Who Pays For The Comparator? Insurance Cost-Sharing And The Wall That Got Taller

By John O'Brien

Part 4 of a five-part series: “Beyond Bulk — Rethinking Comparator Supply for Modern Clinical Trials”

The first three articles in this series stayed on solid ground: a documented problem, a model that can be stood up in weeks, and economics I could show my work on. The last two are different, and I want to say so up front. Whether a patient’s health insurance should help pay for a comparator dispensed in a clinical trial is less a supply chain question than a financial, legal, and ethical one, and much of what follows is argument rather than evidence. Where the ground is soft, I will say so; where the law gets thick, I will slow down and translate. But the pull model forces the question into the open — once a comparator moves through the same rails as any other prescription, someone in finance will ask why the sponsor is paying full price for a drug the insurer would have covered anyway — and the answer has changed since 2022. This article is about the wall that question runs into. The next is about a door in it that I did not know was there.

The Question The Pull Model Makes Askable

In a push model the question never arises: the sponsor buys, labels, ships, and dispenses the drug as clinical supply, and there is no claim to adjudicate. In the pull model from part 2, the comparator is a prescription filled against a study card that runs on the same back end as an ordinary insurance claim — so it can be positioned as a secondary payer. The patient’s plan pays first, the study card absorbs the rest, the patient pays nothing, and the sponsor’s cost falls by whatever the plan paid. In my 2022 modeling, a conservative assumption — one-quarter of U.S. subjects privately insured, plans covering roughly 80% — produced about a 30% cost avoidance on that subset, or $650,000 to $1.3 million over two years against a $6.5 million U.S. comparator spend.¹

The legal footing is real. Section 2709 of the Public Health Service Act, added by the Affordable Care Act and in force since 2014, requires most private plans to cover the routine patient costs of people in approved clinical trials for life-threatening conditions.² Medicare has covered those costs since 2000,³ Medicaid since January 2022.⁴ Research sites already run the machinery: the coverage analysis, or billing grid, that assigns every protocol item to sponsor, payer, or patient before the study opens.⁵

Three points of precision, because this is where supply chain people get it wrong. First, “standard of care” is not a payer concept; “routine cost” is, and the test is whether the patient would have received the item anyway, outside the trial. A protocol-required comparator at its labeled dose and indication generally passes. Second — and this trips up oncology teams constantly — a comparator given in combination with the investigational product usually does not. Bendamustine plus rituximab as the control arm in chronic lymphocytic leukemia is standard of care; the same pair plus a new investigational agent is an investigational regimen, and insurers treat the whole thing that way. The lever exists only for the pure control arm. Third, the coverage analysis belongs to the site’s billing compliance function, not the sponsor.⁵ You do not decide your comparator is a routine cost; the site’s analyst does.

Where It Works: Privately Insured Patients

The best published evidence that Section 2709 changed payer behavior comes from The University of Texas MD Anderson Cancer Center (MD Anderson)’s early-phase program. Kehl and colleagues analyzed 2,404 insurance-clearance referrals before and after the requirement took effect: clearance rates for privately insured patients rose sharply — an odds ratio of 4.72 — while Medicare and Medicaid rates, already high, did not move.⁶ It is one institution, it measures clearance to enroll rather than payment for a drug, and it is nine years old. I searched for anything newer on insurance payment for comparators specifically and found nothing — a familiar finding in this series.

Concretely, with the comparator from part 3: at roughly $9,900 per 28-day cycle, a full-dose ibrutinib patient costs the sponsor about $129,000 a year.⁷ If that patient’s commercial plan covers 80%, the study card’s share falls to roughly $26,000. Across a cohort, that is not a rounding error.

It is also not free money, and any pharmacy card vendor will walk a sponsor through the frictions when the feature is requested. Plans deny claims, especially when the trial site is out of network — Section 2709’s network protections are narrower than most assume, and older “grandfathered” plans are exempt.² Plans impose prior authorization on a comparator whose timing is set by a protocol, not a formulary. To a pharmacy benefit manager a study card looks exactly like a manufacturer co-pay card, and a growing number of plans run “accumulator” programs that refuse to count such payments toward the deductible — the fight behind the HELP Copays Act now before Congress.⁸ Every denial becomes a reconciliation item, muddying the spend visibility I praised in part 3. And there is a fairness question worth saying out loud: the sponsor chose the comparator, dictates its dose and duration, and profits from the trial, so the plan is subsidizing a commercial program it had no voice in. The routine-cost logic holds when the comparator is exactly what the patient would have received anyway, and weakens as the protocol departs from it — a judgment for the coverage analyst and the institutional review board, and one the consent form must disclose plainly.

Where It Breaks: The Wall Got Taller

Government-insured patients are where the model meets the federal anti-kickback statute, so a plain-language primer. The statute is a criminal law that says, in effect: you may not give anyone something of value to move them toward buying something Medicare or Medicaid will pay for. Two words carry the weight — “remuneration,” anything of value, which a card that pays your co-pay squarely is; and “induce,” to move someone toward a decision. “Safe harbors” are arrangements the government has pre-cleared. An “advisory opinion” is when you describe a proposed arrangement to the U.S. Department of Health and Human Services (HHS) Office of Inspector General (OIG) and it tells you, in advance, whether it would pursue you.

In 2022, a study team I partnered with — running a large Phase 3 with an expensive comparator, and asked by senior leadership to evaluate the pull model — wanted to do for Medicare patients what we planned for privately insured ones: let Part D pay first and cover the rest with the study card. OIG Advisory Opinion 20-05, from 2020, had already analyzed a close cousin: a manufacturer’s card covering Medicare beneficiaries’ Part D cost-sharing for its own drug. OIG found the card was remuneration to induce a federally paid purchase, that no safe harbor applied, that inducement need only be one purpose of the payment, and that it would insulate beneficiaries from price and skew prescribing.⁹ There was no clinical trial exception. The team’s choices for Medicare, Medicaid, and TRICARE patients were to let them pay out of pocket, or to have the sponsor pay everything. It paid everything.¹

I wrote in part 1 that this terrain had grown harder since then. Here is the sequence. The Second Circuit went first, in 2022, upholding OIG’s rejection of a Pfizer co-pay program for Medicare patients.¹⁰ In January 2025 the Fourth Circuit followed in Pharmaceutical Coalition for Patient Access, upholding OIG’s rejection of a manufacturer-funded oncology assistance program: “any remuneration” means what it says, no corrupt intent required, and charitable framing does not help when the help is tied to buying the funders’ drugs.¹¹ Then, in September 2026, the D.C. Circuit decided Vertex Pharmaceuticals v. HHS, holding that a fertility-preservation program for Medicare and Medicaid patients receiving Vertex’s gene therapy would violate the statute — the same broad reading of the same two words.¹² Circuit courts sit one step below the Supreme Court; when three read a law the same way, that is the working law of the land. The Sixth Circuit reads it more narrowly,¹³ so the Supreme Court could eventually weigh in — but no sponsor can plan a trial on that.

At the pharmacy counter, this splits into two paths with very different risk.

Path A — the study card is the primary payer for government-insured patients. Medicare never sees a claim for the drug. This is what the study team chose, and it remains the lower-risk path — not risk-free, since any sponsor is subject to the statute once Medicare covers routine trial costs¹⁴ and a free comparator is a thing of value — but it is the posture every push-model trial already occupies. I know of no enforcement theory that has treated protocol-required, sponsor-supplied comparator as a kickback, and the industry has run trials this way for decades.

Path B — Medicare is primary and the sponsor covers the cost-sharing. This is 20-05 territory, and the Inflation Reduction Act has made it more tempting, not less. Since 2025, Part D out-of-pocket spending has been capped — $2,100 in 2026.¹⁵ On $129,000 a year of ibrutinib, Medicare-primary would leave the sponsor covering at most $2,100: a 98% reduction in comparator cost, and precisely the problem. Nearly the whole cost of the sponsor’s protocol lands on a federal program, and the statute does not care that the sponsor’s payment is small. Three circuits now agree that “we are only covering the co-pay” describes remuneration to induce.

What Supply Teams Can Do Now

None of the above is settled, and the next article will argue that it is less settled than it looks. But three things are true today.

• Know your payer mix before you count the savings. Kehl’s cohort was 41% government-insured, and most oncology indications skew older than that.⁶ The private-insurance lever is only as big as the privately insured share of your U.S. enrollment; in a Medicare-heavy population it may be worth very little.
• Get the coverage-analysis view of the comparator early. Ask the lead site’s billing compliance office whether the comparator, as your protocol uses it, is a routine cost or a research cost. Combination with the investigational product usually makes it a research cost. If it is research, there is no lever, and the question is closed before it opens.
• Treat cost-sharing as a legal decision, not a platform setting. Any card vendor can switch secondary adjudication on. Before contracting, ask how the vendor handles government-insured patients, accumulator plans, and a denial at the counter — and route the answers through legal and compliance at feasibility, not after launch.

Looking Ahead

Everything in this article points one way: for privately insured patients the lever exists and is worth something; for government-insured patients the wall has only grown taller, and the two-tier structure that results — insurance-first for some patients, sponsor-pays-all for others — raises a fairness question no one has answered. That is where my 2022 research stopped, and it is where most sponsors stop today.

But in the years since, OIG did something I did not expect: it looked at a sponsor paying Medicare participants’ cost-sharing inside a clinical trial, said the arrangement implicated the statute — and declined to pursue it. And in the same Vertex decision that added a third circuit to the wall, the court took away OIG’s ability to sit on advisory opinion requests indefinitely. The final article in this series is about what those two developments open up, what they do not, and why — after four years of fighting for this lever — I have come around to the view that the model’s real gift lies somewhere else.

Acknowledgments. Special thanks to Toni-Joy Motomura and Brian Horan for their review and editorial input across this series, and to Scott Bailey, Shawn Fletcher, Michelle Novak, and Ken Getz, whose mentorship shaped the research it draws on. Nothing in this article is legal advice; sponsors should consult counsel before structuring any insurance adjudication arrangement.

References

1. O’Brien J. Improving the Comparator Sourcing Paradigm in Clinical Trials. Master’s Capstone/CCP. Brown University; 2022.
2. Public Health Service Act §2709, 42 U.S.C. §300gg-8 (coverage for individuals participating in approved clinical trials), as added by the Patient Protection and Affordable Care Act; effective for plan years beginning on or after January 1, 2014.
3. Centers for Medicare & Medicaid Services. National Coverage Determination 310.1: Routine Costs in Clinical Trials (Clinical Trial Policy). Effective September 19, 2000; revised July 9, 2007.
4. Consolidated Appropriations Act, 2021, Pub. L. 116-260, Div. CC, §210 — Medicaid coverage of routine patient costs in qualifying clinical trials; effective January 1, 2022.
5. Institutional research billing compliance guidance on Medicare coverage analysis, including the principle that sponsor-covered items must be covered consistently for all participants and that billing designations may not be directed by the sponsor — e.g., UC San Diego Health, Office of Coverage Analysis Administration FAQ (updated August 19, 2025); University of Arizona Health Sciences, Coverage Analysis; Weill Cornell Medicine, Research Billing Compliance Handbook.
6. Kehl KL, Fullmer CP, Fu S, et al. Insurance clearance for early-phase oncology clinical trials following the Affordable Care Act. Clin Cancer Res. 2017;23(15):4155-4162.
7. O’Brien J. The Fifth Model: Where It Works, Where It Doesn’t, And The Economics Behind It. Clinical Supply Leader. July 21, 2026 (ibrutinib 420 mg tablet, ~$9,900 per 28-day supply; annual figure assumes 13 cycles).
8. Help Ensure Lower Patient (HELP) Copays Act, S. 864, 119th Cong. (introduced March 7, 2025).
9. U.S. Department of Health and Human Services, Office of Inspector General. Advisory Opinion No. 20-05 (September 2020).
10. Pfizer Inc. v. U.S. Department of Health and Human Services, 42 F.4th 67 (2d Cir. 2022).
11. Pharmaceutical Coalition for Patient Access v. United States, 126 F.4th 947 (4th Cir. 2025); see also Congressional Research Service, Legal Sidebar LSB11298, “Legal Challenge to Patient Assistance Programs Puts Anti-Kickback Statute in the Spotlight” (May 1, 2025).
12. Vertex Pharmaceuticals Inc. v. United States Department of Health and Human Services, No. 25-5133 (D.C. Cir. Sept. 4, 2026), affirming in part and reversing in part; see also OIG Advisory Opinion No. 24-06 (July 2024).
13. United States ex rel. Martin v. Hathaway, 63 F.4th 1043 (6th Cir. 2023).
14. Sidley Austin LLP. Clinical Trial Growing Pains: Compliance Challenges and Solutions for Development-Stage Biotechs. Global Life Sciences Update, March 4, 2025.
15. Centers for Medicare & Medicaid Services. Medicare Part D benefit parameters, contract year 2026 (annual out-of-pocket threshold $2,100), as amended by the Inflation Reduction Act of 2022.

About The Author:

John O’Brien is a clinical supply chain professional with experience managing comparator sourcing and distribution for global oncology clinical trials. His 2022 master’s capstone (CCP) project focused on re-engineering comparator supply through commercial pharmacy network integration and insurance cost-sharing, including collaboration with researchers at the Tufts Center for the Study of Drug Development, BeiGene (now BeOne Medicines), Myonex, TrialCard (now Valeris), and SupplyRx (now a part of Inceptua). This is the fourth article in a five-part series for Clinical Supply Leader.