I have had two friends in the last week tell me stories about their health insurance experiences that make me angry about our healthcare system. As I write this, I’m sitting in a Senate building waiting to do a briefing for staff and I want to go office to office and ask – what are you doing about it? But that would be naïve… I’m just in a MOOD.
A Tier-able Situation. On Tuesday, Health Affairs published a Forefront article from the National Pharmaceutical Council that makes the point that benefit design is having a moment because the old pharmacy benefit manager model (PBM) model is looking a little tired.
Employers and patients are paying more. And PBMs are under growing scrutiny for formularies shaped more by rebate math than clinical value. That matters because PBMs decide which drugs are covered, what utilization management applies, and how much patients pay at the pharmacy counter. If rebates are tied to list prices, the system rewards higher-priced drugs with bigger rebates, even when lower-cost or clinically comparable options exist.
That’s where value-based insurance design (VBID) comes in. Long story short, high-value medicines and services should be easy to access with low or no cost sharing; lower-value care faces more scrutiny. The article points to examples like zero-dollar copays for diabetes medicines, preventive service coverage, and Medicare Advantage demonstrations.
The timing matters. In October 2025, Cigna’s Evernorth announced a rebate-free model that would pass drug discounts directly to patients at the pharmacy counter. This would potentially reduce brand drug costs by 30% for people in high-deductible plans. Federal reforms taking effect by 2029 will require PBMs to pass through rebates and prohibit cost-based compensation. That means there’s this moment in time to rethink how we do things.
Here’s the catch: employers have become reliant on rebate revenue to offset other health care costs. The model only works if employers are willing to trade rebate revenue for pricing transparency.
For manufacturers, the shift would mean arguments about access need to be tied to value, not just coverage. Fee-based PBMs and VBID principles make the negotiation cleaner, although not necessarily cheaper.
Way Plaque When. On Monday, STAT reported that Medicare spending on Leqembi and Kisunla is running at about 20% of what Centers for Medicare & Medicaid (CMS) actuaries projected. In 2024, Medicare spent $90 million on Leqembi and $5 million on Kisunla across roughly 9,200 patients. Through the first three quarters of 2025, total spending hit $213 million across 19,000 patients. Annualized, that’s about $280 million.
CMS originally projected Medicare would spend $3.5 billion on Leqembi alone in 2025. Even if you double the reported figures to account for Medicare Advantage, actual spending is still under 20% of forecast.
These drugs (Leqembi from Eisai and Biogen, Kisunla from Lilly) are complicated to use. IV infusions every few weeks, PET scans to confirm early-stage disease, serial MRIs to monitor for brain swelling or bleeding. Patients on blood thinners don’t qualify. The eligible population is narrow. And the clinical benefit is modest: the drugs slow cognitive decline but don’t halt progression or produce noticeable improvements clinicians can reliably observe at the individual level.
CMS now says spending is incorporated into standard baseline projections with no separate line item, signaling the agency no longer views these drugs as a budget risk.
For manufacturers, this is the operational reality of launching a drug with a strong mechanistic hypothesis (clear amyloid plaques equals treat Alzheimer’s) but weak real-world evidence of benefit. Surveillance requirements, administration burden, unclear benefit at the individual level, eligibility restrictions that keep the addressable population small.
Two things can be true: these drugs represent scientific progress in a devastating disease, and they’re not going to be blockbusters under current use paradigms. And they become case studies in how expensive innovation gets when clinical friction is high.
340B Spread Sheet. Last week, MinnPost published a deep dive on Minnesota’s 340B debate, centered on a state bill that would give hospitals enforcement power against manufacturers restricting contract pharmacy arrangements. The Minnesota Senate passed it in April with bipartisan support. It’s stalled in the House.
In 2024, Minnesota hospitals generated $1.3 billion in 340B revenue. Five hospitals captured over half of this revenue. M Health Fairview University of Minnesota Medical Center alone pulled in $335 million (26% of the state total) while providing less than $8 million in charity care (2% of its 340B haul). Hennepin County Medical Center, by contrast, provided $90 million in uncompensated care against roughly $100 million in 340B revenue.
Remember, hospitals buy drugs at 20% to 50%+ discounts but charge full price. The spread is revenue. Some hospitals maximize that spread with dozens of contract pharmacy arrangements, including pharmacies in wealthy suburbs. Small rural clinics like Scenic Rivers Health generate $200,000 to $300,000 annually from 340B. Fairview generates $335 million.
Since 2020, manufacturers have restricted contract pharmacy arrangements, arguing the spread has turned 340B into a profit center. The 8th Circuit ruled for Arkansas in 2024, holding federal 340B law doesn’t preempt state enforcement. But then a North Dakota federal judge struck down a similar law one day before Minnesota House Republicans met to discuss priorities, writing that hospitals and contract pharmacies are in coordinated collusion.
The takeaway? 340B is now a state-by-state enforcement fight. Manufacturers can restrict contract pharmacies in states without enforceable laws, but face penalties in states that pass Arkansas-style statutes. The strategic question is whether state-by-state restriction is sustainable or whether federal legislative reform becomes the better long-term play.
Near Zero is Hard to Discount. Late last week, KFF published an analysis of CMS’s new GENEROUS model, the voluntary Medicaid supplemental rebate program tied to international reference pricing. The White House projects $64.3 billion in Medicaid savings over ten years.
I’m not convinced.
In the model, CMS negotiates Most Favored Nation (MFN)-based supplemental rebates for participating manufacturers, calculated as the second-lowest net price across eight countries. States can opt in, select which drugs to cover, and adopt uniform coverage criteria negotiated by CMS. Seventeen manufacturers have signed MFN agreements to avoid tariffs.
Here’s the problem: Medicaid already gets massive rebates. Statutory rebates average 53% across all drugs, 62% for brands. Some brand drugs hit 77% due to Medicaid’s best price provision and inflation penalty. Drugs with rapid price growth can hit 100% or more. Medicaid already pays near-zero.
KFF ran the math. If US gross prices are 4.22 times international prices, an MFN approach delivers a 76% rebate. That’s higher than Medicaid’s current 62% average for brands, but not by much. And for drugs already carrying 77% or 100% rebates, MFN adds nothing. The model only generates savings on drugs where existing Medicaid rebates are small relative to the gap between US and international net prices.
Translation: GENEROUS will save money on newer drugs with limited competition and small rebates. It will save little or nothing on older drugs in crowded classes.
Also unknown: which drugs are in the model. If manufacturers can exclude high-rebate drugs and only include low-rebate drugs, the model saves very little.
GENEROUS is a margin management problem disguised as a transparency initiative. For drugs where Medicaid already pays near-zero, MFN pricing doesn’t move the revenue needle. For drugs with smaller rebates, MFN pricing could cut net Medicaid revenue. I remain skeptical this delivers $6.4 billion in annual savings.
Decoupled, not Defanged. On Tuesday, Healthcare Dive reported that Optum Rx is shifting to a fee-based PBM model, decoupling its revenue from drug list prices and prescription volume. Clients will pay monthly per-member fees instead of spread-based or rebate-linked compensation. The company says margins stay the same (3% to 5%) and the change isn’t mandatory for clients right away.
This follows similar moves by CVS Caremark and Express Scripts, which reached an FTC settlement earlier this year after being sued for allegedly steering patients toward higher-cost insulin to maximize rebate revenue. It also follows February legislation that prohibited PBMs from linking compensation to manufacturers’ list prices in Medicare Part D.
Optum is highlighting transparent flat fees, elimination of spread pricing, 100% rebate pass-through to clients by 2028. Here’s what Optum isn’t highlighting: how formulary decisions get made when the PBM’s fee isn’t tied to rebate size.
If Optum Rx gets paid the same per member regardless of which drugs are on formulary, what determines formulary placement?
Two possibilities. One: it becomes a pure cost-to-plan-sponsor negotiation. The rebate still matters but goes to the client rather than the PBM. Manufacturers still compete on net costs. Two: it becomes a utilization management negotiation, where manufacturers compete on how much prior auth and step therapy they’re willing to accept in exchange for formulary access.
If rebate competition continues but flows to plan sponsors, manufacturers face the same margin pressure as before. If the game shifts toward utilization management, manufacturers face a different tradeoff: access versus rebate.
For manufacturers, the shift removes one political argument: PBMs can no longer be accused of inflating list prices to maximize rebate-linked revenue. But it doesn’t remove the underlying tension. Fee-based PBMs make the negotiation cleaner. They don’t make it cheaper.
Reimbursement Fundamentals – Part D OOP Cap Works. Other Changes May Be Needed.
I made a mistake. I worked on a paper, it was released, and I forgot to talk about it. That’s how fast things are moving. But let’s get into it because the paper is great and the executive summary is chef’s kiss.
In mid-March, the MAPRx Coalition published a white paper arguing that the Inflation Reduction Act’s (IRA’s) Part D redesign is undermining affordability and access even as it delivers an annual out-of-pocket (OOP) cap. MAPRx supported the IRA reforms. They still do. The $2,000 annual OOP cap (increasing to $2,100 in 2026) is a huge deal for beneficiaries with high drug spending who previously faced unlimited exposure. The Department of Health and Human Services projected the cap would benefit 6.1 million beneficiaries in 2025, up from 1.5 million who reached catastrophic coverage in 2022. That’s meaningful financial protection.
Patient advocates understood achieving this would involve tradeoffs, including higher premiums. The Congressional Budget Office initially projected a 5% increase in per-enrollee costs. That seemed acceptable in exchange for removing unlimited OOP liability.
But the IRA didn’t just cap beneficiary costs. It fundamentally redesigned the benefit’s financial structure. Plans now bear 60% of drug costs in the catastrophic phase, compared to just 15% before the redesign. This altered incentives across plan bidding, benefit design, formulary construction, and utilization management. Plans have strong incentives to control drug spending before beneficiaries reach the annual OOP cap, when plan liability increases most sharply.
MAPRx’s argument: those incentives are reshaping plan behavior in ways that erode the OOP cap’s affordability gains. Early evidence shows higher premiums, narrower formularies, increased utilization management, fewer plan choices especially in standalone PDPs, and a shift from fixed copays to coinsurance that makes OOP costs less predictable earlier in the year. These effects hit beneficiaries who never reach the annual OOP cap, meaning savings at the catastrophic level coexist with increased burden elsewhere in the benefit.
The data tells the story. Between 2022 and 2026, standalone prescription drug plans (PDPs) reduced average branded drug coverage by 6%, from 658 drugs to 614. Medicare Advantage Prescription Drug (MA-PD) plans reduced coverage by 5%, from 743 to 705. While year-over-year formulary changes are expected due to patent expirations and new drug launches, the overall trend across both plan types is toward fewer branded products covered.
Nearly half of branded products covered by standalone PDPs and MA-PD plans in 2026 are subject to prior authorization. These requirements have increased steadily since 2020. For beneficiaries, the impact shows up as barriers to care. Prior authorization delays treatment initiation or continuation while documentation is reviewed. Quantity limits require repeated physician intervention to maintain therapy. Step therapy forces beneficiaries to cycle through plan-preferred treatments before accessing the medication originally prescribed.
And then there’s this: the standalone PDP market is contracting sharply. The number of PDPs offered nationally declined approximately 22% in 2026 compared to 2025. That’s an almost 50% decrease over two years. For millions of beneficiaries with Original Medicare and Medigap, this poses an acute access problem. Medigap policies don’t include prescription drug coverage, so standalone PDPs are their only pathway to Part D. MA-PD plans aren’t always realistic substitutes, particularly in rural regions where plan availability is limited or where longstanding clinicians may not be part of MA networks.
The declining availability of standalone PDPs also hits the program’s most vulnerable enrollees: LIS beneficiaries. LIS enrollees receive additional federal assistance to reduce premiums and cost-sharing and often enroll in benchmark PDPs, plans priced below a regional subsidy threshold, to receive premium-free coverage. As plan participation contracts, the number of benchmark options has collapsed. In 2026, some states like Florida and Texas have only one premium-free benchmark plan available. Fewer benchmark choices mean more beneficiaries are automatically enrolled in other plans by CMS because their prior plan is no longer premium-free. Fewer options means higher risk of medication disruption and less ability for beneficiaries to select plans that align with their drug needs.
Premiums are rising despite temporary protections. The IRA capped growth in the Part D base beneficiary premium at 6% annually through 2030. CMS launched the Part D Premium Stabilization Demonstration beginning in 2025 as a voluntary, temporary set of parameters to moderate beneficiary premium disruption while plans adjusted to the redesigned benefit. These steps have tamed the immediate impact, but are temporary stabilizers rather than structural solutions.
The shift from copays to coinsurance is also troubling. Historically, plans met actuarial equivalence requirements primarily through fixed-dollar copays for many drugs, particularly non-specialty branded products. Fixed copays provided beneficiaries with predictable costs at the pharmacy counter. Coinsurance requires beneficiaries to pay a percentage of a drug’s cost and was traditionally concentrated in specialty tiers. In recent years, plans have increasingly applied coinsurance across a broader range of drug tiers, including preferred and nonpreferred brand drugs that were previously subject to fixed copays. Under coinsurance, beneficiary cost-sharing rises in direct proportion to a drug’s price, increasing both the amount paid per fill and the variability of OOP costs over time.
Plans now have stronger incentives to manage when and how beneficiary spending accumulates over the course of the year. Shifting from fixed copays to coinsurance allows plans to shift a greater share of costs to beneficiaries earlier in the benefit year before the cap is reached and plan responsibility rises most sharply. Some non-LIS beneficiaries with high drug spending will benefit substantially from the annual OOP cap and experience lower total costs over the course of the year. Beneficiaries with moderate prescription drug spending may face higher and less predictable OOP costs without ever reaching the cap.
For manufacturers, the IRA redesign creates dual pressure: pricing pressure through negotiation, rebates, and potential MFN proposals, and access pressure through formulary exclusions, tier placement, and utilization management. Plans managing 60% catastrophic liability means they control utilization aggressively before beneficiaries reach the cap. That shows up as tighter formulary reviews, more aggressive prior authorization and step therapy even for protected class drugs that must be covered, tier placement pressure to shift brands to higher cost-sharing tiers, and reduced formulary stability as plans become more willing to drop drugs mid-contract or move them to non-preferred status.
Manufacturers can’t just manage rebates anymore. They must manage access. And access is harder to compete on when plans have structural incentives to restrict utilization.
MAPRx is clear about the core tension. The introduction of an annual OOP cap and the Medicare Prescription Payment Plan represents a meaningful advance in protecting Part D beneficiaries from catastrophic prescription drug costs. For years, millions of beneficiaries, particularly those with complex or chronic conditions, faced unlimited OOP exposure that could lead to financial hardship and nonadherence to essential therapies. Establishing a fixed annual cap provides an important and predictable financial safeguard.
At the same time, achieving this core affordability goal can’t come at the expense of access. Rising deductibles, greater reliance on coinsurance, narrower formularies, and expanded utilization management increase upfront and administrative barriers for beneficiaries who may never reach the annual cap. For these beneficiaries, affordability challenges shift rather than disappear.
The question is whether Congress intended this tradeoff or whether the benefit design didn’t account for how plans would behave when their financial exposure increased. If it’s the former, the tradeoff was explicit: catastrophic protection in exchange for tighter access controls earlier in the benefit. If it’s the latter, the policy design didn’t account for operational reality.
Either way, the OOP cap is real. The access barriers are also real.
The manufacturer response depends on whether you’re inside or outside the negotiation window, inside or outside protected classes, and whether your drugs are in therapeutic areas where plans have formulary flexibility. The IRA fixed catastrophic costs by shifting liability to plans. Plans are responding rationally by controlling utilization before beneficiaries hit the cap.
That’s not a bug, it is the design of the program.
