BALANCE

Medicare has been prohibited by statute from covering weight loss drugs since 2003. The Medicare Prescription Drug, Improvement, and Modernization Act excluded agents used for weight loss from the definition of a covered Part D drug, a restriction rooted in the fen-phen safety scandal of the late 1990s and the congressional judgment that weight management medications were elective rather than medically necessary. For two decades, that exclusion held. More than 40 percent of Medicare beneficiaries age 60 and older meet the clinical definition of obesity, and none of them could access the most effective pharmacological treatments for it through their Medicare drug benefit.

The BALANCE model, announced December 23, 2025, is the Trump administration’s mechanism for breaching that wall without changing the statute. Using CMMI’s Section 1115A demonstration authority, CMS is testing whether negotiated GLP-1 pricing paired with lifestyle support requirements can expand coverage for weight management while generating or maintaining program savings. It is a voluntary model for manufacturers, states, and Part D plans. It is also the most politically visible CMMI model in the 2025 portfolio, the one the administration has staked the most public capital on, and the one whose cost trajectory is the hardest to predict.

The Statutory Obstacle and the Administrative Workaround

#

The statutory exclusion at Section 1860D-2(e)(2) of the Social Security Act is specific: Part D plans may not cover drugs when used for weight loss. The exclusion applies to the indication, not the molecule. The same semaglutide sold as Ozempic is covered by Part D when prescribed for type 2 diabetes. Sold as Wegovy and prescribed for weight management, it is excluded. Tirzepatide as Mounjaro for diabetes is covered. As Zepbound for weight management, it is not. This indication-level exclusion has created a system in which Medicare spends billions annually on GLP-1 medications for diabetes and cardiovascular indications while prohibiting coverage of the same molecules for the condition most likely to generate the chronic diseases driving that spending.

The Biden administration attempted to resolve this through regulatory reinterpretation. A November 2024 proposed rule would have reinterpreted the statutory exclusion to permit Part D coverage of GLP-1s for the treatment of obesity as a chronic disease, making an estimated 3.4 million beneficiaries newly eligible. The Trump administration declined to finalize that proposal. The April 4, 2025 Part D final rule stated the reinterpretation was “not appropriate at this time,” citing CBO’s estimate that full coverage would increase federal spending by $35.5 billion over nine years.

BALANCE takes a different path. Rather than reinterpreting the statute, CMS invokes CMMI’s demonstration authority to test a payment model that waives the statutory exclusion for participating plans and states. The legal mechanism is the same one that authorizes every other CMMI model: Section 1115A permits the Secretary to waive Medicare and Medicaid requirements as necessary to carry out the test. The demonstration authority is broad, but it is time-limited. BALANCE runs through December 2031. If the model is not certified for expansion or replaced by legislation, the statutory exclusion returns for any beneficiary whose coverage depended on the waiver.

The separate Medicare GLP-1 Bridge, operating from July through December 2026 under Section 402(a)(1)(A) of the Social Security Amendments of 1967, uses a different legal authority entirely: the Secretary’s power to conduct demonstrations testing changes in payment or reimbursement methods. The Bridge is not a CMMI model. It is a short-term payment demonstration that precedes the CMMI model, creating Medicare GLP-1 coverage six months before BALANCE’s Part D component launches.

What BALANCE Actually Does

#

BALANCE establishes CMS as a direct price negotiator with GLP-1 manufacturers on behalf of state Medicaid agencies and Medicare Part D plan sponsors. This is structurally distinct from Part D’s existing rebate and formulary negotiation system, in which manufacturers negotiate with plans and their pharmacy benefit managers. Under BALANCE, CMS negotiates the key terms centrally: guaranteed net pricing, cost-sharing limits, coverage eligibility criteria, and lifestyle support requirements. Those terms then flow through to participating states and Part D plans via standardized agreements.

The November 6, 2025 White House announcement of pricing agreements with Eli Lilly and Novo Nordisk preceded the formal BALANCE model by seven weeks. The deals established the foundational price point: $245 per monthly supply as the net price Medicare and Medicaid will pay for injectable GLP-1 medications across all approved indications, including weight management. That figure represents a reduction from list prices that ranged from approximately $1,000 to $1,350 per month. Starting doses of oral GLP-1 formulations, if approved by the FDA, will be priced at approximately $149 to $150 per month.

CMS completed formal negotiations with both Eli Lilly and Novo Nordisk by late February 2026, with participation agreements executed by February 28. The eligible products include Wegovy (semaglutide) and Zepbound (tirzepatide) for the initial coverage period, with Mounjaro (tirzepatide for diabetes) also covered at the negotiated price across indications. If the FDA approves oral formulations, specifically Novo Nordisk’s oral semaglutide for weight management and Eli Lilly’s orforglipron, they can enter the model through subsequent application cycles.

The model’s eligibility threshold for the weight management indication is narrower than the FDA labeling for these drugs. The FDA approves GLP-1s for weight management in adults with BMI of 30 or above, or BMI of 27 or above with at least one weight-related comorbidity. BALANCE and the Bridge set a higher bar: a BMI of 35 or above, or a BMI of 30 or above with at least one qualifying condition, including type 2 diabetes, noncirrhotic metabolic dysfunction-associated steatohepatitis with moderate to advanced liver fibrosis, or obstructive sleep apnea. Beneficiaries must also be on current and ongoing lifestyle modification including structured nutrition and physical activity. Prior authorization is required for all prescriptions.

The administration has framed these eligibility constraints as targeting the patients most likely to benefit clinically. The practical effect is also fiscal: approximately 10 percent of Medicare beneficiaries are expected to meet the criteria, compared to the roughly 25 percent who might qualify under the broader FDA-approved labeling. That narrowing is what makes the model’s cost trajectory manageable relative to the CBO estimate for full statutory coverage.

The Three-Phase Timeline

#

BALANCE unfolds across three overlapping phases, each with a different legal authority, different operational structure, and different implications for plans and beneficiaries.

Phase 1: Medicaid (May 2026). State Medicaid agencies can begin joining BALANCE on a rolling basis starting May 2026. Participation is voluntary, and states execute supplemental rebate agreements with participating manufacturers. States that participate gain access to the negotiated $245 net price for all covered indications, including weight management, which Medicaid has always had the legal authority to cover but which most states have declined to cover because of cost. As of early 2025, only approximately 13 to 15 states covered GLP-1s for weight loss through Medicaid, and some, including California, eliminated that coverage effective January 2026 due to budget pressure. BALANCE’s price reduction is designed to change that calculus.

Phase 2: The Medicare GLP-1 Bridge (July through December 2026). The Bridge is the mechanism for getting GLP-1s into Medicare beneficiaries’ hands before BALANCE’s Part D component is operational. It operates entirely outside the Part D benefit structure. CMS administers the Bridge through a central processor that handles prior authorization, claims adjudication, and pharmacy reimbursement. Part D plans carry no risk for Bridge prescriptions. Beneficiaries do not need their Part D plan to opt in. The $50 copayment does not count toward the Part D deductible or the $2,100 annual out-of-pocket spending cap. Pharmacies are reimbursed at wholesale acquisition cost minus the beneficiary copay, plus a dispensing fee and applicable sales tax.

The Bridge’s separation from the Part D benefit has important implications. Beneficiaries already receiving GLP-1s for Part D-covered indications, diabetes, cardiovascular disease, sleep apnea, continue accessing those drugs through their Part D plan at their plan’s cost-sharing level, even if that cost-sharing exceeds the Bridge’s $50 copayment. The Bridge covers only the weight management indication that Part D itself cannot. CMS has stated it will monitor to ensure plans do not shift existing covered utilization to the Bridge to reduce their own costs.

Phase 3: BALANCE in Medicare Part D (January 2027). The full model integrates GLP-1 coverage for weight management into the Part D benefit structure. Part D sponsors must apply and be accepted to participate. Participating plans incorporate the negotiated cost-sharing limits for model drugs within the Part D benefit. Most beneficiaries in participating plans will have out-of-pocket costs capped at $50 per month after the deductible is met. Before reaching the deductible, cost-sharing will be limited to $245 per month plus a dispensing fee. Plans that do not participate in BALANCE will not cover GLP-1s for weight management, because the statutory exclusion still applies outside the model waiver. Beneficiaries who accessed GLP-1s through the Bridge in the second half of 2026 and want to continue in 2027 must enroll in a BALANCE-participating plan during the Open Enrollment period.

The transition from Bridge to BALANCE is the model’s most significant operational vulnerability. A beneficiary who starts a GLP-1 in July 2026 through the Bridge and does not switch to a BALANCE-participating Part D plan for 2027 will lose coverage. The information asymmetry is substantial: many beneficiaries will not know which plans are participating in BALANCE until fall 2026 at the earliest, and the Annual Election Period for 2027 coverage runs October 15 through December 7, 2026. That window is tight for a population already navigating one of the most complex enrollment cycles in MA and Part D history.

The Manufacturer Negotiation and What It Reveals

#

The November 2025 pricing agreements with Eli Lilly and Novo Nordisk were announced from the Oval Office as a political event. They were also, structurally, the first time a presidential administration has negotiated drug prices directly with manufacturers outside of the IRA’s statutory negotiation framework for Part D. The IRA negotiation process, which produced maximum fair prices for a first cohort of ten drugs effective in 2026, operates under statutory authority with explicit procedural requirements, excise tax enforcement, and manufacturer litigation rights. The BALANCE negotiation operates under CMMI demonstration authority with none of those statutory guardrails. The manufacturers participated voluntarily. The pricing terms were agreed bilaterally. The enforcement mechanism is the participation agreement itself, not a statutory penalty.

This distinction matters for what it reveals about the administration’s theory of pharmaceutical pricing. The IRA framework is legislative, mandatory, and slow: two years from selection to price implementation, with judicial review available. The BALANCE framework is administrative, voluntary, and fast: seven weeks from Oval Office announcement to formal model launch. If BALANCE demonstrates that negotiated pricing at scale can work through CMMI authority, it establishes a template for future administrations to negotiate drug prices outside the IRA’s statutory framework for any therapeutic category where manufacturer participation can be secured.

The manufacturers’ willingness to participate reflects a calculation that the political and market risk of declining exceeded the revenue impact of the price concession. GLP-1 medications are the highest-revenue pharmaceutical category in the world. Novo Nordisk and Eli Lilly together generate more than $60 billion annually from semaglutide and tirzepatide products. Opening Medicare’s 67 million beneficiaries as a covered market for the weight management indication, even at $245 per month rather than $1,000 or more, represents volume expansion that could offset per-unit revenue reduction. The manufacturers also received concessions: Eli Lilly secured a three-year exemption from the tariffs the administration has threatened to impose on imported branded pharmaceuticals, and both companies received commitments that their drugs would be the exclusive covered products in the demonstration’s initial phase.

The Lifestyle Support Requirement

#

Every BALANCE beneficiary receiving a GLP-1 for weight management must also receive access to a manufacturer-provided lifestyle support program at no cost. This requirement is where the model’s name becomes operational: “Better Approaches to Lifestyle and Nutrition for Comprehensive hEalth.” The lifestyle supports must be evidence-based and designed to help beneficiaries incorporate a reduced-calorie, nutrient-dense diet and increased physical activity into daily living, consistent with the FDA labeling that recommends GLP-1s be used alongside appropriate lifestyle modifications.

The RFA specifies that lifestyle supports must be delivered on a recurrent basis, enabling beneficiaries to log weight, review goals, and engage with the program regularly. Manufacturers proposed their lifestyle support offerings as part of the application process, and CMS expects to revisit requirements annually based on model performance.

The lifestyle support requirement serves two functions. The first is clinical: clinical trial data for both semaglutide and tirzepatide showed weight loss in the context of lifestyle modification, and the FDA labeling reflects that context. A model that covers the drug without the accompanying intervention is testing something different from what the evidence supports. The second function is political: the MAHA (Making America Healthy Again) framing that pervades the administration’s health policy agenda requires visible pairing of pharmacological intervention with behavioral health improvement. A model that simply covered GLP-1s at negotiated prices without lifestyle requirements would be, in the administration’s framing, incomplete.

The question is whether manufacturer-provided lifestyle programs are the right delivery vehicle. Pharmaceutical companies are not behavior change organizations. Their core competency is drug development, manufacturing, and distribution, not sustained nutrition counseling or physical activity programming. The RFA allows manufacturers to contract with third parties for program delivery, and most will. Whether a manufacturer-funded digital wellness program can achieve meaningful behavior change in a Medicare population with high rates of functional limitation, food insecurity, digital illiteracy, and social isolation is an empirical question the model is designed to test. It is also the question most likely to determine whether BALANCE’s weight loss outcomes prove durable or temporary.

The MAHA Intersection

#

BALANCE does not exist in isolation within the 2025 CMMI portfolio. It is the pharmaceutical arm of a three-model chronic disease prevention platform that also includes ACCESS and MAHA ELEVATE.

ACCESS, the 10-year voluntary model for chronic condition management via technology, includes hypertension, obesity, prediabetes, diabetes, and chronic kidney disease among its eligible conditions. ACCESS enables digital health companies and Part B providers to deliver technology-supported care management for Medicare FFS beneficiaries with these conditions. A beneficiary who enters ACCESS for prediabetes management and subsequently meets BALANCE eligibility criteria has a pathway from digital health intervention to pharmacological treatment within the same CMMI policy architecture.

MAHA ELEVATE, the lifestyle medicine model launching September 2026, focuses on nutrition, physical activity, sleep, stress management, harmful substance avoidance, and social connection. It is the most explicitly MAHA-aligned model in the portfolio. Its incubation-phase design means it will operate initially at small scale, testing whether functional medicine approaches can generate measurable health improvements and cost reductions. For beneficiaries who start on a GLP-1 through BALANCE and achieve sufficient weight loss to reduce or discontinue the medication, MAHA ELEVATE’s lifestyle medicine framework offers a potential maintenance pathway.

The three models together represent a theory of chronic disease management that moves from technology-enabled monitoring (ACCESS) to pharmacological intervention (BALANCE) to sustained lifestyle medicine (MAHA ELEVATE). Whether that theory holds depends on whether the models’ participant populations actually overlap, whether the timelines align in practice, and whether CMS has the implementation capacity to operate all three simultaneously alongside WISeR, AHEAD, LEAD, and the rest of the 2025 CMMI portfolio.

Impact on Part D Plan Design and Formulary Strategy

#

For Part D plan sponsors, BALANCE creates a new strategic decision with bid-cycle implications. Plans must decide whether to participate in BALANCE for their 2027 plan year, a decision that must be made during the bid development process in spring 2026. Participation means incorporating the negotiated cost-sharing limits into the plan’s benefit design and accepting the utilization and cost risk associated with a new covered therapeutic category. Non-participation means offering a plan that does not cover GLP-1s for weight management, which may be a competitive disadvantage in markets where beneficiaries have experienced the Bridge and expect continued access.

The actuarial uncertainty is real. CBO estimated that full statutory coverage of weight loss drugs would cost $35.5 billion over nine years, but that estimate assumed broader eligibility criteria and higher per-unit prices than BALANCE provides. The American Action Forum estimated that every one million new Medicare GLP-1 users under the negotiated pricing would cost approximately $1.74 billion annually in federal Part D outlays. CMS’s own projection that approximately 10 percent of beneficiaries will qualify narrows the eligible pool, but adherence rates, discontinuation patterns, and downstream medical cost offsets are all uncertain at this scale.

Part D plans participating in BALANCE will face formulary design questions they have not previously encountered for this indication. Tier placement, step therapy requirements, quantity limits, and the interaction between BALANCE-negotiated cost-sharing and the plan’s standard benefit structure all require actuarial modeling with limited historical data. The $2,100 annual Part D out-of-pocket cap, fully effective in 2026, creates a ceiling on beneficiary exposure but does not limit plan liability. Plans in the catastrophic phase bear 20 percent of costs, with the federal reinsurance subsidy covering 60 percent and manufacturers covering 20 percent under the IRA’s manufacturer discount program.

The interaction between BALANCE and the IRA drug negotiation process adds further complexity. Ozempic, Rybelsus, and Wegovy were selected for the second round of IRA price negotiations, with maximum fair prices to take effect in 2027. If a drug is both a BALANCE model drug and an IRA-selected drug, the pricing interactions between the negotiated MFP and the BALANCE net price will need to be resolved. CMS has not provided detailed guidance on how these two pricing frameworks will coexist for the same molecule.

Why This Is Also a Dual Eligible Story

#

The dual eligible population, the 12 million Americans enrolled in both Medicare and Medicaid, sits at the intersection of BALANCE’s two coverage pathways. A dual eligible beneficiary enrolled in a D-SNP with Part D coverage could potentially access GLP-1s through both the Medicaid component (if their state participates in BALANCE) and the Medicare Part D component (if their D-SNP participates in BALANCE). The coordination of benefits, cost-sharing, and the interaction between Medicaid supplemental rebate agreements and Part D plan design creates administrative complexity that CMS has not yet fully addressed.

For dual eligible beneficiaries specifically, BALANCE’s weight management coverage addresses a long-standing inequity. Obesity prevalence is significantly higher among Medicaid populations than among commercially insured or Medicare-only populations. Dual eligibles have the highest rates of multiple chronic conditions, including the metabolic syndrome components, diabetes, hypertension, cardiovascular disease, that GLP-1 medications treat across multiple indications. Many dual eligibles currently receive GLP-1s for diabetes through Part D but cannot access the same medications for weight management. BALANCE closes that gap for those enrolled in participating plans and residing in participating states.

The low-income subsidy interaction with the Bridge is a specific concern that KFF and other analysts have identified. During the July through December 2026 Bridge period, low-income beneficiaries who qualify for Part D’s Extra Help program cannot apply those subsidies to GLP-1 prescriptions filled under the Bridge, because the Bridge operates outside the Part D benefit. For beneficiaries accustomed to $0 or near-$0 cost-sharing on their Part D medications, the Bridge’s $50 copayment may be a meaningful barrier. Whether BALANCE’s full 2027 launch resolves this depends on how participating plans structure cost-sharing for LIS-eligible beneficiaries within the negotiated limits.

The Cost Neutrality Question

#

BALANCE’s viability as a CMMI model depends on demonstrating that it preserves or enhances quality of care while reducing or maintaining program expenditures. The model’s theory of savings rests on a premise that has intuitive appeal but limited long-term Medicare-specific evidence: that treating obesity pharmacologically will reduce downstream spending on the chronic diseases obesity causes.

The evidence base is substantial for short-term clinical outcomes. The STEP and SURMOUNT clinical trial programs demonstrated average weight loss of 15 to 22 percent of body weight with semaglutide and tirzepatide, along with improvements in blood pressure, lipid profiles, blood glucose, and cardiovascular risk markers. Wegovy’s SELECT trial demonstrated a 20 percent reduction in major adverse cardiovascular events in adults with cardiovascular disease and obesity or overweight. These are among the most clinically significant outcomes in modern pharmacotherapy.

The evidence is more limited for the proposition that Medicare-specific medical cost offsets will materialize within the model’s testing period. CBO’s analysis found that drug costs would exceed medical spending reductions over the 10-year budget window, producing a net federal spending increase of $35.5 billion even with assumed government-negotiated prices. The USC Schaeffer Center estimated substantially larger medical cost offsets, $176 billion to $245 billion over 10 years, but did not net those against drug costs and relied on assumptions about sustained adherence and population-level metabolic improvement that have not been validated at Medicare scale.

The discontinuation problem is the model’s most significant clinical and fiscal risk. GLP-1 medications for weight management are effective while taken. Weight regain after discontinuation is substantial and well-documented. If a significant share of BALANCE beneficiaries initiate treatment, achieve weight loss, discontinue, regain weight, and reinitiate, the cost trajectory looks very different from one in which beneficiaries achieve durable weight loss and avoid or delay chronic disease progression. The lifestyle support requirement is designed to address this, but whether a manufacturer-funded wellness program can prevent the weight regain that occurs when the pharmacological signal is removed is unproven at scale.

CMS will evaluate the model rigorously. CMMI’s evaluation framework will track costs, adherence, clinical outcomes, and beneficiary experience. But the evaluation timeline creates a political economy problem: if BALANCE is popular with beneficiaries and politically visible, discontinuing it because the cost-neutrality evidence is unfavorable becomes politically difficult regardless of what the evaluation shows. The model’s five-year testing period, running through December 2031, spans at least one and possibly two presidential transitions. The beneficiaries who begin GLP-1 treatment under BALANCE will expect continued access. The political constituency that forms around a popular benefit is the most durable force in Medicare policy.

What Comes Next

#

BALANCE occupies an unusual position in the CMMI portfolio. It is voluntary, which typically means CBO will score it conservatively for federal savings. It depends on manufacturer participation that could change if pricing terms become unfavorable relative to commercial market alternatives. It relies on state Medicaid participation that will be uneven, particularly in states with restrictive fiscal postures or philosophical opposition to weight management coverage. And it operates under demonstration authority that expires, requiring either congressional action or model expansion certification to become permanent.

The most consequential near-term questions are operational. How many Part D plans will participate for 2027, and will they include the largest national carriers? How many states will join the Medicaid component, and will the states with the highest obesity prevalence and dual eligible populations be among them? Will the Bridge-to-BALANCE transition create a coverage cliff for beneficiaries who start treatment in July 2026 and do not switch plans? And can CMS operate the Bridge’s central processor, the BALANCE negotiations, and the concurrent rulemaking for the CY 2027 Part D benefit simultaneously without implementation failures?

The longer-term question is whether BALANCE establishes a durable pathway for Medicare coverage of weight management or remains a time-limited demonstration that ends when the political alignment that created it changes. The statutory exclusion remains on the books. The Treat and Reduce Obesity Act, which would remove it legislatively, has been introduced in multiple congressional sessions without enactment. If BALANCE succeeds clinically but fails the cost-neutrality test, Medicare will have demonstrated that GLP-1s improve health outcomes for seniors while also demonstrating that covering them increases net federal spending. That is the result the statute’s exclusion was designed to prevent. Whether it is a result the program can sustain is a question BALANCE will answer, but not quickly.

Related Reading

#

MCR-04_09 Part D in 2026-2027: Drug Negotiation, Formulary Disruption, and the BALANCE Bridge MCR-10_01 The LIS Landscape: Extra Help, Medicare Savings Programs, and the Low-Income Non-Dual Population MCR-04_12 The IRA Drug Negotiation Process: First Cohort MFPs, Manufacturer Litigation, and What Comes Next