Newsletter · · Ashutosh Agarwal

Pill Penalty, MFN Rebates, and Medicare's GLP-1 Turn - Healthcare Policy: Drug Pricing, IRA & Managed Care - Week of July 11, 2026

Healthcare policy newsletter for the week of July 11, 2026. A thematic week: a Sanofi insider put hard numbers on the IRA 'pill penalty' curbing cancer research, policy veterans dissected how 'most favored nation' rebates would actually work, Medicare Advantage's coding wars escalated into an AI billing fight and an $80 billion overpayment debate, and GLP-1s crossed a real policy line as Medicare began covering them for weight loss on July 1.

Healthcare Policy: Drug Pricing, IRA & Managed Care

Week of July 11, 2026: Pill Penalty, MFN Rebates, and Medicare's GLP-1 Turn

Intro

This was a thematic week more than a headline week. No single blockbuster announcement, but underneath, the podcasts told a coherent story about where US healthcare policy is actually biting. On the drug side, industry insiders are starting to put hard numbers on how the Inflation Reduction Act (the 2022 law that lets Medicare negotiate drug prices) is quietly reshaping what companies choose to research, and they are nervously eyeing the next big idea, "most favored nation" pricing. On the insurance side, Medicare Advantage, the privately-run version of Medicare that now covers more than half of all seniors, is caught in a widening fight over how plans code patients to get paid more, and whether the whole payment system is being gamed. And running through everything: GLP-1s, the weight-loss and diabetes drugs (Ozempic, Wegovy, Zepbound, Mounjaro), which just crossed a real policy line as Medicare began covering them for weight loss on July 1.

Below is what the week's podcasts actually said, who said it, and why it matters, with a clear line drawn between people running these businesses and outside commentators watching them.

TL;DR

  • The "pill penalty" is showing up in the data. A Sanofi policy executive walked through new research claiming the IRA has already cut small-molecule (pill-form) cancer drug research sharply, a 27% drop in lead programs, because pills get negotiated by Medicare years sooner than injected biologic drugs. The fear now: "most favored nation" pricing does even more damage.
  • Medicare Advantage's coding wars went nuclear. One reporter described insurers and hospitals now using dueling AI tools, one side to add billing codes, the other to deny them, while a former Medicare chief said the plans are overpaid by roughly $80 billion a year. Star-rating bonuses (over $13 billion this year) are being challenged in court.
  • GLP-1s crossed a policy threshold. Medicare started covering weight-loss drugs on July 1 at about $50 a month out of pocket, a PBM executive described capping members' GLP-1 cost at $200 a month, and brand prices that were $1,500 have fallen toward $150, with one operator arguing "the age of unrestrained pharma margins is gone."

What's New

1. A Sanofi insider puts numbers on the "pill penalty," and warns it is already curbing cancer research. On Vital Health Podcast (July 9), Michael Penn, U.S. Head of Reimbursement and Public Policy at Sanofi (an operator/insider), laid out fresh research his company funded through Vital Transformation. The core mechanic: under the IRA, small-molecule drugs (ordinary pills) become eligible for Medicare price negotiation after nine years, while biologics (injected drugs made from living cells) get thirteen. That four-year gap, he argues, is enough to redirect where the industry spends its money. Penn framed it plainly: "we wanted to understand the impact of what we call the pill penalty or the disadvantaging of the small molecules in terms of drug selection for Medicare negotiation as compared to, you know, their biological brethren." He said the study, a balanced comparison of the four years before and after the IRA, found a 27% reduction in lead assets and a 35% reduction in follow-on research in small-molecule oncology (cancer). His blunt math on the economics: "that nine-year pill penalty is somewhere between 40% or 60% of the profit goes away." He added that 2025 was the second-worst year for the oncology pipeline since the IRA passed, even though it was supposed to be a recovery year, and that developing a drug still runs "north of $2 billion and north of 10 years." Why it matters: this is the industry's evidence-building campaign to convince Congress and the Congressional Budget Office they went too far. Whether or not you accept a Sanofi-funded study at face value, drugmakers are clearly reweighting pipelines away from pills, and that shapes what treatments exist a decade from now.

2. A veteran policy insider explains how "most favored nation" pricing would actually work, and why it could backfire on pharma. On RealPharma: Conversations with Pharma Pathfinders (July 10), Deborah Williams, a longtime health-policy insider (former staffer at the agency that ran Medicare and at the House Ways and Means Committee), demystified the mechanics that get buried in slogans. "Most favored nation" (MFN) means the US tries to pay no more than a basket of other rich countries. She described concrete legislative versions circulating, nicknamed "Guard" and "Globe," that don't set prices directly but force a rebate back to the government if a company's US price exceeds that international basket: roughly a 25% discount on drugs given in doctors' offices and 50% on pharmacy drugs. Her key warning is technical but important: because MFN would tie into "best price" rules, it would ripple into the 340B discount program and lower the reported "average sales price" that everything else keys off, so, she argued, "legislating on any of these components will make things worse for pharma and will basically destroy their objectives." She also doubts other countries will simply agree to pay more ("Recall us dropping aid shipments lately into Germany and Switzerland. And yet they pay less"), and noted about two-thirds of pharma revenue comes from the US anyway. On the legal odds: a version that compels companies is probably unconstitutional, but one routed through the IRS (like 1970s wage-and-price controls) could survive. Why it matters: MFN sounds simple on a bumper sticker; the plumbing is where it lives or dies, and insiders think a poorly-designed version hurts the very companies it targets in unpredictable ways.

3. Medicare Advantage's "AI billing war," with one company arguably on both sides. On the Becker's Healthcare Podcast (July 8), reporter Jakob Emerson of Becker's (an outside commentator) described a genuinely striking conflict. UnitedHealth's Optum sells AI coding tools to more than 100 health systems that help them document more billable diagnoses, while UnitedHealthcare, the insurance arm, then has to fight those same codes on the claims it receives. Emerson said he asked UnitedHealthcare's CEO how the company avoids its own Optum AI tools worsening costs; the CEO conceded it is "a driver of higher costs right now" but expects a "saturation moment." Emerson's co-host put it less diplomatically: it is "like selling arms to both Russia and Ukraine at the same time." He also flagged that the Medicare Advantage star-rating system, worth over $13 billion in bonus payments this year, is unraveling in court: after smaller insurer Clover Health won a case against Medicare in May 2026 over how its 2026 ratings were calculated, the rest of the industry piled in with lawsuits, and the government is recalculating. His pointed question: studies now suggest the star ratings "don't actually correlate with the quality of the plans" and have become "a big algorithmic gaming system." Why it matters: the tools meant to measure and control Medicare Advantage costs are themselves becoming the battleground, and the litigation is now big enough to move the largest insurers' bonus revenue.

4. Medicare quietly started paying for weight-loss drugs, and the 340B hospital program is next on the chopping block. On Health:Further (July 4), the hosts (investors and outside commentators) ran through a dense policy week. The headline change: starting July 1, certain Medicare members can get weight-loss drug coverage for the first time, at about $50 a month out of pocket, mostly with prior-authorization requirements. They expect "huge adoption" and a big near-term spending outlay, with hoped-for health savings later. They also detailed a proposed 2027 outpatient payment rule that would slash 340B drug payments (a program that lets certain hospitals buy drugs cheaply) to "average sales price minus 33.4%," down from the current average sales price plus 6%, and accelerate clawback of $7.8 billion in prior overpayments. The hospital lobby, they noted, called it a "continued assault on the 340B drug pricing program" within hours. And they flagged that 26 states are suing the Trump administration over new Medicaid work requirements. Why it matters: two things that were theoretical are now live or imminent: Medicare paying for GLP-1s (a spending event for the whole system) and a serious cut to hospital drug economics.

The Debate

The week's clearest argument was about Medicare Advantage: is it a better way to deliver care, or an $80-billion-a-year overpayment machine? Both sides showed up with real substance.

The case for (from an operator): On the Becker's Healthcare Podcast (July 9), Dr. Ken Cohen, Chief Medical Officer of Optum Health (a UnitedHealth operator/insider), presented peer-reviewed research (published in the American Journal of Managed Care) arguing that when doctors take full financial responsibility for a patient's total cost of care, outcomes improve, especially for the most vulnerable patients. Using the CDC's Social Vulnerability Index, his team found that full-risk Medicare Advantage patients are actually sicker and poorer than traditional Medicare patients (about 68% live in high-vulnerability areas versus roughly 50% in traditional Medicare), a direct rebuttal to critics who say the plans cherry-pick healthy people. And the results, he said, were largest for the neediest: full-risk MA showed 17% fewer hospital admissions among more socially vulnerable patients (versus 13% for the less vulnerable), 14% fewer emergency-driven admissions, and 30% lower use of high-risk medications. For dual-eligible patients (those on both Medicare and Medicaid, 19% of Medicare beneficiaries but 35% of the spending), admissions and readmissions ran 24% and 33% lower. His explanation: the extra money funds infrastructure traditional Medicare won't pay for, such as AI-driven risk stratification, embedded pharmacists, behavioral-health staff, and care in the home. Cohen argued that if every primary-care doctor practiced like his risk-experienced ones, savings to Medicare would run "in the tens of billions of dollars yearly."

The case against (from a former regulator): On TCN Talks (July 10), Dr. Don Berwick, former head of the Centers for Medicare and Medicaid Services (an authoritative outside critic), told the mirror-image story. Medicare Advantage, he said, was supposed to be "5% or 6% cheaper with better care." Instead, "they're 10% or more, 17% depending on the plan" more expensive, because "they just upcode everybody." His example: "50% of the patients in one large Medicare Advantage plan were coded as having peripheral vascular disease, which, ka-ching, you know, added nearly $3,000 a year to payment for them." He cited the nonpartisan Medicare Payment Advisory Commission's finding that MA is paid about $80 billion a year more than traditional Medicare for the same patients, and noted UnitedHealthcare's stock fell more than 50% at its worst last year as the government started clawing back on aggressive coding. His broader point: the administrative machinery of private insurance eats 15–22% of the bill, versus 1–3% for traditional Medicare.

Where they actually agree: notably, both men, the operator and the critic, accept that the extra Medicare Advantage dollars are real. They just disagree on whether that money buys better care (Cohen's infrastructure) or is largely captured through coding games and marketing (Berwick's upcoding). And both point to the same pivot: Medicare (through CMS) is tightening the coding rules that fed the profits. That is the swing factor to watch.

The Names in Play

Even in a thematic week, a few specific companies came up by name:

  • Sanofi: the source of the pill-penalty research and the industry's evidence campaign against the IRA and MFN (Vital Health Podcast, July 9).
  • UnitedHealth / Optum: at the center of the coding debate on both the "for" side (Optum Health's outcomes research) and the "against" side (upcoding critique; the Optum-sells-to-both-sides conflict). One commentator noted UnitedHealthcare stock fell 50%+ last year on coding pushback (Becker's, July 8 and 9; TCN Talks, July 10).
  • Centene: its head of pharmacy clinical programs detailed how the insurer manages GLP-1 coverage differently across Medicaid, Medicare Advantage, and the ACA marketplace (Becker's, July 10).
  • Cigna / Express Scripts: its trade-relations SVP described capping members' GLP-1 costs at $200 a month and guaranteeing cost "trend" to clients (Bright Spots in Healthcare, July 7; note this was a reposted episode).
  • AbbVie: cited by the RealPharma guest as the maker of Humira, the best-selling drug in pharma history, in a 340B example where a penny-priced drug still generates thousands in hospital spread (RealPharma, July 10).
  • Novo Nordisk and Eli Lilly: the GLP-1 leaders, repeatedly named as the companies whose brand prices fell toward $150 under pricing and regulatory pressure (Prof G Markets, July 5; Equity Mates, July 8).
  • Hims & Hers: the operator that undercut branded GLP-1s with $150 compounded versions (Prof G Markets, July 5).

Read-Throughs

PBMs (pharmacy middlemen): On Bright Spots in Healthcare (July 7, a repost), Harold Carter, SVP of Trade Relations at Express Scripts (part of Cigna; an operator/insider), described capping members' GLP-1 out-of-pocket cost at $200 a month, and increasingly guaranteeing clients a spending "trend," putting Express Scripts' own money at risk if costs run over. His sobering data point: "Over 50% of patients that are on a GLP-1 fall off therapy within the first year," which reframes the PBM job from "cut the unit price" to "keep patients on the drug long enough to get value."

Insurer GLP-1 exposure: On Becker's Healthcare Podcast (July 10), Angel Ballew, Head of Pharmacy Clinical Programs at Centene (an operator/insider), stressed there is "not really a one-size-fits-all approach," coverage bends to each line of business, with prior authorization as the most-used tool and step therapy used sparingly. She flagged the new Medicare Bridge Program that launched July 1, and said Centene is watching indications expand into "Alzheimer's disease and substance abuse disorder," meaning the utilization (and cost) base could keep growing.

Biosimilars and 340B: the RealPharma discussion (July 10) is a warning for anyone betting on biosimilars (cheaper copies of biologic drugs). Deborah Williams argued the 340B program actively distorts incentives, hospitals favor the higher-list-price product because the spread is bigger, which helps explain why biosimilars "have never really fulfilled the potential" many expected. If the proposed 340B cuts land, that math changes.

Ex-US / generics: On Prof G Markets (July 5), a Hims & Hers operator noted semaglutide (the ingredient in Ozempic/Wegovy) "just went generic in Canada this year" and will go generic in the US "in about four years," a clock ticking on branded GLP-1 pricing power. On Equity Mates Investing Podcast (July 8), the hosts noted Trump's April 2026 100% tariff on imported pharmaceuticals, avoidable by signing an MFN deal and committing to US manufacturing, and said 17 major drugmakers have signed such deals covering about 86% of the branded market, with Novo Nordisk hit when the policy was announced.

Hospitals: the Health:Further hosts (July 4) were bracing. Between the 340B cuts, expanded site-neutral payments (paying the same for a scan whether done in a hospital or a suburban clinic), and Medicaid changes from last year's budget law, they called 2027 shaping up to be "an incredibly difficult year to be a hospital operator," and sensed a policy "willingness to see it break."

GLP-1s as the systemic wildcard: the Prof G operator argued GLP-1s "will have more impact in global society over the next five years than AI," with downstream effects on diabetes, heart disease, kidney and liver disease, and even addiction, and that brand prices collapsing from $1,500 to $150 signals that "the age of unrestrained pharma margins is gone."

What Changed

Two things genuinely moved this week, both on July 1:

  • Medicare began covering weight-loss GLP-1s for certain members at roughly $50 a month out of pocket (mostly with prior authorization), the first time Medicare has paid for these drugs for weight loss specifically (Health:Further, July 4).
  • Centene's Medicare Bridge Program launched, part of the shifting coverage architecture for GLP-1s in Medicare Advantage (Becker's, July 10).

Everything else, the pill-penalty research, the MFN "Guard/Globe" proposals, the 340B cut, and the star-ratings litigation, remains in the argument-and-lawsuit stage, not yet settled policy.