self-report measure for population health research on racism and health. Social Science & Medicine, 61(7), 1576–1596. https://doi.org/10.1016/j.socscimed.2005.03.006

Latendresse, G. (2009). The interaction between chronic stress and pregnancy: Preterm birth from a biobehavioral perspective. Journal of Midwifery & Women’s Health, 54(1), 8–17. https://doi.org/10.1016/j.jmwh.2008.08.001

LaVeist, T. A., Pérez-Stable, E. J., Richard, P., Anderson, A., Isaac, L. A., Santiago, R., Okoh, C., Breen, N., Farhat, T., Assenov, A., & Gaskin, D. J. (2023). The economic burden of racial, ethnic, and educational health inequities in the US. JAMA, 329(19), 1682–1692. https://doi.org/10.1001/jama.2023.5965

Lavner, J. A., Hart, A. R., Carter, S. E., & Beach, S. R. H. (2022). Longitudinal effects of racial discrimination on depressive symptoms among Black youth: Between- and within-person effects. Journal of the American Academy of Child & Adolescent Psychiatry, 61(1), 56–65. https://doi.org/10.1016/j.jaac.2021.04.020

Lawrence, W. R., Jones, G. S., Johnson, J. A., Ferrell, K. P., Johnson, J. N., Shiels, M. S., Diez Roux, A. V., & Forde, A. T. (2023). Discrimination experiences and all-cause and cardiovascular mortality: Multi-ethnic study of atherosclerosis. Circulation: Cardiovascular Quality and Outcomes, 16(4). https://doi.org/10.1161/circoutcomes.122.009697

Lewis, T. T., Cogburn, C. D., & Williams, D. R. (2015). Self-reported experiences of discrimination and health: Scientific advances, ongoing controversies, and emerging issues. Annual Review of Clinical Psychology, 11(1), 407–440. https://doi.org/10.1146/annurev-clinpsy-032814-112728

Liu, S. Y., & Kawachi, I. (2017). Discrimination and telomere length among older adults in the United States. Public Health Reports, 132(2), 220–230. https://doi.org/10.1177/0033354916689613

Mekawi, Y., Hyatt, C. S., Maples-Keller, J., Carter, S., Michopoulos, V., & Powers, A. (2021). Racial discrimination predicts mental health outcomes beyond the role of personality traits in a community sample of African Americans. Clinical Psychological Science, 9(2), 183–196. https://doi.org/10.1177/2167702620957318

Meyer, C. S., Schreiner, P. J., Lim, K., Battapady, H., & Launer, L. J. (2019). Depressive symptomatology, racial discrimination experience, and brain tissue volumes observed on magnetic resonance imaging. American Journal of Epidemiology, 188(4), 656–663. https://doi.org/10.1093/aje/kwy282

Miller, H. N., LaFave, S., Marineau, L., Stephens, J., & Thorpe, R. J. (2021). The impact of discrimination on allostatic load in adults: An integrative review of literature. Journal of Psychosomatic Research, 146(110434). https://doi.org/10.1016/j.jpsychores.2021.110434

Moody, D. L. B., Taylor, A. D., Leibel, D. K., Al-Najjar, E., Katzel, L. I., Davatzikos, C., Gullapalli, R. P., Seliger, S. L., Kouo, T., Erus, G., Rosenberger, W. F., Evans, M. K., Zonderman, A. B., & Waldstein, S. R. (2019). Lifetime discrimination burden, racial discrimination, and subclinical cerebrovascular disease among African Americans. Health Psychology, 38(1), 63. https://doi.org/10.1037/hea0000638

Ndugga, N., Hill, L., Rao, A., Pillai, A., & Artiga, S. (2025, December 16). Key data on health and health care by race and ethnicity. KFF. https://www.kff.org/racial-equity-and-health-policy/key-data-on-health-and-health-care-by-race-and-ethnicity/?entry=executive-summary-introduction

Oshri, A., Reck, A. J., Carter, S. E., Uddin, L. Q., Geier, C. F., Beach, S. R. H., Brody, G. H., Kogan, S. M., & Sweet, L. H. (2024). Racial discrimination and risk for internalizing and externalizing symptoms among Black youths. JAMA Network Open, 7(6), e2416491. https://doi.org/10.1001/jamanetworkopen.2024.16491

Pacheco, N. L., Noren Hooten, N., Wu, S. F., Mensah‐Bonsu, M., Zhang, Y., Chitrala, K. N., De, S., Mode, N. A., Ezike, N., Beatty Moody, D. L., Zonderman, A. B., & Evans, M. K. (2025). Genome‐wide transcriptome differences associated with perceived discrimination in an urban, community‐dwelling middle‐aged cohort. The FASEB Journal, 39(3). https://doi.org/10.1096/fj.202402000r

Pantesco, E. J., Leibel, D. K., Ashe, J. J., Waldstein, S. R., Katzel, L. I., Liu, H. B., Weng, N., Evans, M. K., Zonderman, A. B., & Beatty Moody, D. L. (2018). Multiple forms of discrimination, social status, and telomere length: Interactions within race. Psychoneuroendocrinology, 98, 119–126. https://doi.org/10.1016/j.psyneuen.2018.08.012

Paradies, Y., Ben, J., Denson, N., Elias, A., Priest, N., Pieterse, A., Gupta, A., Kelaher, M., & Gee, G. (2020). Racism as a determinant of health: A systematic review and meta-analysis. PLOS ONE, 10(9), 1–48. https://doi.org/10.1371/journal.pone.0138511

Robinson, M. N., & Thomas Tobin, C. S. (2021). Is John Henryism a health risk or resource?: Exploring the role of culturally relevant coping for physical and mental health among Black Americans. Journal of Health and Social Behavior, 62(2), 136–151. https://doi.org/10.1177/00221465211009142

Shammas, M. A. (2011). Telomeres, lifestyle, cancer, and aging. Current Opinion in Clinical Nutrition and Metabolic Care, 14(1), 28–34. https://doi.org/10.1097/mco.0b013e32834121b1

Spears, I. D., Gorelik, A. J., Norton, S. A., Boudreaux, M. J., Wolk, M. W., Siudzinski, J., Paul, S. E., Cox, M. A., Rogers, C. E., Oltmanns, T. F., Hill, P. L., & Bogdan, R. (2026). Cumulative lifespan stress, inflammation, and racial disparities in mortality between Black and White adults. JAMA Network Open, 9(1). https://doi.org/10.1001/jamanetworkopen.2025.54701

Thames, A. D., Irwin, M. R., Breen, E. C., & Cole, S. W. (2019). Experienced discrimination and racial differences in leukocyte gene expression. Psychoneuroendocrinology, 106, 277–283. https://doi.org/10.1016/j.psyneuen.2019.04.016

Torres, L., Geier, T. J., Tomas, C. W., Bird, C. M., Timmer‐Murillo, S., Larson, C. L., & deRoon ‐Cassini, T. A. (2024). Racial discrimination increases the risk for nonremitting posttraumatic stress disorder symptoms in traumatically injured Black individuals living in the United States. Journal of Traumatic Stress, 37(4). https://doi.org/10.1002/jts.23051

Tynes, B. M., Maxie-Moreman, A., Hoang, T.-M. H., Willis, H. A., & English, D. (2024). Online racial discrimination, suicidal ideation, and traumatic stress in a national sample of Black adolescents. JAMA Psychiatry, 81(3). https://doi.org/10.1001/jamapsychiatry.2023.4961

Wallace, M., Crear-Perry, J., Richardson, L., Tarver, M., & Theall, K. (2017). Separate and unequal: Structural racism and infant mortality in the US. Health & Place, 45, 140–144. https://doi.org/10.1016/j.healthplace.2017.03.012

Webb, E. K., Bird, C. M., deRoon-Cassini, T. A., Weis, C. N., Huggins, A. A., Fitzgerald, J. M., Miskovich, T., Bennett, K., Krukowski, J., Torres, L., & Larson, C. L. (2022). Racial discrimination and resting-state functional connectivity of salience network nodes in trauma-exposed Black adults in the United States. JAMA Network Open, 5(1), e2144759. https://doi.org/10.1001/jamanetworkopen.2021.44759

Williams, D. R., Lawrence, J. A., & Davis, B. A. (2019). Racism and health: evidence and needed research. Annual Review of Public Health, 40(1), 105–125. https://doi.org/10.1146/annurev-publhealth-040218-043750

Williams, D. R., Priest, N., & Anderson, N. B. (2016). Understanding associations among race, socioeconomic status, and health: Patterns and prospects. Health Psychology, 35(4), 407–411. https://doi.org/10.1037/hea0000242

Yip, T. (2015). The effects of ethnic/racial discrimination and sleep quality on depressive symptoms and self-esteem trajectories among diverse adolescents. Journal of Youth and Adolescence, 44(2), 419–430. https://doi.org/10.1007/s10964-014-0123-x

Yip, T., Cheon, Y. M., Wang, Y., Cham, H., Tryon, W., & El‐Sheikh, M. (2019). Racial disparities in sleep: Associations with discrimination among ethnic/racial minority adolescents. Child Development, 91(3), 914–931. https://doi.org/10.1111/cdev.13234

More On
Racial Equity and Health Policy Adolescents Children Maternal and Child Health Race/Ethnicity Substance Use

Related Content

Five Key Facts About People Experiencing Homelessness

Racial Disparities in Maternal and Infant Health: Current Status and Key Issues

Optional Medicaid and CHIP Coverage for Lawfully Present Immigrant Children Could Mitigate Coverage Losses Due to 2025 Reconciliation Law

The Business of Health with Chip Kahn

AI: Show Me the Outcomes

June 30, 2026

Video

Audio

Listen On:

Apple

Spotify

YouTube Music

About this Episode

Episode 10, AI Series: Chip talks with Dr. Toyin Ajayi, co-founder and CEO of Cityblock Health, which delivers value-based care to more than 100,000 Medicaid and dual-eligible members across ten states, many of them people of color managing chronic conditions. Ajayi makes a pointed case: Roughly 60 percent of health care AI investment goes to billing, coding, and risk adjustment — making sure someone gets paid — while only a fraction goes to delivering care. If we continue to concentrate AI there, she warns, it will drive up cost without improving outcomes. She says there is a better way — AI built for care can lower costs by improving care for those hardest to reach. She and Chip discuss what that looks like and how Cityblock is using AI now to improve care and the patient experience for its members.

The Host

Headshot photo of Chip Kahn wearing a navy blue suit with a red tie, red pendant on lapel, and glasses.

Chip Kahn

Sr. Visiting Fellow

Charles N. Kahn III is a senior visiting fellow at KFF. He is also a visiting senior fellow at the American Enterprise Institute and a nonresident senior scholar at the University of Southern California’s Schaeffer Center for Health Policy & Economics. He serves as co-chair of the international Future of Health collaborative.

Guest

Toyin Ajayi, MD, MPhil

Co-founder and Chief Executive Officer, Cityblock Health

Dr. Toyin Ajayi is a Board-certified Family Medicine physician and CEO of Cityblock, a value-based healthcare provider for Medicaid and dually eligible beneficiaries. Prior to Cityblock, she served as Chief Medical Officer of Commonwealth Care Alliance, an integrated health plan and care delivery system for Medicare and Medicaid beneficiaries. Dr. Ajayi serves on the Board of Directors of Evolent Health and Foodsmart and is a co-founder of Coalition Partners. She’s an Aspen Institute Henry Crown Fellow and a member of the National Academy of Medicine. She’s been named to Inc.’s Female Founders 500 list, TIME100 Next, Modern Healthcare’s Top Women Leaders in Healthcare, and the STATUS List.

Dr. Ajayi received her undergraduate degree from Stanford University, an MPhil from the University of Cambridge, her medical degree, with Distinction in Clinical Practice, from King’s College London School of Medicine, and in 2024 was awarded an honorary Doctorate of Science from Georgetown University. Board certified in Family Medicine, Dr. Ajayi completed her residency training at Boston Medical Center and practiced as a hospitalist and primary care provider with a focus on patients with chronic, complex and end-of-life needs.

PREVIOUS EPISODES

Guardrails for AI in Health Care — How High?

The AI Arms Race in Administrative Health Care

AI: Rewiring the Machine

SERIES

The Business of Health with Chip Kahn

This weekly podcast features insightful conversations between host Chip Kahn and his guests, who discuss the business of health care, connecting the dots between the health care business, policy, and patients.

The podcast’s first series on AI in health care illuminates how AI is changing health care, and features guests who are deploying this technology, managing its consequences, and designing policy around it.

More On
Other Health Access to Care Affordability Artificial Intelligence Care Coordination Delivery System Dual Eligible Individuals Financing Health Care Utilization Health I.T. Health System Performance Health Workforce Medicaid Work Requirements Medical Technology Medicare's Future Race/Ethnicity
News Release

Poll: People Without a Trusted Health Care Provider Are More Likely to Endorse Vaccine Myths, As Are Those Who Often Use Social Media or AI for Health Information

While More People Identify Vaccine Myths as “Definitely False” than “Definitely True,” At Least Half Are Uncertain About What to Believe

Published: Jun 30, 2026

People who don’t have a trusted health care provider are more likely than people with one to believe or lean toward believing several common myths about vaccines, a new KFF Tracking Poll on Health Information and Trust reveals. Similarly, people who use social media or artificial intelligence (AI) chatbots at least weekly for health information are more likely than those who don’t to endorse these false vaccine claims.

One example: Among adults who say they do not have a doctor or other health provider they trust to answer questions about their health, about 4 in 10 (39%) incorrectly believe that it is either “definitely” or “probably true” that MMR vaccines have been proven to cause autism in children, compared to a quarter (24%) among those who say they have a trusted provider.

Similarly, more than a third of people who report using social media (37%) or AI chatbots (35%) at least weekly for health information incorrectly say this myth is true, about twice the share among those who never use social media (16%) or AI (20%) for health information.

The poll finds a similar pattern for most of the other vaccine myths tested for people without a trusted doctor as well as for people who frequently use social media or AI for health information. The differences remain significant even when controlling for other factors such as age, race and ethnicity, education, partisanship, and insurance status.

Exposure to each of these false claims has been fairly steady in KFF polls over the past several years, though the share who report hearing the myth that mRNA vaccines can alter a person’s DNA dropped by 9 percentage points since April 2025 (from 45% to 36%). Exposure to the myth that measles vaccines are more dangerous than measles rose between 2024 and 2025, but has remained steady since then (29% now).

Across the four false vaccines claims, far more people say the claims are “definitely false” than say they are “definitely true,” but at least half of the public is less certain what to believe, falling into the malleable middle and saying each of these claims are either “probably true” or “probably false.”

While many parents who skip or delay recommended vaccines for their children express uncertainty over vaccine myths, they are also about twice as likely as parents who keep their children up to date on vaccines to believe or lean toward believing false claims about the measles and COVID-19 vaccines.

The pattern is true for each of the four false claims: that MMR vaccines cause autism in children (57%  among those who delay or skip vaccines v. 30% among those who stay up to date), that more people died from COVID-19 vaccines than the virus itself (55% v. 29%), that mRNA vaccines alter DNA (52% v. 23%), and that measles vaccines are more dangerous than measles (43% v. 18%). This relationship remains significant even when controlling for factors like age, education, and partisanship.

The poll also includes a new analysis that identifies patterns of belief across the four false claims and sorts them into a new belief typology. A small share (8%) are consistent or leaned myth believers (saying all four claims are either “probably” or “definitely true”) and just over half (55%) are consistent or leaned myth deniers (saying all four claims are either “probably” or “definitely false”). About 3 in 10 (31%) are in the mixed middle, providing a range of true and false answers and lacking certainty on at least half of the claims.

Designed and analyzed by public opinion researchers at KFF, this survey was conducted May 7-31, 2026, online and by telephone among a nationally representative sample of 2,480 U.S. adults in English and in Spanish. The margin of sampling error is plus or minus three percentage points for the full sample. For results based on other subgroups, the margin of sampling error may be higher.

More On
Health Information and Trust Public Opinion Artificial Intelligence Social Media Vaccines

Contact

Craig Palosky
Poll Finding

KFF Tracking Poll on Health Information and Trust: Update on Common Vaccine Myths

Published: Jun 30, 2026

News Release: Poll: People Without a Trusted Health Care Provider Are More Likely to Endorse Vaccine Myths, As Are Those Who Often Use Social Media or AI for Health Information

Section:
0 / 0

Findings

Copy link to Findings

Key Takeaways

  • With childhood vaccination rates in the U.S. continuing to decline as measles cases rise across the U.S., KFF’s latest Tracking Poll on Health Information and Trust shows that several commonly circulated vaccine myths remain pervasive among the public. Many adults say they have heard false claims about the measles and COVID-19 vaccines, including that the measles, mumps, and rubella (MMR) vaccines have been proven to cause autism in children (66%), that more people have died from the COVID-19 vaccines than the virus (46%), that mRNA vaccines can alter DNA (36%), or that measles vaccines are more dangerous than measles itself (29%).
  • While many have heard of these myths, smaller shares are convinced they are true. Fewer than one in ten adults express ardent belief in each myth, while larger shares (between 31% and 44%) say each are “definitely false” and at least half fall in the “malleable middle,” saying each of these claims is either “probably true” or “probably false.”
  • Adults who have a relationship with a trusted health care provider are less likely than those who don’t have such a relationship to believe or lean toward believing vaccine falsehoods. For instance, nearly half (46%) of adults who say they do not have a health care provider they trust to answer questions about their health say it is “probably” or “definitely true” that more people have died from COVID-19 vaccines than from the virus, which is nearly twice the share among those with a trusted provider (24%). While younger adults, Hispanic adults, and uninsured adults are more likely than their counterparts to say they don’t have a trusted provider, the connection between lacking a trusted provider and belief in vaccine myths holds even when controlling for factors like age, race and ethnicity, education, partisanship, and insurance coverage.
  • Those who use social media and artificial intelligence (AI) chatbots for health information are also more likely to endorse many of these vaccine myths. For example, adults who use social media for health information at least weekly are more than twice as likely as those who don’t use social media for health to say the myth linking MMR vaccines to autism is “probably” or “definitely true” (37% v. 16%). Use of artificial intelligence (AI) for health information is also correlated with views on some of these myths, with adults who regularly use AI for health information more likely than non-users to believe or lean toward believing myths about the MMR and mRNA vaccines. While younger adults, Black and Hispanic adults, and those without a college degree are all more likely to use to social media for health information, the connection between frequent use and belief in vaccine myths holds even when controlling for factors like age, race and ethnicity, education, and partisanship.
  • Parents’ views on vaccine myths are also correlated with their decisions about childhood vaccinations. Parents who report skipping or delaying recommended childhood vaccines are consistently at least 25 percentage points more likely than those who keep their children up-to-date to say vaccine myths are “definitely” or “probably true,” including the false claims that MMR vaccines cause autism in children (57% v. 30%), that more people died from COVID-19 vaccines than the virus itself (55% v. 29%), that mRNA vaccines alter DNA (52% v. 23%), and that measles vaccines are more dangerous than measles (43% v. 18%). This relationship remains significant even when controlling for factors like age, education, and partisanship.
  • When looking at patterns of belief across the four false vaccine claims, a new analysis shows that some adults are consistent or leaned myth believers (8% who say all four claims are either “probably” or “definitely true”), or myth deniers (55% who deny all four claims, saying they are either “probably” or “definitely false”). At the same time, about three in ten (31%) fall in a “mixed middle” group, providing a range of true and false answers across the four vaccine myths and lacking certainty on at least half of the false claims. This group may be an important focus for those looking to counter vaccine misinformation and dispel confusion. Black adults, Hispanic adults, Republicans, younger adults, and those without a college degree are all more likely than their counterparts to fall into this “mixed middle” group, as are individuals who go to social media or AI for health information. Notably, nearly half of parents who report skipping or delaying recommended vaccines for their children fall in the “mixed middle,” indicating that these parents’ decisions may be driven, at least in part, by uncertainty and confusion.

Exposure to Common Vaccine Myths

The latest KFF Tracking Poll on Health Information and Trust examines the pervasiveness of several commonly circulated vaccine myths. These false or unproven claims about vaccines have remained persistent in terms of exposure over the past several years, with little change to the share of the public have heard most of these myths.

The false claim that MMR vaccines have been proven to cause autism in children – a myth associated with a since retracted study from the 1990s – remains one of the most widely heard vaccine myths, with two-thirds of adults saying they have heard or read this. Nearly half (46%) of adults say they have heard the false claim that more people have died from the COVID-19 vaccines than the virus itself, and about one third of adults (36%) say they have heard the myth that mRNA vaccines can change your DNA (mRNA is a vaccine technology utilized by some COVID-19 vaccines and others under development). About three in ten adults (29%) say they have heard the false claim that measles vaccines are more dangerous than measles infections.

Exposure to each of these false claims has been fairly steady in KFF polls over the past several years, though the share who report hearing the myth that mRNA vaccines can alter a person’s DNA dropped by 9 percentage points since April 2025 (from 45% to 36%). Exposure to the myth that measles vaccines are more dangerous than measles rose between 2024 and 2025, but has remained steady since then.

Figure 1

Four line charts showing the percentage of the public who have heard four false claims about vaccines over time, from June 2023 to June 2026.

The Share of the Public Who Have Heard Several Vaccine Myths Has Remained Fairly Steady Over Time

Percent who have heard the following false claims about vaccines:

Note: MMR refers to the measles, mumps, and rubella vaccine. See topline for full question wording.

Source: KFF Tracking Poll on Health Information and Trust (May 7-31, 2026)

Uncertainty Surrounding False Vaccine Claims

While many report having heard some false claims about vaccines, very few adults are ardent believers in these myths, with larger shares (but fewer than half) stating the myths are “definitely false.” At the same time, and consistent with past KFF polling, at least half of adults fall into the “malleable middle” across the myths, expressing some uncertainty and saying these vaccine myths are either “probably true” or “probably false.” Across these four vaccine falsehoods, most parents fall into the malleable middle, expressing some uncertainty for each claim.

Stacked bar chart showing the share of the public who believe four false vaccine claims are definitely true, probably true, probably false, or definitely false.

At Least Half of the Public Express Some Uncertainty When it Comes to Belief in Falsehoods About the Measles and COVID-19 Vaccines

Stacked bar chart showing the share of the public who believe four false vaccine claims are definitely true, probably true, probably false, or definitely false.

See topline for full question wording.

KFF Tracking Poll on Health Information and Trust (May 7-31, 2026)

Over time, the share who fall in the “malleable middle” for these vaccine myths has been relatively stable, with at least half saying each is either “probably true” or “probably false.” At the same time, there have been some minor shifts in the share who view some of these myths as “definitely false” over the past few years, underscoring how the public’s willingness to endorse vaccine falsehoods is not completely static. Somewhat larger shares of adults now say it is “definitely false” that mRNA vaccines can change your DNA (31% now v. 24% in April 2025), and that measles vaccines are more dangerous than getting infected with measles (44% now v. 38% in March 2024). Conversely, the share of adults who think it is “definitely false” that more people have died from COVID-19 vaccines than from the virus has declined (39% now compared to 47% in June 2023) alongside a 5-percentage point jump in the share who say this myth is “probably true.” There have been no notable changes in views of the long-standing myth that MMR vaccines cause autism.

Stacked bar chart showing how belief in four false vaccine claims has changed over time, from June 2023 to June 2026.

Over Time, at Least Half of Adults Consistently Fall in the “Malleable Middle” on Belief in Vaccine Myths

Stacked bar chart showing how belief in four false vaccine claims has changed over time, from June 2023 to June 2026.

See topline for full question wording.

KFF Tracking Poll on Health Information and Trust (May 7-31, 2026)

Belief in Vaccine Myths is Tied to Lacking a Trusted Health Care Provider And Use of Social Media and AI For Health Information

Previous KFF polls have shown that health care providers are the most trusted source of health information among the public, and this latest poll shows that individuals who have a trusted provider are less likely than those without a trusted provider to endorse vaccine-related myths. For example, among adults who say they do not have a doctor or health care provider they trust to answer questions about their health (16% of all adults), nearly half (46%) say it is either “definitely” or “probably true” that more people have died from COVID-19 vaccines than from the COVID-19 virus, compared to a quarter (24%) among those who say they have a trusted health care provider.

Unsurprisingly, adults who say they do not have a trusted health care provider to answer their health questions are much more likely to be uninsured than those who have a provider they trust (36% v. 7%). At the same time, lacking a trusted provider could be related to low trust in providers and doctors more broadly and not necessarily related to health care access. Across demographics, adults under age 50, LGBT adults, and Hispanic adults are all more likely than their counterparts to say they don’t have a provider they trust to answer questions about their health. However, the connection between not having a trusted provider and belief in vaccine myths remains significant even when controlling for factors like age, race and ethnicity, education, partisanship, and insurance coverage.

Grouped bar chart showing the percentage of adults who believe four false vaccine claims are definitely or probably true, comparing adults with and without a trusted health care provider.

Adults Without a Trusted Health Care Provider Are More Likely to Believe Vaccine Myths

Grouped bar chart showing the percentage of adults who believe four false vaccine claims are definitely or probably true, comparing adults with and without a trusted health care provider.

Adults with or without a trusted health care provider are those who say they do or do not have a doctor or other medical provider who they trust to answer questions about their own health. See topline for full question wording.

KFF Tracking Poll on Health Information and Trust (May 7-31, 2026)

Frequent health information-seeking on social media and via artificial intelligence (AI) chatbots is also tied to a tendency to believe vaccine myths. Adults who say they use social media for health information and advice at least weekly (26% of all adults) are more likely than those who never use social media for health to say each false vaccine claim is “probably” or “definitely true.” Similarly, adults who report using AI tools or chatbots at least weekly for health advice (20% of all adults) are more likely than those who never use these chatbots to endorse myths about the measles vaccines and mRNA vaccines. For example, the share who say it is “probably” or “definitely true” that MMR vaccines have been proven to cause autism is higher among adults who seek health advice at least weekly from social media (37%) or AI (35%) compared to those who never use social media (16%) or AI (20%) for health advice.

The relationship between belief in vaccine myths and use of social media or AI for health information continues to be significant even when controlling for factors like age, race and ethnicity, education, and partisanship.

Bar chart showing the percentage of adults who believe four false vaccine claims are definitely or probably true, broken down by how often they use social media and AI tools for health information.

Use of Social Media and AI Is Tied to Belief in Vaccine Myths

Bar chart showing the percentage of adults who believe four false vaccine claims are definitely or probably true, broken down by how often they use social media and AI tools for health information.

See topline for full question wording.

KFF Tracking Poll on Health Information and Trust (May 7-31, 2026)

Belief in Vaccine Myths is Correlated with Parents’ Decisions to Skip or Delay Recommended Childhood Vaccines

Parents who skip or delay recommended vaccines for their children are about twice as likely as parents who keep their children up to date on vaccines to believe or lean toward believing false claims about the measles and COVID-19 vaccines, underscoring how false health claims may shape parents’ decisions.

Overall, at least three in ten parents say it is “probably” or “definitely true” that MMR vaccines have been proven to cause autism in children (36%), that more people have died from COVID-19 vaccines than the virus (35%), or that mRNA vaccines can alter DNA (29%). About one in four (23%) parents believe or lean toward believing the myth that measles vaccines are more dangerous than measles.

The tendency to endorse these false vaccine claims, however, rises substantially among parents who report not keeping their children up to date on recommended vaccines. Six in ten (57%) parents who report having skipped or delayed recommended childhood vaccines (excluding seasonal vaccines for COVID-19 and flu), say it is either “definitely true” or “probably true” that the MMR vaccines have been proven to cause autism in children, while around half believe or lean toward believing that the COVID-19 vaccines killed more people than the virus (55%), or that mRNA vaccines alter DNA (52%). About four in ten (43%) parents who skipped or delayed childhood vaccines say it is true that the measles vaccines are more dangerous than measles itself. Each of these shares is at least 25 percentage points higher than among parents who report keeping their children up to date on recommended childhood vaccines, a relationship that remains significant even when controlling for factors like age, education, and partisanship. 

Bar chart showing the percentage of parents who believe four false vaccine claims are definitely or probably true, broken down by total parents and by whether they have skipped or delayed their children's vaccines or kept them up to date.

Skipping or Delaying Recommended Vaccines for Children Is Tied to Parents’ Belief in Vaccine Myths

Bar chart showing the percentage of parents who believe four false vaccine claims are definitely or probably true, broken down by total parents and by whether they have skipped or delayed their children's vaccines or kept them up to date.

Parents are parents or guardians of a child under the age of 18 living in their household. See topline for full question wording.

KFF Tracking Poll on Health Information and Trust (May 7-31, 2026)

Digging Deeper on The Malleable Middle: Patterns of Belief Across Vaccine Myths

While KFF polling has routinely found that at least half the public fall in the “malleable middle” when it comes to a wide variety of false health claims, there are nuances within this group that can be examined by looking at patterns of belief across multiple myths rather than examining a single question.

This new typology identified five groups based off patterns of belief across the four false vaccine claims included in this survey:

  • Consistent myth believers (1% of the public) say all four vaccine myths are true, including at least three out of four as “definitely true.”
  • Leaned myth believers (6% of the public) say all four vaccine myths are true but are somewhat uncertain in their beliefs, saying at least two of the four myths are “probably true.”
  • The mixed middle (31% of the public) provide a range of true and false responses and at least half of the time provide a “probably” response (in either the true or false direction).
  • Leaned myth deniers (26% of the public) say all four vaccine myths are false but are somewhat uncertain, saying at least two of the four myths are “probably false.”
  • Consistent myth deniers (29% of the public) say all four vaccine myths are false, including at least three out of four as “definitely false.”

The “mixed middle” group, making up 31% of all adults, reflects a portion of the “malleable middle” that expresses the most uncertainty and does not routinely land on one side when it comes to commonly circulating vaccine falsehoods. The share who fall in the “mixed middle” differs by partisanship, education, race and ethnicity, and age. For example, four in ten Hispanic adults and about a third (35%) of Black adults are part of this group compared to fewer than three in ten white adults (28%). Republicans (44%) are more than twice as likely as Democrats (18%) to be part of this group, while independents (31%) fall in between the two. Those without a college degree (36%) are also 10 percentage points more likely than college graduates (23%) to fall into this mixed middle group. These findings suggest these groups who disproportionately fall in the “mixed middle” may be an important focus for those looking to counter vaccine misinformation and dispel confusion. 

Stacked bar chart showing the percentage of adults who fall into five belief categories — consistent myth believers, leaned myth believers, the mixed middle, leaned myth deniers, and consistent myth deniers — across four vaccine-related myths, broken down by total adults, total parents, age, race and ethnicity, party identification, and education.

Few Adults Are Consistent Believers Across Vaccine Myths, While Three in Ten Fall in The "Mixed Middle" Providing a Mix of True, False, and Uncertain Answers

Stacked bar chart showing the percentage of adults who fall into five belief categories — consistent myth believers, leaned myth believers, the mixed middle, leaned myth deniers, and consistent myth deniers — across four vaccine-related myths, broken down by total adults, total parents, age, race and ethnicity, party identification, and education.

<details> <summary> Category definitions: </summary> Consistent myth believers are those who say all four vaccine myths are true, including at least three out of four as “definitely true.” Leaned myth believers say all four vaccine myths are true but are somewhat uncertain in their beliefs, saying at least two of the four myths are “probably true.” The mixed middle are those who provide a range of true and false responses and at least half of the time provide a “probably” response (in either the true or false direction). Leaned myth deniers say all four vaccine myths are false but are somewhat uncertain, saying at least two of the four myths are “probably false.” Consistent myth deniers say all four vaccine myths are false, including at least three out of four as “definitely false.” </details> <details> <summary> Vaccine myths: </summary> The four vaccine myths include: 1) the MMR vaccines have been proven to cause autism in children; 2) getting the measles vaccine is more dangerous than becoming infected with measles; 3) more people died from the COVID-19 vaccines than from COVID-19 virus; 4) mRNA vaccines can change your DNA </details> See topline for full question wording.

KFF Tracking Poll on Health Information and Trust (May 7-31, 2026)

Adults who are frequently using social media for health advice are also more likely to lack certainty across vaccine myths, providing a range of mixed answers and saying at least half of the four myths are either “probably true” or “probably false.” When looking at belief across different vaccine myths, adults who report using social media for health information on at least a weekly basis are twice as likely to fall in the “mixed middle” group as those who never use social media for health (41% v. 21%). Adults who use AI tools for health information at least occasionally are also somewhat more likely to fall into the “mixed middle” compared to those who never use AI for health, but the difference is smaller for AI than it is for social media use.

Stacked bar chart showing the percentage of adults who fall into five belief categories — consistent myth believers, leaned myth believers, the mixed middle, leaned myth deniers, and consistent myth deniers — across four vaccine-related myths, broken down by how often they use social media and AI tools for health information.

Those Who Report Using Social Media at Least Weekly Are More Likely To Fall Into the “Mixed Middle” on False Vaccine Claims

Stacked bar chart showing the percentage of adults who fall into five belief categories — consistent myth believers, leaned myth believers, the mixed middle, leaned myth deniers, and consistent myth deniers — across four vaccine-related myths, broken down by how often they use social media and AI tools for health information.

<details> <summary> Category definitions: </summary> Consistent myth believers are those who say all four vaccine myths are true, including at least three out of four as “definitely true.” Leaned myth believers say all four vaccine myths are true but are somewhat uncertain in their beliefs, saying at least two of the four myths are “probably true.” The mixed middle are those who provide a range of true and false responses and at least half of the time provide a “probably” response (in either the true or false direction). Leaned myth deniers say all four vaccine myths are false but are somewhat uncertain, saying at least two of the four myths are “probably false.” Consistent myth deniers say all four vaccine myths are false, including at least three out of four as “definitely false.” </details> <details> <summary> Vaccine myths: </summary> The four vaccine myths include: 1) the MMR vaccines have been proven to cause autism in children; 2) getting the measles vaccine is more dangerous than becoming infected with measles; 3) more people died from the COVID-19 vaccines than from COVID-19 virus; 4) mRNA vaccines can change your DNA </details>See topline for full question wording.

KFF Tracking Poll on Health Information and Trust (May 7-31, 2026)

Nearly half (45%) of parentswho have skipped or delayed recommended vaccines for their children fall in this “mixed middle” group, underscoring the connection between confusion, uncertainty and parents’ decisions to forgo recommended vaccines for their children.

Stacked bar chart showing the percentage of parents who fall into five belief categories — consistent myth believers, leaned myth believers, the mixed middle, leaned myth deniers, and consistent myth deniers — across four vaccine-related myths, broken down by total parents and by whether they have skipped or delayed their children's vaccines or kept them up to date.

While Parents Who Skip or Delay Recommended Childhood Vaccines Are More Likely To Believe Vaccine Myths, a Larger Share Express Mixed, Uncertain Beliefs

Stacked bar chart showing the percentage of parents who fall into five belief categories — consistent myth believers, leaned myth believers, the mixed middle, leaned myth deniers, and consistent myth deniers — across four vaccine-related myths, broken down by total parents and by whether they have skipped or delayed their children's vaccines or kept them up to date.

<details> <summary> Category definitions: </summary> Consistent myth believers are those who say all four vaccine myths are true, including at least three out of four as “definitely true.” Leaned myth believers say all four vaccine myths are true but are somewhat uncertain in their beliefs, saying at least two of the four myths are “probably true.” The mixed middle are those who provide a range of true and false responses and at least half of the time provide a “probably” response (in either the true or false direction). Leaned myth deniers say all four vaccine myths are false but are somewhat uncertain, saying at least two of the four myths are “probably false.” Consistent myth deniers say all four vaccine myths are false, including at least three out of four as “definitely false.” </details> <details> <summary> Vaccine myths: </summary> The four vaccine myths include: 1) the MMR vaccines have been proven to cause autism in children; 2) getting the measles vaccine is more dangerous than becoming infected with measles; 3) more people died from the COVID-19 vaccines than from COVID-19 virus; 4) mRNA vaccines can change your DNA </details> See topline for full question wording.

KFF Tracking Poll on Health Information and Trust (May 7-31, 2026)

Methodology

Copy link to Methodology

This KFF Tracking Poll on Health Information and Trust was designed and analyzed by public opinion researchers at KFF. The survey was conducted May 7 – 31, 2026, online and by telephone among a nationally representative sample of 2,480 U.S. adults in English (2,407) and Spanish (73).

The sample includes 1,977 who were reached through an address-based sample (ABS) and completed the survey online (1,819) or over the phone (158). An additional 503 respondents were reached through a random digit dial telephone (RDD) sample of prepaid (pay-as-you-go) cell phone numbers. Among this prepaid cell phone component, 223 were interviewed by phone and 280 were invited to the web survey via short message service (SMS). Marketing Systems Groups (MSG) provided both the ABS and RDD samples. All fieldwork was managed by SSRS of Glen Mills, PA; sampling design and weighting was done in collaboration with KFF.

Both the ABS and RDD sample frames included disproportionate stratification aimed at reaching Hispanic and non-Hispanic Black respondents. The ABS was also stratified based on model-based prediction of household-members’ party identification (Republican, Democratic, or independent).

Respondents received a $15 incentive for their participation, with interviews completed by phone receiving a mailed check and web respondents receiving an electronic gift card incentive.

In order to ensure data quality, cases were removed if they failed two or more quality checks: (1) attention check questions in the online version of the questionnaire, (2) had over 30% item nonresponse, or (3) had a length less than one quarter of the mean length by mode. Likewise, cases that were reached through ABS who reported a living in a different state than the sampled address were removed for quality assurance. Based on this criterion, 39 cases were removed.

The combined ABS and cell phone samples were weighted to match the sample’s demographics to the national U.S. adult population using data from the Census Bureau’s 2025 Current Population Survey (CPS). The combined sample was weighted by gender by age, gender by education, age by education, race/ethnicity by education, education, race, census region, population density, frequency of internet usage, recalled 2024 vote by quintiles of the county-level 2024 vote share. The weights also take into account differences in the probability of selection for each sample type (ABS and prepaid cell phone). This includes adjustment for the sample design and geographic stratification of the samples, and within household probability of selection. The population density benchmark was from the 2026 Claritas Pop-Facts Premier. The internet frequency benchmarks was from the 2025 National Public Opinion Reference Survey (NPORS) data. The county-level 2024 vote share was from CNN-provided file of 2024 election results by county

The margin of sampling error including the design effect for the full sample is plus or minus 3 percentage points. Numbers of respondents and margins of sampling error for key subgroups are shown in the table below. For results based on other subgroups, the margin of sampling error may be higher. Sample sizes and margins of sampling error for other subgroups are available by request. Sampling error is only one of many potential sources of error and there may be other unmeasured error in this or any other public opinion poll. The following questions included in this survey were designed, analyzed, and paid for by KFF. The demographic questions included in this study were developed and funded jointly by CNN and KFF, with each organization having independent editorial control over its portion of the survey. KFF Public Opinion and Survey Research is a charter member of the Transparency Initiative of the American Association for Public Opinion Research.

GroupN (unweighted)M.O.S.E.
Total2,480± 3 percentage points
   
Race/Ethnicity  
White, non-Hispanic                                                                                       1,355± 3 percentage points
Black, non-Hispanic435± 6 percentage points
Hispanic420± 7 percentage points
   
Age  
18-29399± 7 percentage points
30-49888± 4 percentage points
50-64590± 5 percentage points
65+556± 6 percentage points
   
Party ID  
Democrats774± 5 percentage points
Independents876± 5 percentage points
Republicans607± 5 percentage points
   
Parents of children under 18682± 5 percentage points

More On
Health Information and Trust Public Opinion Children Measles Public Health Vaccines

DOWNLOAD

TOPLINE & METHODOLOGY

Related Content

KFF Tracking Poll on Health Information and Trust: The Public’s Views on Measles Outbreaks and Misinformation

KFF Tracking Poll on Health Information and Trust: Tylenol-Autism Link and Vaccine Policies

KFF Tracking Poll on Health Information and Trust: Vaccine Safety and Trust