GLP-1 (glucagon-like peptide-1) drugs were originally developed to help people with type 2 diabetes manage blood sugar levels but have gained widespread attention for their effectiveness as a treatment for obesity. Due to their cost, however, coverage of GLP-1s for obesity treatment in Medicaid, ACA Marketplace plans, and most large employer firms remains limited, and GLP-1 coverage in Medicare for treatment of obesity is prohibited under current law. While state Medicaid programs must cover nearly all Food and Drug Administration (FDA) approved drugs, a long-standing statutory exception allows states to choose whether to cover weight-loss drugs under Medicaid. As a result, Medicaid coverage of GLP-1 drugs for obesity treatment is optional for states, while coverage for other indications (diabetes, cardiovascular disease, and sleep apnea) is required.

The upfront costs of GLP-1s are an ongoing concern for both public and private payers, and some employers and state Medicaid programs are now restricting coverage, despite recognizing their effectiveness at treating obesity. Expanded obesity drug coverage can increase Medicaid spending and put pressure on overall state budgets, and states are now facing tighter budget conditions and longer-term fiscal uncertainty, due in part to the federal Medicaid cuts in the 2025 reconciliation law, causing state Medicaid programs to re-evaluate their obesity drug coverage. However, almost four in ten adults and a quarter of children with Medicaid have obesity, meaning expanding Medicaid coverage of these drugs could provide access to effective obesity treatments for millions. In the longer term, reduced obesity rates among Medicaid enrollees could also result in reduced Medicaid spending on chronic diseases associated with obesity, though the evidence is mixed. Any savings on health spending because of obesity drugs may take many years and may not accrue to the Medicaid program if individuals experience shifts in coverage, so states may not be factoring long-term savings into coverage decisions.

At the federal level, the Trump administration decided not to proceed with a Biden administration proposal to allow Medicare and require Medicaid to cover obesity drugs but recently launched their own obesity drug coverage initiatives to reduce costs and increase access (see Box 1). While lower prices for state Medicaid programs could help alleviate cost concerns for states and result in expanded coverage of obesity drugs, how the new lower costs compare to the net prices state Medicaid programs currently pay and how states will respond amid tightening budget conditions remain unclear. Further, the recent announcements will not impact costs for Medicaid enrollees as they already pay little or no copays for prescription drugs, and the costs of purchasing drugs directly from manufacturers through TrumpRx will likely still be prohibitive for people on Medicaid who must have a low income to qualify for the program. Unless Medicaid covers obesity medications, enrollees are unlikely to have access to them given the high out-of-pocket cost even at lower prices. 

This brief discusses the current landscape of Medicaid GLP-1 coverage and examines recent trends in Medicaid prescriptions and gross spending on GLP-1s. Key takeaways include:

• Obesity drug coverage in Medicaid remains limited, with 13 state Medicaid programs covering GLP-1s for obesity treatment under fee-for-service (FFS) as of January 2026.
• The number of Medicaid prescriptions and gross spending on GLP-1s have both increased substantially since 2019.
• Increased utilization of Ozempic and Wegovy (semaglutide) as well as Mounjaro and Zepbound (tirzepatide) have contributed substantially to recent growth.

Does Medicaid cover GLP-1s for obesity treatment?

States can decide whether to cover obesity drugs under Medicaid. Under the Medicaid Drug Rebate Program, Medicaid programs must cover nearly all of a participating manufacturer’s FDA-approved drugs for medically accepted indications. However, weight-loss drugs are included in a small group of drugs that can be excluded from coverage¹ (though the statutory exception refers to agents used for “weight loss”, “obesity drugs” is used to refer to this group of medications in this analysis). As a result, coverage of GLP-1 drugs for the treatment of obesity remains optional for states, while coverage is required for drugs approved for the treatment of diabetes and, since March 2024 and December 2024, for the treatment of cardiovascular disease (Wegovy) and moderate to severe obstructive sleep apnea in adults with obesity (Zepbound), respectively (Table 1). Coverage is also required if deemed medically necessary for children under Medicaid’s Early and Periodic Screening, Diagnostic and Treatment (EPSDT) benefit.

Obesity drug coverage in Medicaid remains limited, with 13 state Medicaid programs covering GLP-1s for obesity treatment under FFS as of January 2026 (Figure 1). When covered, GLP-1s are typically subject to utilization controls such as prior authorization, which can further limit access. Notably, KFF’s 2025 Medicaid budget survey found 16 state Medicaid programs covered GLP-1s for obesity treatment as of October 2025; however, since then, four states (California, New Hampshire, Pennsylvania, and South Carolina) have eliminated coverage of GLP-1s for obesity treatment, likely reflecting recent state budget challenges and the significant costs associated with coverage. North Carolina eliminated GLP-1 coverage beginning October 2025 due to a budget stalemate in the legislature, but coverage was reinstated in December 2025, bringing the total number of states covering GLP-1s for obesity to 13 as of January 2026. A few other states are planning or considering obesity drug restrictions in state fiscal year 2026 or 2027, and state interest in expanding coverage of obesity drugs is also waning according to this year’s survey, with states continuing to report cost as the key factor contributing to obesity drug coverage decisions. The state obesity drug coverage landscape will continue to evolve as states respond to the recent announcement of the BALANCE model (see Box 1) and as states contend with budget challenges and the federal Medicaid spending cuts in the 2025 reconciliation law.

How have Medicaid prescriptions and gross spending on GLP-1s changed in recent years?

The number of Medicaid prescriptions and gross spending on GLP-1s have increased substantially since 2019 (Figure 2). Not all GLP-1s are approved for obesity treatment, and this analysis includes all FDA-approved GLP-1s, including those approved for obesity (Saxenda, Weogvy, Zepbound) as well as those approved for type 2 diabetes (see Table 1). Overall, the number of GLP-1 prescriptions increased sevenfold, from about 1 million in 2019 to over 8 million in 2024. At the same time, gross spending increased ninefold, from about $1 billion in 2019 to almost $9 billion in 2024, and gross spending per GLP-1 prescription reached $1,000 in 2024. Preliminary trends through June 2025 (data not shown) show rapid growth will continue in 2025. Those prices and spending numbers do not account for rebates, and states typically receive substantial rebates on brand drugs. In response to growing criticism of the cost of their drugs, Novo Nordisk, the company that manufactures Ozempic and Wegovy, reported last year that rebates and other fees (across all payers) accounted for about 40% of the cost of the two drugs and that they expected rebates to grow. GLP-1s still account for a relatively small share of the total number of Medicaid prescriptions, accounting for about 1% of all Medicaid prescriptions in 2024 (up from about 0% in 2019). However, GLP-1s accounted for over 8% of all Medicaid prescription drug spending before rebates in 2024 (up from 1% in 2019).

Specifically, increased utilization of Ozempic and Wegovy (semaglutide) as well as Mounjaro and Zepbound (tirzepatide) have contributed substantially to recent growth. Prescriptions and spending on Ozempic, approved for type 2 diabetes (not obesity) in 2017, have grown considerably over the period. By 2024, Ozempic had surpassed Trulicity, also approved for type 2 diabetes (not obesity) to make up the largest share of GLP-1 prescriptions and spending (39% in 2024). Looking from 2023 to 2024, the latest year of data available, prescriptions and gross spending for Wegovy (first approved for obesity in 2021, approved for cardiovascular disease in 2024) and Mounjaro (approved for type 2 diabetes in 2022) more than doubled, and prescriptions and gross spending for Zepbound (first approved for obesity in 2023, approved for sleep apnea in 2024) increased more than fivefold. These drugs are produced by Novo Nordisk and Eli Lilly, which both recently announced agreements with the Trump administration to lower prices. From Medicaid data publicly available, there is no way yet to disentangle how much of the growing use of GLP-1s is related to treatment for diabetes, cardiovascular disease, or sleep apnea versus obesity, or a combination.

1. Drugs that may be excluded from coverage under the MDRP include drugs used for: a) anorexia, weight loss, or weight gain, b) promoting fertility, c) cosmetic purposes or hair growth, d) symptomatic relief of cough and colds, e) prescription vitamins and mineral products except prenatal vitamins and fluoride preparations , f) nonprescription drugs, g) a manufacturer’s covered outpatient drug in which tests and monitoring services have to be purchased from that manufacturer, h) sexual or erectile dysfunction unless used to treat a condition. For more information see, 42 U.S.C. § 1396r-8.  ↩︎

Introduction

The Trump Administration has taken action to share Medicaid data provided by states to the Centers for Medicare and Medicaid Services (CMS) with the Department of Homeland Security (DHS) and Immigration and Customs Enforcement (ICE) for immigration enforcement purposes. This action represents a reversal in prior policy that asserted CMS would not share such information and ICE would not use such information for immigration enforcement purposes, including assurances made to the public, applicants and enrollees, health care providers, and states that information collected to determine eligibility for health coverage programs would not be used for immigration enforcement.

This issue brief provides an overview of a new data sharing agreement for CMS to share Medicaid data with ICE and its potential implications. This sharp departure in policy to allow sharing of Medicaid enrollee data with ICE for immigration enforcement raises several key issues:

• Data limitations may make it difficult for CMS to comply with a current court order to share data only for individuals who are not lawfully present for the 22 plaintiff states challenging the data sharing. Under pending litigation, CMS is prohibited from sharing certain information from the plaintiff states, including information for lawfully present individuals or citizens. If the data available to CMS are similar to the data researchers have access to, it is not possible to isolate immigrants who are not lawfully present in the dataset that will be shared with ICE.
• Going forward, the sharing of Medicaid enrollee data with ICE will likely make immigrant families, including citizen children in those families, more reluctant to access health coverage and care and other programs and services. The policy exacerbates already heightened fears among immigrant families due to increased immigration enforcement activity overall. Even prior to the public notice of the new data sharing policy, the KFF/New York Times 2025 Survey of Immigrants found that about half (51%) of immigrant adults were “very concerned” or “somewhat concerned” that health officials or health care providers may share their information with ICE or U.S. Customs and Border Protection (CBP).
• The data sharing raises broader questions and concerns about data privacy and protections, which may extend beyond immigrant families. Individuals enrolled in Medicaid did so under prior assurances that information would not be shared for immigration enforcement purposes. Applying new policy retroactively to people who took actions under previous policy and safeguards may broadly increase concerns about privacy and safeguards of data. Health care providers, including community health centers and hospitals, may also have growing concerns about potentially being required to share patient information.

What Medicaid data do states report to CMS?

Medicaid is jointly administered by the federal government and states, and states must share certain information with CMS for program administration. State Medicaid agencies collect and maintain personal and health information for applicants and enrollees to determine eligibility and provide care. Undocumented immigrants are not eligible for Medicaid or other federally-funded coverage, but payments for emergency services may be made to hospitals on behalf of individuals who are otherwise eligible for Medicaid but for their immigration status. States include information about these payments in their reports to CMS. Some states provide health coverage to some immigrants regardless of immigration status using state-only funds. However, states are not required to submit this state-only program information to CMS. States cannot require applicants to provide citizenship or immigration status for any family or household members not applying for coverage.

Under the data sharing agreement, CMS will be sharing data from the Transformed Medicaid Statistical Information System or T-MSIS with ICE, which provides robust information on Medicaid enrollees but is significantly lagged. T-MSIS data also are made available to researchers, providing insight into what data it includes. The data available to researchers are generally almost two years old (e.g., calendar year 2024 data are available in 2026) and include demographic and eligibility files with information on age, gender, race and ethnicity, Medicaid eligibility, Medicaid enrollment, and coverage through Medicaid managed care. There are also claims files that include claims for all services covered by the states, premium payments from the states to managed care plans, and records of services used by managed care enrollees. In some cases, states provide more data to the federal government than are made available to researchers such as KFF. It is likely that CMS has access to more timely data too, but it is unknown how timely those data are and the extent to which timeliness varies across the states.

In the T-MSIS data available to researchers, there are a few types of information related to immigration status, but no indicator that allows for separate identification of immigrants who are undocumented. Specifically, T-MSIS includes the following data elements:

• Immigration status, which indicates whether a person is one of the following:
  • “A qualified alien,” which is a specific subset of noncitizens with lawful immigration status, such as lawful permanent residents (LPRs or “green card” holders), who may be eligible for Medicaid and CHIP and certain other federal programs. Many must wait five years after obtaining qualified status before they may enroll in coverage.
  • “Lawfully present under CHIPRA 214,” which includes lawfully residing pregnant people and children in states that have taken up the option to eliminate the five-year wait to enroll in coverage for eligible immigrants with qualified status and extend coverage to lawfully present pregnant people and/or children beyond those with a qualified status.
  • “Eligible only for payment of emergency Medicaid services,” which includes individuals who would otherwise be eligible for Medicaid but for their immigration status. This group includes undocumented immigrants, noncitizen immigrants with a qualified status who are in the five-year waiting period for coverage, and lawfully present immigrants without a qualified status.
  • “A U.S. citizen or U.S. national.”
• Whether or not the enrollee’s immigration status has been verified, and
• The date on which the five-year waiting period for receiving Medicaid and other federally funded benefits ends.

Undocumented immigrants are included in the population that is “eligible for payment of emergency Medicaid services,” but that group also includes some groups of lawfully present immigrants, and the dataset available to researchers does not separately identify these groups within that population. Additionally, because of the data lag, immigration status may have changed for some people since the data were collected.

What actions has the Trump Administration taken to share Medicaid enrollee data?

In July 2025, CMS and ICE established an Information Exchange Agreement that enables immigration enforcement officials to access the personal data of millions of Medicaid enrollees. A news organization made the agreement public in January 2026, after it was released as part of a lawsuit several states brought against the Department of Health and Human Services and DHS. Under the agreement, CMS will provide ICE, “access to the CMS Integrated Data Repository to retrieve information concerning the identify and location of aliens in the United States.” The agreement indicates this information will allow ICE to, “receive information concerning the identity and location of aliens in the United States, such as address, telephone number, banking information (routing number, account type, account number), email address, internet protocol (IP) addresses and other information relevant to identifying and locating aliens in the United States.”  

The agreement specifies that CMS will provide direct access to the T-MSIS to a select set of ICE employees for renewable two-month periods, and that ICE employees will access the T-MSIS to receive information concerning the identification and location of aliens in the U.S. It identifies Medicaid recipients’ name, address, Medicaid identification number, social security number, date of birth, sex, phone number, locality, ethnicity and race as the data elements to be shared. The agreement does not specify whether ICE access to Medicaid data is limited to a certain subset of Medicaid enrollees, such as noncitizens. As such, the agreement could provide ICE employees access to these data for all Medicaid enrollees, regardless of their immigration status. Also, the stated purpose of the access is to identify and locate aliens in the United States, not specifically those who are undocumented.

This action represents a reversal in prior policy that asserted CMS would not share and ICE would not use health coverage eligibility information for immigration enforcement purposes. Under this prior policy, ICE, CMS, and states made assurances to the public, applicants and enrollees, and health care providers that information collected to determine eligibility for and administer health coverage programs, including Medicaid and the Affordable Care Act (ACA) Marketplace, would not be used for immigration enforcement. CMS previously stated on its website that the information regarding individuals that it collects will only be used for administration of its programs: “To protect your privacy, we’ll tell you before we collect any personal information we need to run our health care programs and only use it for that purpose.” CMS also previously stated on its website that it will not use immigration status for immigration enforcement purposes: “We won’t use any immigration status you share with us for immigration enforcement purposes.”

A preliminary injunction temporarily halted the Trump Administration’s actions to share Medicaid enrollee data with ICE in some states. In June 2025, there were reports that the Trump administration shared the personal and health data of millions of noncitizen Medicaid enrollees living in California, Illinois, Washington, and D.C. with immigration enforcement officials. Following court challenges, a preliminary injunction temporarily blocked the data sharing in the 20  plaintiff states challenging the data sharing in August 2025. The court found that while “there does not appear to be anything categorically unlawful about DHS obtaining data from agencies like HHS for immigration enforcement, that the data sharing was implemented without a “reasoned decision making process” and likely violated the Administrative Procedures Act (APA). The injunction did not govern data sharing in other states that are not involved in the lawsuit.

On November 25, 2025, CMS issued a notice outlining its plans to begin sharing Medicaid data it receives from states with DHS and ICE once the preliminary injunction is lifted. The notice specifies that ICE, “contemplates requesting from CMS available biographical, contact, and location information, though ICE reserved the right, in a specific case, to consider requesting other information on a case-by-case basis as permitted by law.”  CMS indicates that the notice makes the public aware that it intends to disclose certain information to DHS and that it will update relevant websites to reflect its authority to share information with DHS. CMS asserts that the policy change is exempt from a notice and comment period because it is a policy statement outlining actions that are consistent with federal law. ICE also issued a policy memorandum to rescind its previous policy of not using health coverage eligibility information to pursue immigration enforcement action.  

In December 2025, the court updated its ruling, allowing CMS to share a subset of information it intends to share as outlined in the CMS notice. Under the injunction, CMS may only share information with DHS and ICE that is associated with the Medicaid program (not any other health programs); pertains only to individuals who are not lawfully present in the U.S.; and is limited to citizenship and immigration status, address, phone number, date of birth, and Medicaid ID. It prohibits the sharing of any additional information, such as health information, or information related to lawfully present individuals or citizens and precludes sharing additional information that may be requested on a case-by-case basis. It also specifies that, “to the extent that such basic biographical, contact, or location information about unlawfully present aliens is not severable from other information that DHS and ICE are not entitled to obtain (e.g., information about lawful permanent residents or citizens, sensitive health records), HHS or CMS may not share it.” The injunction only applies in the 22 plaintiff states involved in the lawsuit as of December 29, 2025, meaning the data sharing can proceed as outlined in the agreement and notice in other states.

Issues to Consider

Courts will ultimately determine the legality of the data sharing policy within the context of existing data privacy and protection laws. Beyond that determination, there are also issues to consider regarding how the data sharing policy can be operationalized in ways consistent with what the courts have allowed so far and broader implications for families, health care providers, and states.

Data limitations may make it difficult for CMS to comply with the current court order to share data only for individuals who are not lawfully present for the 20 states challenging the data sharing agreement. If the T-MSIS data available to CMS are similar to the data researchers have access to, it is not possible to isolate immigrants who are not lawfully present. Existing T-MSIS data elements identify individuals eligible only for payment of emergency Medicaid services. However, any data shared with ICE using that identifier will include people who are lawfully present (those subject to the five-year waiting period and those who do not have qualified status) as well as individuals who are not lawfully present. The data may also include some individuals who may not have been lawfully present at the time their information was entered into the system, but who have since transitioned to lawfully present status. In the future, the 2025 reconciliation law’s restrictions on Medicaid eligibility for certain lawfully present immigrants (including refugees and asylees among others), which begin in October 2026, will mean that the share of lawfully present immigrants identified as eligible only for payment of emergency Medicaid services will increase.

Going forward, the sharing of health coverage data with ICE will likely make immigrant families, including citizen children in those families, more reluctant to access health coverage and care as well as other programs and services. The policy exacerbates already heightened fears among immigrant families due to increased immigration enforcement activity overall as well as other restrictive immigration policies, including changes in public charge policy and the recission of protections from immigration enforcement in sensitive locations such as health care facilities. Prior to the CMS notice of plans to share data, the KFF/New York Times 2025 Survey of Immigrants found that about half (51%) of immigrant adults were “very concerned” or “somewhat concerned” that health officials or health care providers may share their information with ICE or CBP (Figure 1).  These fears may contribute to individuals delaying or avoiding medical care. About one in seven (14%) immigrant adults, including 48% who are likely undocumented and 20% who are parents, say that they or a family member have avoided seeking medical care since January 2025 due to immigration-related concerns.

The new data sharing policy may increase questions and concerns about data privacy and protections, not only for immigrants but for the public more broadly. People already enrolled in Medicaid did so under prior assurances that information would not be shared for immigration enforcement purposes. Applying new policy retroactively to people who took actions under previous policy and safeguards may broadly increase concerns about privacy and safeguards of data. Health care providers, including community health centers and hospitals, may also have growing concerns about potentially being required to share patient information. Additionally, the policy raises questions about what other data could be shared in the future. This policy represents one piece of broader Trump Administration actions to share data for immigration enforcement. While being challenged in courts, the Trump administration has directed the Internal Revenue Service to share personal information of individuals for immigration enforcement purposes. The Transportation Security Administration is sharing the names of passengers with ICE to seek out and expel people under deportation orders, although ICE previously had avoided interfering with domestic travel. It is unknown whether the Trump Administration might share additional information from other programs in the future for immigration enforcement; or whether future data sharing will extend to other purposes beyond immigration enforcement.

Some Affordable Care Act (ACA) Marketplace enrollees are choosing between plans that charge higher premium payments and plans with higher deductibles. Many enrollees are considering these tradeoffs following the expiration of the ACA’s enhanced premium tax credits at the end of 2025, according to a new Health System Tracker analysis.

While switching from a silver plan to a bronze plan could lower premium payments, the loss of cost-sharing reductions for deductibles and copays and the higher cost-sharing associated with a bronze plan may leave these enrollees worse off financially, depending on how much care they need.

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

1. Not included in this analysis are several regional mechanisms that are still in the planning stages but not yet fully  operational. In addition, we did not include nutrition-related commodities or time-limited arrangements by some of these mechanisms during the COVID-19 emergency (e.g., through COVAX). ↩︎
2. World Bank, Improving Health Product Procurement Efficiency, 2019;  Center for Global Development, Financing Supply Chains: The Missing Link in Global Health, 2025, https://www.cgdev.org/blog/financing-supply-chains-missing-link-global-health; Center for Global Development, Aggregating Demand for Pharmaceuticals is Appealing, but Pooling Is Not a Panacea, 2019, https://www.cgdev.org/publication/aggregating-demand-pharmaceuticals-appealing-pooling-not-panacea;  Global Fund, Sourcing and Management of Health Products, https://www.theglobalfund.org/en/sourcing-management/; Global Fund, Market Shaping Strategy, https://www.theglobalfund.org/en/sourcing-management/market-shaping-strategy/; Parmaksiz K, Bal R, van de Bovenkamp H, Kok MO, “From promise to practice: a guide to developing pooled procurement mechanisms for medicines and vaccines”, J Pharm Policy Pract. 2023 Jun 14;16(1):73. ↩︎
3. PAHO, Essential Medicines for Noncommunicable Diseases Available through the PAHO Strategic Fund, 2021, https://iris.paho.org/handle/10665.2/55092. ↩︎
4. Global Fund, Technical Evaluation Reference Group: Thematic Evaluation of the wambo.org pilot for non-Global Fund financed orders, 2022, https://archive.theglobalfund.org/media/12667/archive_terg-wambo-pilot-evaluation_report_en.pdf; Global Fund, wambo.org Benefits for Buyers, https://www.theglobalfund.org/media/11064/psm_non-grant-program-potential-benefits_table_en.pdf. ↩︎
5. Global Fund, Procurement and NextGen Market Shaping Update, 2025, https://archive.theglobalfund.org/media/skedqy4c/archive_bm53-05-procurement-nextgen-market-shaping_update_en.pdf.
   ↩︎
6. UNICEF handling fees, https://www.unicef.org/supply/handling-fees; PAHO, Report on the Charge Assessed on the Procurement of Public Health Supplies For Member States, 2024, https://www.paho.org/sites/default/files/2024-08/cd61-16-e-charge-assessed-procurement.pdf ↩︎

After failed Senate votes late last year and no subsequent bipartisan agreement, the enhanced premium tax credits expired as of January 1. Some states, particularly those operating State-Based Marketplaces (SBMs), have been preparing for this possibility for months and are moving to blunt the impact on consumers by implementing their own state-funded subsidies and implementing other programs aimed at stabilizing the cost of unsubsidized premiums. 

State-Specific Subsidies

SBMs have the flexibility under the Affordable Care Act to offer additional state-based subsidies on top of federal premium tax credits to further lower monthly premium payments for Marketplace enrollees. A few SBMs have enacted their own supplemental premium subsidies to maintain affordability and enrollment now that the enhanced premium tax credits have lapsed.  

New Mexico has advanced two measures that would backfill the lost enhanced premium tax credits in their entirety for all consumers in 2026. BeWell, New Mexico’s Health Insurance Marketplace, will backfill all of the lost federal tax credits for enrollees with annual incomes up to 400% FPL. Additionally, for enrollees making above 400% of poverty, New Mexico financial assistance will cap premium payments for a benchmark plan at 8.5% of their household income, mirroring the structure of the enhanced premium tax credits. 

Other states have moved to fully backfill the expired tax credits for a portion of enrollees. Maryland, for example, adopted a single-year state premium assistance program that replaces 100% of the lost federal subsidy for enrollees below 200% of the federal poverty level (FPL) ($31,300 for an individual signing up for coverage in 2026) and partial replacement of the lost enhanced tax credit for those with incomes above 200% up to 400% FPL. However, there is no additional state assistance to replace tax credits lost by people with annual incomes above 400% FPL, who are now completely ineligible for any tax credits with the reinstatement of the “subsidy cliff.” 

California is allocating funds to fully replace the premium tax credits for enrollees making up to 150% FPL and partially replace the lost credits for those with incomes between 150% and 165% FPL in 2026. With California receiving about $2 billion annually in enhanced premium tax credits, only a small share of the federal tax credits will be backfilled by these state-specific subsidies. Like in Maryland, California also does not have any additional state assistance to replace enhanced premium tax credits for people with annual incomes above 400% who have lost eligibility for any federal premium assistance since January 1. 

Some states, like Colorado and Washington, have also retooled or created state-specific subsidy programs to provide an additional flat dollar amount to enrollees. However, in many cases, these dollar amounts fall short of entirely backfilling lost federal tax credits. In Colorado, a maximum dollar amount of $80 per month for an individual enrollee (and an additional $29 for each subsequent family member) is provided to households making between 100% to 400% of poverty. This plan will backfill about 40% of the lost federal assistance. Washington is retooling their existing Cascade Care Savings program, setting new fixed dollar maximums for 2026 to provide some relief to consumers ($55 per member per month for those receiving federal tax credits and $250 per member per month for those not receiving subsidies). 

Prior to the creation and expiration of the enhanced tax credits, some SBMs already had additional state assistance in place on top of federal premium tax credits. States such as New York, Connecticut, Vermont, Massachusetts, and New Jersey all have state-specific subsidies that are not directly related to the expiration of the enhanced premium tax credits and will remain in place whether or not they are extended. Additionally, New York and Oregon operate a basic health plan or similar program that provides coverage to certain low-income residents who would otherwise be eligible for Marketplace plans, offering lower premiums and cost sharing regardless of changes to federal tax credit policy. 

Reinsurance Programs

Several states also operate Section 1332 reinsurance programs that reduce unsubsidized premiums by reimbursing insurers for a share of high-cost claims. These programs do not replace lost federal subsidies, but they help stabilize the unsubsidized premiums some consumers will face the full cost of in 2026. Specifically, these are individuals and families with annual incomes over 400% FPL who will be entirely locked out of eligibility for financial assistance with the reinstatement of the “subsidy cliff” and face some of the largest increases in premium payments. 

Maryland’s reinsurance program, in place since 2019 and extended through 2028, has lowered premiums by as much as 35% relative to what they would have been, according to state findings, and is expected to continue softening premium growth when enhanced premium tax credits expire. Colorado, New Jersey, Georgia, and Oregon operate similar programs. Colorado and New Jersey say that these programs have reduced statewide unsubsidized premiums by roughly 20% and provide even greater relief in rural rating areas. In Georgia and Oregon, their reinsurance programs have reduced the cost of unsubsidized premiums by at least 10%. 

Enrollment Assistance

On top of state-based subsidies and reinsurance programs, some states may also increase outreach to help consumers shop for lower-cost options. But navigator programs have faced repeated federal funding cuts, limiting the ability of states, particularly those without robust state-based outreach budgets, to rely on enrollment assistance as a tool for mitigating affordability challenges. 

Limited Relief

A small subset of states (and just a portion of those that operate SBMs) have moved to blunt the impact of the enhanced premium tax credits expiration on enrollees. While expanded state subsidies and reinsurance programs may soften the impact of the enhanced premium tax credit expiration for some consumers, they are unlikely to substantially alter the projected coverage losses. Collectively, these efforts from a small handful of states represent only a small fraction of the roughly $35 billion it would take to extend the enhanced premium tax credits each year, underscoring that state-level actions can mitigate, but not replace, the role of federal policy in sustaining ACA Marketplace coverage. 

As widely expected, and following a recent Presidential memorandum, the Department of Health and Human Services (HHS) issued a memo on January 5, 2026 implementing major changes to the government’s recommended vaccination schedule for children, adding to other changes previously made in 2025. Collectively, these changes reduce the number of vaccines recommended for all children and, as such, have important implications for childhood immunizations and U.S. public health broadly, especially given the context of already declining childhood vaccination rates and ongoing outbreaks of diseases such as influenza and measles. While states, not the federal government, ultimately determine which vaccines are recommended for children and required for school entry, the federal government has significant influence over vaccine policy and access. This policy brief provides an overview and identifies several implications of these changes.

Starting in 2025, changes by HHS to routine vaccine recommendations for children have reduced the number of diseases targeted from 17 to 11 and the number of routine vaccines from 13 to 7 (see Table). Due to changes starting in October of last year, there are now six vaccines no longer recommended for routine use by all children in the United States: rotavirus, COVID-19, influenza, hepatitis A, hepatitis B, and meningococcal vaccines. Instead, some of these vaccines are now recommended for a narrower group of children, based on certain risk criteria or other specific factors, and all six have been moved from the routine category of vaccination to “shared clinical decision making” (SCDM), a process that is “individually based and informed by a decision process between the health care provider and the patient or parent/guardian.”  Beyond these changes, while HPV vaccine remains routinely recommended, HHS has reduced the number of recommended doses from two or three (depending on age of initial vaccination) to one.

Insurance coverage for childhood vaccines is not expected to change, although there is one exception. Most insurers are required to cover the Advisory Committee on Immunization Practices (ACIP)/Centers for Disease Control and Prevention (CDC) recommended vaccines at no-cost (including those recommended through SCDM), either due to requirements of the Affordable Care Act or other federal statutes and as such, these changes should not affect coverage, as also stated by HHS officials. Thus, parents wishing to vaccinate their children against diseases no longer recommended for all children but available through SCDM may still do so without having to be concerned about out- of-pocket costs. One potential exception relates to the HPV vaccine: now that the federal government recommends just one dose of HPV (instead of the previously recommended two to three doses), insurers will not be required to cover an additional dose should parents seek one (the current commercial price for one dose is more than $300). Through the end of 2026, however, health insurers have pledged to continue to cover all vaccines that were recommended in the prior childhood vaccine schedule and several states have moved to mandate free coverage by state-regulated insurers, and could choose to include additional doses.

The new schedule positions the U.S. as an outlier among peer nations in routinely recommending so few vaccines for children. One of the main reasons cited for the new schedule was that the U.S. was an outlier compared to peer countries in recommending so many routine vaccinations for children. Denmark was cited, both at a recent federal advisory committee meeting and in the HHS decision memo, as a model for the U.S. to emulate. The changes made now closely align the U.S. with Denmark (with the only difference being the varicella vaccine which is recommended by the U.S. but not Denmark). No other country among the 20 peer nations that HHS compares the U.S. to in its decision memo recommends vaccines against as few diseases as Denmark, which makes that country, and now the U.S. also, an outlier in this group. For comparison, Australia, Austria, Germany, Greece, Ireland, Italy, Japan, New Zealand, Spain, and the United Kingdom all recommend vaccinating children against 14 or more diseases. More broadly, each country has its own process and unique set of circumstances when it comes to determining vaccine recommendations, and each has developed their childhood vaccination schedule after years of reviewing and weighing available evidence in light of many factors, such as differences in health delivery systems, insurance coverage, public health system capacity, and national priorities. To date no high-income country has made its vaccination decisions based only on what a “peer” country does.

The process used to make these changes marks a departure from the past and further changes are signaled ahead. Historically, any major changes to federal vaccine recommendations were developed through an established, deliberative process that included internal government review with experts from the Center for Disease Control and Prevention (CDC) and other agencies, as well as consideration and public debate via meetings of the Advisory Committee of Immunization Practices (ACIP), the external expert advisory group to CDC. Recent changes made under HHS Secretary Kennedy, however, have circumvented this process, and instead changes have been announced without much internal or external consultation, and without prior public notice. This occurred with Secretary Kennedy’s announcement on social media that HHS was narrowing its COVID-19 recommendations, and now with an announcement of a new childhood vaccine schedule that was not reviewed by CDC experts nor given a public hearing through ACIP.  The White House and HHS have also stated they are examining other aspects of vaccine policy including examining whether the measles, mumps, rubella combination vaccine should be divided into separate shots, how vaccines are tested and safety is monitored, and whether liability protections for manufacturers should be reexamined, which could mean further changes to policy may be forthcoming without a standard review processes.

While HHS cites decreased vaccine uptake and declining trust as additional reasons for changing pediatric vaccine recommendations, it is unclear if these changes will address these issues and they could have the opposite effect. By narrowing the groups recommended to receive certain vaccines, the federal government’s changes could lead parents and providers to choose to vaccinate less often than they have previously. Taking a cue from the new federal vaccine schedule, some states may also loosen vaccine requirements for school attendance, which could also result in lower vaccination coverage. In addition, moving vaccines from a routine, universal recommendation to a shared clinical decision-making (SCDM) recommendation can add an additional step or introduce other barriers to vaccine access, further depressing vaccination rates. With six vaccines newly placed in the SCDM category, such barriers could lead to more missed opportunities for vaccinations. In addition, even though federal health officials state that the new schedule is designed to create more public trust in vaccines, it’s not clear how it may affect trust and it could very well do the opposite. Most parents in the U.S. do not believe there are too many recommended vaccines, though MAGA Republican and MAHA parents – a minority of parents overall – are more likely to agree that the government should recommend fewer vaccines. Many states and expert groups are already making vaccine recommendations that differ from those of the federal government, and the new schedule is likely to exacerbate the conflicting messages that parents and providers hear about vaccines. For now, parents and providers will have to navigate a landscape of conflicting messages and recommendations about vaccines, which sows confusion and ultimately makes it harder for people to know what to do. 

There will be differential impacts of these changes across the country. Ultimately, it is state and local jurisdictions that hold primary responsibility for determining key childhood vaccine policies, including which vaccines are recommended for routine use and which are required for school attendance. In light of changing federal guidelines that in many cases have narrowed vaccine recommendations, manystates have taken steps to de-link state policies and recommendations from those coming from the federal government, particularly Democratic-led states. For example, KFF analysis finds that 24 states no longer use HHS/CDC as a source for vaccine recommendations (up from just 13 in September 2025) and instead turn to state level or external expert groups such as AAP for guidance; a smaller number mandate free insurance coverage by state-regulated insurers for their own set of recommended vaccines. The divergence between federal policy and the states is likely to grow after the latest changes to federal recommendations, which means vaccine coverage and access could increasingly vary according to where one lives. More limited access in some states could, in turn, lead to decreased vaccine coverage and increased incidence of vaccine preventable diseases.

Tracking how the new vaccine schedule impacts vaccination rates in the U.S. may be challenging due to other HHS policies. It remains to be seen how these new recommendations play out in terms of vaccination rates, but recent Trump administration policies may make tracking these changes more challenging. For example, CDC and other federal staffing and funding going to state and local public health efforts across the U.S. has been cut, jeopardizing data collection and analysis. Further, the Trump administration in December announced that, starting in 2026, states will no longer be required to report several measures related to immunization status to HHS as part of their Medicaid and CHIP reporting requirements. Given that nearly four in 10 children in the U.S. are covered by Medicaid, visibility on their vaccine status will be reduced going forward.

In a letter to state health officials last month, the Centers for Medicare & Medicaid Services (CMS) announced the removal of immunization measures from the Child Core Set and Adult Core Set. These “Core Sets” measure health care performance annually, with the goal of improving the health of Medicaid and Children’s Health Insurance Program (CHIP) enrollees. Dropping Medicaid vaccine reporting requirements may make it more challenging to monitor vaccination trends and the impact of recent vaccine policy changes. This policy watch describes the recent Trump administration changes to Medicaid vaccine reporting requirements and explores what impact this change may have on state Medicaid programs and enrollees.

The “Core Sets” are a set of quality measures designed to measure and improve health care quality and access. The Core Set of Children’s Health Care Quality Measures for Medicaid and CHIP, or the “Child Core Set”, includes quality measures that capture various aspects of children’s health such as behavior health care, preventive care, maternal health, acute and chronic conditions, oral health care, and experience of care as well as vaccination status. The Child Core Set was developed in 2009, voluntary annual state reporting began in 2010, and reporting became mandatory in 2024. There is also a separate set of quality measures for adult enrollees, the “Adult Core Set”. The Adult Core Set was developed in 2012, voluntary annual state reporting began in 2014, and the behavioral health measures became mandatory for states to report in 2024. To enforce compliance with mandatory measures, CMS has the authority to withhold federal Medicaid payments, though states can request a one-year exemption if they are unable to report for a specific population and measure. The Core Set data are made publicly available and are designed to measure health care access and quality for Medicaid/CHIP enrollees, allowing states to monitor health care quality, identify improvement opportunities, and address health disparities.

The Trump administration recently removed four immunization measures from the Core Sets, making them voluntary for states to report. CMS’s December 2025 letter to state health officials made all of the immunization-related measures in the Child Core Set and one of two immunization-related measures in the Adult Core Set “voluntary utilization measures” for 2026 and 2027 (see Table 1). The letter also states that federal Medicaid payments are not tied to performance on the immunization quality measures and encourages states not to use the immunization measures in any payment arrangements (for example, incentives for managed care plans). Two other measures for 2027 were retired (Medical Assistance with Smoking and Tobacco Use Cessation, MSC-AD, and the Asthma Medication Ratio, AMR-CH and AMR-AD), though CMS notes that they will explore adding other tobacco use and asthma measures in the future.

The recent Core Set changes did not follow the typical process for updates. To advance and improve the quality measures, CMS is required by federal law to update the Core Sets, ensuring the measures “reflect the testing, validation, and consensus process for the development of pediatric quality measures”. As finalized in federal regulation under the Biden Administration in 2023, the development of the Core Sets each year begins with a workgroup of Medicaid/CHIP stakeholders and quality measure experts who review the Core Sets and make recommendations for changes through the annual review and selection process. The workgroup’s recommendations are then published for public comment and submitted to CMS, who ultimately releases the final Core Sets based on the recommendations. CMS’s recent decision to remove the immunization measures did not follow this process, though the letter notes the Secretary has the “discretion to make changes to the Core Sets that he deems best to improve and strengthen the Core Sets” under federal law.

While seemingly a small, technical change, the removal of vaccine reporting in Medicaid and CHIP may make it more difficult to monitor and understand vaccination trends for a large share of children in the U.S. State-level data from the 2024 Child Core Set for the now removed immunization measures show variation in vaccination rates across states and vaccination type (Figure 1) and trends across the 2022-2024 Core Sets show declines in vaccination rates for some vaccines. These data can help state Medicaid programs understand vaccination trends, compare their state to others, inform state Medicaid policy, and improve rates to reach longstanding vaccination rate goals. Nearly four in 10 children in the U.S. are covered by Medicaid, making even small changes to the program relevant to broader children’s health trends. 

It remains unclear how many states will continue to report the voluntary immunization measures and what will happen to these measures in future years. Prior to mandatory reporting, the number of states reporting the Child Core Set measures increased over time, with voluntary responses to CIS-CH and IMA-CH (show in Figure 1) reaching 46 to 48 responding states by 2023 depending on the vaccine. Given the infrastructure is already in place to report, states may continue to voluntarily report the immunization measures. However, CMS has indicated they will be considering options for new immunization measures to replace the now voluntary measures. They plan to engage stakeholders, including states, quality measure experts, providers as well as vaccine registry managers and electronic health record vendors, to develop measures that capture “whether parents and families were informed about vaccine choices, vaccine safety and side effects, and alternative vaccine schedules” and to explore whether data can account for “person and family preferences related to vaccines” and religious exemptions for vaccinations. Without the Child Core Set immunization measures, at this time it may still be possible to monitor Medicaid/CHIP children's vaccination trends through other data sources, such as the National Immunization Survey or individual state collected data, but these data are not easily accessible and may not be comparable across states. 

Dropping vaccine reporting requirements could also make it more challenging to monitor recent declines in childhood vaccinations rates and the impact of vaccine policy changes. Children’s routine and seasonal vaccination rates have declined in recent years, due in part to rising vaccine hesitancy fueled by vaccine misinformation, increasingly partisan views on vaccine requirements, and a decline in trust of health authorities. In addition, the Administration recently announced significant changes to the children’s vaccination schedule, reducing the number of diseases targeted from 17 to 11 and the number of routine vaccines from 13 to 7. Changes to the vaccine schedule at the federal level (and confusion about the changes) coupled with other public health policy actions under the Trump administration could further drive down vaccination rates among children and increase incidence of vaccine preventable diseases. Limited vaccination status data means reduced visibility into vaccination trends and what impact recent policy changes may have.

In his latest JAMA Forum column, KFF’s Larry Levitt explores how unaffordable health care is in the U.S. in the context of the debate over extending enhanced Affordable Care Act premium tax credits and an upcoming election where affordability will likely be front and center.

Note: This brief was updated on May 21, 2026 to incorporate new data on abortion statistics.

• In the three years since the Supreme Court ruling that overturned Roe v. Wade, the total number of abortions nationally has slightly increased. The most recent data from the Society for Family Planning’s #WeCount project show that the average monthly abortion volume during the first half of 2025 was higher than the monthly average in 2024. From January to June 2025, there have been more than 590,000 abortions compared to 1.14 million abortions in all of 2024 and 1.06 million abortions in 2023. For most of the decade prior to the Dobbs ruling, there was a steady decline in abortion rates nationally, with a slight uptick in the years just before the ruling.
• The upward trend in abortion volume is likely due to multiple reasons, including expanded telehealth capacity, the ability to mail medication abortion pills to patients, and the lower costs for telehealth abortions through virtual clinics compared to in-person care. Medication abortion via telehealth now accounts for 27% of all abortions.
• In contrast to the abortion bans, several states have passed laws to protect abortion access for their residents and expand access to people seeking abortions from other states which have contributed in part to the increased the number of abortions in those states compared to pre-Dobbs time frame. Twelve (12) states require state-regulated private plans to cover abortion, many without cost-sharing, and 20 Medicaid programs use state-only funds to cover nearly all medically necessary abortions. Twenty-three (23) states passed shield laws intended to reduce the legal risks for clinicians who provide abortion care to patients who live in states where abortion is banned or restricted.
• The upward trend in abortion volume can also be attributed to increased interstate travel. The travel rate for abortion care across state lines nearly doubled from 2020 to 2024, with Illinois, North Carolina, New Mexico, and Kansas experiencing the highest volume of out-of-state abortion patients last year.

Following the 2022 ruling in Dobbs v. Jackson Women’s Health Organization, it was generally expected that the abortion rate would drop due to the number of states that rapidly adopted abortion bans (13 states) and early gestational restrictions (6 states). There is no doubt these policies have made abortion access much more challenging or even impossible for those seeking abortion who live in restrictive states; yet, contrary to expectations, recent data show that the number of abortions in the U.S. overall has slightly increased in three years following the Supreme Court ruling. The combination of growth in telehealth availability for abortion care, lower telehealth costs, increased legal reproductive health care protections through state efforts, and higher rates of interstate travel, all likely contributed to the unexpected trajectory in abortion volume. However, the possibility of more state bans and restrictions combined with the ongoing legal challenges seeking to further restrict access may reverse this trend. Additionally, future actions that the Trump administration could take at the federal level could further limit abortion availability and access even in states that have enshrined the right to abortion, particularly if the administration restricts the distribution of medication abortion pills through the Comstock Act or targets the provision of telehealth abortions through regulatory revisions at the Food and Drug Administration.

This brief reviews the different sources of abortion data in the U.S., the factors that have affected abortion rates across the country before and after Dobbs, and what we may see as the Trump administration, Republican majorities in the House and Senate, and a conservative federal judiciary shape policy in the coming years.

How is abortion tracked at the state and federal level?

Three major organizations collect and report national and state-level data on abortion volume and rates: the federal Centers for Disease Control and Prevention (CDC), the Guttmacher Institute, and most recently, the Society of Family Planning through its (SFP) #WeCount project.

For decades, the federal CDC Abortion Surveillance System has requested data from the central health agencies of the 50 states, D.C., and New York City to document the number and characteristics of women obtaining abortions. Reporting to the CDC is voluntary and not all states participate in the surveillance system. Notably, California, Maryland, and New Hampshire have not reported data on abortions to the CDC system for years. Most states collect and report data on the demographic characteristics of patients, gestational weeks, and type of abortion procedure. CDC publishes data from the surveillance system annually, with the most recent data on abortions in 2022, reflecting a 2-year lag. Following the termination of federal staff from the agency’s the Reproductive Health Division, it was unclear whether the CDC would continue to update its Abortion Surveillance System which currently presents data from 2022 that was released in November 2024. It has been reported that the next Abortion Surveillance report is expected to be released during spring of 2026 with data from 2023.

Prior to the Dobbs ruling, the Guttmacher Institute, an independent research and advocacy organization, periodically conducted the Abortion Provider Census (APC), collecting data on abortion incidence, abortion facilities, and patient characteristics. Data from the APC are based primarily on questionnaires completed by known facilities that provide abortion in the country, information from state health departments, and Guttmacher estimates for a small portion of facilities. The most recent APC reports data from 2020. Following the Dobbs ruling, the Guttmacher Institute established an additional data collection initiative, the Monthly Abortion Provision Study, to track abortion volume within the formal U.S. health care system. This ongoing effort collects data on and provides national and state-level estimates on abortions while also tracking the changes in national abortion volume since 2020.

While the CDC and Guttmacher APC data differ in terms of collection methods, timeframe, and completeness, both have shown similar trends in abortion rates over the past decade. One notable difference is that Guttmacher’s survey has included continuous reporting from all states, which explains at least in part the higher abortion volume in their data.

Society of Family Planning’s (SFP) #WeCount is a newer national reporting initiative that measures changes in abortion access following the Dobbs ruling. The project provides semiannual reports on the monthly number of abortions by state and includes data on abortions provided through in-person health care settings and through telehealth. The #WeCount report started collecting data in April 2022 and has published two full years of abortion data since Dobbs.

How has the abortion rate changed over time?

For most of the decade prior to the Dobbs ruling, there was a steady decline in abortion rates nationally, but there was a slight increase in the years just before the ruling.

The most recent CDC data are from 2022, the same year as the Dobbs decision, and show that abortion rates declined from 2013 through 2017 and remained steady in the years leading up to the court decision (Figure 1). CDC reported 609,360 abortions in 2022 and a rate of 11.2 abortions per 1,000 women (excludes CA, DC, MD, NH, and NJ). In contrast, the Guttmacher Institute reported 930,160 abortions in 2020 and a rate of 14.4 abortions per 1,000 women. Guttmacher’s study showed a slight upward trend in abortion from 2017 to 2020 whereas CDC’s report showed a stable rate in abortions from 2017 to 2022 except for a slight uptick in 2019 and 2021.

Experts generally attribute the long-term decline in abortion rates to increased use of more effective methods of contraception. The slight increase in the years leading up to the Dobbs decision could be due to greater state-level coverage of Medicaid enrollees that made abortion access more affordable in some states as well as broader financial support from abortion funds to help individuals pay for the costs of abortion care.

Even prior to the Dobbs ruling, abortion rates varied widely between states.

National averages can mask local and more granular differences. Some of the variation in abortion volume and rates has been due to the wide differences in state policies that have shaped the availability of abortion, with some states historically placing restrictions on abortion (such as targeted regulations of abortion providers, requirements for multiple visits, and mandatory waiting periods), that constrained abortion access and availability. In some states, there were only one or two abortion providers even before Dobbs.

What has happened to abortion volume since Dobbs?

The SFP and Guttmacher Institute data both find that while the number of abortions in the U.S. dropped immediately following Dobbs, the total number of abortions nationally has increased two years following the ruling. However, the consistency observed at the national level obscures wide state-level variation and sharp declines in the number of abortions in states with bans and early gestational restrictions.

The latest SFP’s #WeCount data show that there were 591,770 abortions in the first half of 2025. In 2024, there were an estimated 1.14 million abortions, slightly up from 1.05 million in 2023. The monthly average number of abortions steadily increased from 88,180 abortions per month in 2023 to 95,250 abortions in 2024 to 98,630 in 2025 (January to June 2025)(Figure 3).

Why did the number of abortions increase after states instituted bans?

While it was not a total surprise that states without abortion bans had an increase in abortions following the Dobbs ruling, the reasons behind this increase are complex. The upward trend is likely due to a combination of increased interstate travel for abortion access by people coming from abortion ban states, the presence of state-level laws in states that protect providers who offer abortion services, lower costs associated with telemedicine medication abortions, and expanded virtual/telehealth capacity and the ability to mail medication abortions pills to patients among both bricks-and-mortar and telemedicine-only providers.

The Rise of Medication Abortion, Telehealth, and Virtual Clinics

While procedural abortions are only performed in a clinical setting, medication abortion can be provided either in a clinical setting or remotely via telehealth. Medication accounts for nearly two thirds (63%) of abortions nationally. Approved by the U.S. Food and Drug Administration (FDA) in 2000, medication abortion has a solid safety and effectiveness record regardless of whether the pills are dispensed in person by a clinician (either medical doctor or advanced practice clinician) or via telehealth and mailed or dispensed through a retail pharmacy. When taken, medication abortion successfully terminates the pregnancy 91.9% to 99.7% of the time, with a 0.4% risk of major complications, and an associated mortality rate of less than 0.001 percent (0.0005%). Between 2% to 4% of medication abortions may need a procedural intervention with this regimen. The latest Guttmacher data show that in states without bans, medication accounted for the majority of abortions in 2023 (Figure 4). In five states (MT, WY, NE, GA, and VT), more than eight in ten abortions were medication abortions.

Access to medication abortion via telehealth had been historically limited by an FDA policy (Risk Evaluation Mitigation Strategy or REMS) that had permitted only physicians in a health care setting to dispense mifepristone in person. This resulted in a restriction on the ability to mail the pills or for retail pharmacies to dispense. In December 2021, the FDA revised this policy lifting the requirement that clinicians dispense the drug only in-person. This was done, in part, to alleviate the burden placed on the health care delivery system during the COVID-19 public health emergency. In January 2023, the FDA finalized a policy change that allows retail pharmacies to dispense medication abortion pills to patients with a prescription. These changes opened the door to greater use of telehealth for medication abortions.

The increase in telehealth abortions has also been driven in part by the rise in the number of virtual abortion clinics. The number of virtual clinics began to rise after the FDA revised its in-person dispensing requirement in 2021 and now accounts for a quarter (24%) of facilities that offer medication abortion services.

SFP’s #WeCount study breaks out monthly averages for telehealth abortions, and the most recent report shows that telehealth abortions accounted for 27% of all abortions in the second quarter of 2025 (Figure 5). The latest #WeCount reports distinguish between telehealth abortions provided by brick-and-mortar facilities from those provided under shield laws that give some legal protections to clinicians who provide abortion care via telehealth to people living in states with bans and restrictive policies. More than half of these telehealth abortions were performed under shield laws (56%), 6% of abortions were from online services offered by clinics that traditionally operate from physical locations (brick-and-mortar facilities), and four in ten (39%) were from virtual-only clinics. The provision of telehealth abortions varies widely across states, ranging from 8% of all abortions in some states and reaching 39% of all abortions in other states. Note: The counts for medication abortions, particularly those provided by mail, reflect the number of pills dispensed by providers, not necessarily complete abortions. 

Costs for Telemedicine Abortions

The median price of medication abortion offered through brick-and-mortar clinics increased from $580 in 2021 to $600 in 2023. In contrast, the median price of medication abortions via virtual clinics decreased from $239 in 2021 to $150 in 2023, which is 75% less than the cost of in-person care (Figure 6). Virtual clinics do not incur many of the costs of a physical clinic, such as building maintenance, meeting regulations for surgical centers, and security to handle protesters. The increased availability of telehealth and virtual clinics has lowered the costs of care and reduced financial barriers resulting from abortion services as well as travel and other related expenses.

Costs for some have also been offset by the availability of financial assistance and logistical support from national and local networks of abortion funds. Since Dobbs, these networks received a reported 39% more requests for abortion support and financially supported more than 100,000 individuals seeking abortion care. While donations to these networks increased immediately following Dobbs, the frequency of donations has slowed, and funds have begun to taper. Some organizations recently reported suspending operations altogether, signaling that abortion volume may consequently dwindle as demand outpaces donations.

State-Level Protections

Over the past several years, some of the states where abortion remains legal have passed laws to protect abortion access for their residents and expand access to people seeking abortions from other states. For example, residents in California are protected from civil liabilities for providing or receiving abortion services, and providers are protected from professional discipline. Policies that have been implemented include using state funds to cover abortions under Medicaid beyond federal limitations, raising Medicaid reimbursement rates for abortion services, requiring state-regulated private plans to cover abortion, and enacting shield laws to protect clinicians who provide abortions in their states either in person or via telemedicine.

Today, 12 states require state-regulated private plans to cover abortion, some without any cost-sharing (Figure 7).

State actions to use their own revenues to pay for abortions have also expanded access to abortion services. States are not restricted by the federal Hyde Amendment (which bans the use of federal funds for abortion in Medicaid, Medicare and other public programs unless the pregnancy is a result of rape, incest, or if it endangers the woman’s life) and have the option to use state-only funds to cover abortions under other circumstances for women on Medicaid, which 20 states do currently.

A growing number of states passed shield laws to reduce the legal risks for clinicians who provide abortion care to patients who live in states where abortion is banned or restricted. While the details of these laws vary state to state, some policies protect clinicians from professional discipline for offering health care that is criminalized in another state, and others protect clinicians who provide care to patients across state lines, such as by prescribing and mailing abortion pills via telehealth services to patients in their state of residence. Some states also passed broader shield laws to protect patients and people assisting with reproductive services from civil and criminal consequences. As of February 2025, 22 states and Washington D.C. have enacted shield laws, with 8 states extending explicit protections to clinicians regardless of patient location or state of residence (Figure 8).

Interstate Travel

The Guttmacher Institute Monthly Abortion Provision Study is the only data source so far to provide in-depth information on interstate travel pre- and post-Dobbs. Guttmacher estimates that prior to Dobbs, nearly one in ten people obtained an abortion by traveling across state lines in 2020. Even though abortion was legal, there were considerable restrictions in many states that made abortion access very limited, which led to the need for interstate travel for abortion care for some people. The latest data show that 155,000 patients traveled out of state for abortion care in 2024, a slight drop from 170,000 in 2023 but nearly double the number of travelers in 2020 (81,000). The states with the highest number of people traveling inbound for abortion care border at least one state where abortion is banned, including Illinois (36,920 patients), North Carolina (17,080 patients), Kansas (15,610 patients), and New Mexico (12,680 patients)(Figure 9).

The volume of interstate travel into Florida and North Carolina is especially notable as they were two of the last southern states in 2023 where abortion was legal beyond six weeks of gestation. However, the policies in these two states became more restrictive. North Carolina went from a 20-week ban to a 12-week ban in July 2023, and Florida’s 15-week ban changed to a 6-week ban in May 2024. The volume of interstate travel into Florida declined from 9,080 travelers in 2023 to 3,970 travelers in 2024 (data not shown). Conversely, the volume of interstate travel into North Carolina increased from 16,110 in 2023 to 17,080 in 2024 (data not shown). Virginia also had an increase in interstate travel for abortion, from 5,360 out-of-state travelers in 2023 to 9,410 travelers in 2024. This is largely due to Florida’s stricter gestational stage ban, which curtailed abortion access in the region and has resulted in patients in southern states having to travel elsewhere, such as to northern states, for abortion care.

While the data show that abortions slightly increased nearly three years after Dobbs, ongoing and impending legal challenges, state legislative efforts, and federal executive actions could further alter the reproductive care landscape and have impacts beyond abortion counts. A recent JAMA study, for instance, found that fertility rates have increased in states with complete or 6 week abortion bans, namely among populations with the greatest structural disadvantages and barriers to obtaining abortion care. A concurrent study showed infant mortality rates have also risen in these states, many of which are already experiencing some of the worst maternal, infant, and child health outcomes in the U.S. The findings from these studies underscore the widespread repercussions of policy efforts aimed at restricting abortion access.