Blythe Karow · The Device Files: From Concept to Commercialization
This is part 3 of a three-part series on what the breakthrough device coverage problem looks like from inside CMS, in conversation with Dr. Lee Fleisher of Rubrum Advising. Read Part 1 here and Part 2 here.
My conversation with Lee was originally meant to be one article, but it just kept expanding as new information came to light, and THEN CMS and the FDA came out with two major announcements impacting coverage for Breakthrough Devices. So, of course, Lee and I had to jump back on Zoom for another chat so I could pick his brain on the updates.
Here’s the short version of the news:
On April 14, CMS announced that it was proposing to repeal the alternative pathway for the New Technology Add-on Payment (NTAP) program, which let breakthrough-designated devices skip two of the three standard NTAP eligibility criteria since 2020. Then, nine days later, CMS and FDA jointly announced the Regulatory Alignment for Predictable and Immediate Device (RAPID) coverage pathway, designed to dramatically shorten the timeline from FDA market authorization to Medicare national coverage, while also announcing that CMS would be pausing the TCET program we covered in Part 2.
If that all sounded super confusing, never fear, I will walk you through it now that Lee explained it to me!
A quick refresher on how NTAP works. When a hospital uses a new medical technology during an inpatient stay, the device often gets bundled into the standard payment for that diagnosis-related group (DRG). If the new technology costs significantly more than what the DRG pays, the hospital is less likely to adopt due to the price tag. NTAP is the add-on payment that bridges that gap for up to three years, during which time CMS gathers utilization data and then either reprices the DRG or lets the add-on expire.1
To qualify for NTAP, a technology has to meet three criteria:
It must be new (within three years of FDA authorization)
It must be costly enough that the DRG payment is inadequate, and
It must demonstrate substantial clinical improvement over existing technologies.
That third criterion, substantial clinical improvement, is the one that has historically tripped up most applicants. According to one analysis of traditional NTAP applications, 69 percent of denials were attributed to substantial clinical improvement alone, with another 13 percent denied for substantial clinical improvement combined with newness.2
In 2020, the first Trump administration created what’s called the alternative pathway. Under it, FDA-designated breakthrough devices were considered automatically not substantially similar to existing technologies and automatically meeting the substantial clinical improvement criterion. Companies still had to meet the cost criterion, but the substantial clinical improvement bar was waived on the strength of the breakthrough designation.
The numbers tell you the alternative pathway was used (albeit by a small number). CMS received 47 NTAP applications for FY 2027, with 32 of those filed under the alternative pathway and 15 under the traditional pathway. For FY 2020, the year immediately before the alternative pathway took effect, CMS received 18 total applications.3 So the alternative pathway nearly tripled NTAP application volume. A peer-reviewed analysis counted 22 breakthrough device alternative pathway approvals between FY 2021 and FY 2024, of which 9 were cardiovascular technologies.4 You might also notice the math on how many actually made it through, 22 BDDs in 3 years. So still, not a huge number but at least something.
Now CMS is proposing to repeal that alternative pathway. Beginning with FY 2028 NTAP applications, all applicants will need to satisfy all three eligibility criteria, including substantial clinical improvement.5 The repeal also extends to the OPPS device pass-through payment program on the outpatient side, with the same change taking effect for applications received on or after October 1, 2026.6
Lee’s explanation of what CMS is saying with this move came down to one word the CMS political leadership keeps using in his meetings: deinflationary. The argument from CMS is that too many devices that took the alternative pathway have not, in retrospect, demonstrated substantial clinical improvement over what was already available. CMS is no longer willing to take breakthrough designation as a proxy for the substantial clinical improvement criterion. Companies will have to show their work.
I can’t address this better than what was already said here (which is evidently an AI-generated quote from a blog, but why rewrite what works?):7
“CMS’s rationale, at least in this section of the proposed rule, is assertive but not empirical. CMS says it has ‘gained experience’ and has concerns about the limited evaluation process, but the discussion here does not present a visible dataset showing that the alternative pathway led to poor approvals, wasteful spending, inferior outcomes, or systematic misuse. That leaves opponents room to say CMS is replacing a targeted pro-innovation policy with a more restrictive one without demonstrating that the old policy failed.”
Exactly.
And I’ll remind everyone, again, that the WHOLE POINT of this Breakthrough Device coverage stuff is to try and get devices covered faster and give them a chance to prove themselves, and CMS just keeps trying to reset to their old ways.
As an aside, someone recently made a point in a LinkedIn conversation that “prescription drugs automatically have coverage but prescription devices do not.” YES! Thanks for breaking it down to its simplest components in a way I have never been able to articulate, Renee Gagnon. We bury medical device coverage behind 20 different types of insurance payments and wonder why it takes a device 5.7 years to get covered while pharma doesn’t seem to have the same issues – and have never stepped back and thought maybe we can simplify it this way?
I know what CMS would say. “That’s not how the system works.” And then launch into a very complicated explanation of all the different pathways to coverage.
But the big takeaway here, beyond my griping about it, is that if you were planning to go after the alternative NTAP payments you’re now going to need to rethink that strategy. You have until FY 2028 to adjust.
We covered earlier in this series how TCET is a sad, sad program that helps virtually no-one.8
I don’t really feel like rehashing all of the details but really quickly: It was designed to provide early coverage to a select number of Breakthrough Devices and only one device seems to have benefited from it since its launch a few years ago. So, I have no problem with shutting this down and trying something that might work better. Hopefully something that is also better funded so it can actually work with more companies to get their products through.
What’s this NEW program going to do? Well, under RAPID, CMS becomes part of the FDA’s pre-market conversation with the manufacturer earlier in the process. The agencies coordinate on what evidence the company needs to generate, the company runs an Investigational Device Exemption (IDE) study that enrolls Medicare beneficiaries and studies clinical outcomes both agencies agree on, and when FDA grants market authorization, CMS issues a proposed NCD on the same day. The 30-day public comment period runs, and the device can have national Medicare coverage as soon as two months after FDA authorization.9
This is just an announcement and the final guidance will be released later (hopefully soon). I can’t wait to see it because there are some areas I hope they will clarify:
The language in the announcement says the product in question needs to be “the subject of” an Investigational Device Exemption (IDE) study that “enrolls” Medicare beneficiaries. The translation here is that the clinical study used to get FDA authorization will be done via the IDE process and will include endpoints that CMS cares about – the Medicare beneficiaries (elderly). This sounds super fair, but it’s actually quite tricky. The thing is, most devices actually don’t do an IDE study because you only need to get the exemption if your device poses a significant risk to the patient. The vast majority of devices are actually deemed a NonSignificant Risk (NSR) and are directed to skip the IDE process and just work with an IRB (Institutional Review Board) for their trial oversight.
So, the real question is, what happens to the low-risk devices that typically don’t need an IDE study but still generate clinical evidence in a pivotal trial for the FDA? Can they now have IDE discussions to follow this path or is RAPID ruling out low risk devices? That seems counterintuitive to me to rule them out, but it would be in line with the whole “TCET is for PMAs” mentality.
The device must address an unmet medical need among Medicare beneficiaries.9 Um, are there ANY medical needs that you can’t prove have some type of treatment available to them? This is a much, much higher bar than the BDD designation of “providing improved outcomes” and it needs clear guidance. My gut says this is where CMS is going to just rule everything out.
Also, eligibility is even more narrow. To qualify for RAPID, a device must also:
Have FDA Breakthrough Device designation, so you’re already proving you provide better outcomes than the current standard of care
Running an IDE study for your pivotal trial
and
Class III devices, OK so all Class III are allowed now? That is an improvement on TCET maybe, but then we’re back to my old argument on how they’re not really the ones who need help anyway...
or
Class II devices (510k or De Novo) participating in FDA’s Total Product Life Cycle Advisory Program (TAP), so you’re limited to just the therapeutic areas that are currently part of the TAP program: cardiovascular, neurological and physical medicine, radiological health, ophthalmology, and orthopedics.10
This all sounds super limiting and possibly TCET 2.0, wherein no devices benefit from all of this rigmarole. According to a senior CMS official on the announcement press call, about 40 devices currently qualify with another 20 potentially qualifying.11 Is that this year? Or since BDD started? The FDA had granted 1,226 breakthrough device designations as of the end of 2025 so that math isn’t looking great to me.12
For established MedTech leadership and investors evaluating portfolio targets right now, things worth taking from the April announcements:
First, if your target’s reimbursement strategy depends on the NTAP alternative pathway, you have until FY 2028 to revisit it. Companies that have been counting on the substantial clinical improvement waiver to get their device through NTAP need to either generate the substantial clinical improvement evidence on their own, or restructure their model around a different reimbursement path.
Second, if your target is a Class III breakthrough or a Class II breakthrough in TAP, RAPID is a potential opportunity. Honestly I think the chances are slim that it will work for most products. But I have talked before about how much I appreciate the FDA TAP program and how I hope it continues to expand.13 So, if you have the chance to be in that, then definitely jump in and have those conversations with CMS to see if this program actually gets good traction.
Third, if you’re a PMA device, congrats! We maybe just made this slightly easier for you – that is if you can prove you address an unmet need in the Medicare population.
Fourth (where most products actually sit), if your target is outside the RAPID universe, the coverage strategy looks different than it has in the past. TCET is paused. The NTAP alternative is going away. The traditional NCD pathway still exists but is slow and not designed for the typical breakthrough company.
So, what’s left?
Lee made me feel better by breaking down what he thinks is the best strategy for handling CMS as a Breakthrough Device.
Lee’s view is that companies do not need a Local Coverage Determination (LCD) or National Coverage Determination (NCD) to start generating Medicare revenue. Most things Medicare pays for do not have a formal coverage policy. They get paid based on the specific clinical claim the device is making, in the specific patient population the company has evidence to support, when providers submit claims that match.
The strategic move Lee has been making with his clients is to narrow the indication to the population where the data clearly shows benefit, get a small clinical study done with that narrow indication, build coverage on that specific claim, and then expand from there. He told me he has been telling VCs that this is often a more practical path than waiting on a national coverage decision that may not come for years. “We have told many of our companies who have done very well, stay claim by claim, build your evidence, you can get paid during that time,” he said in our interview. “Claim by claim is payment.”
The trap, in Lee’s experience, is companies that take this path and then never bother to generate the evidence at all. They get paid claim by claim and then have nothing to show CMS when the question of broader coverage finally comes around. The companies that have done well with this approach have stayed disciplined about evidence generation even when they didn’t have to.
For established MedTech leadership evaluating a target that doesn’t fit RAPID, the diligence question becomes: has the company been smart about which clinical claims to pursue, which Medicare Administrative Contractors (MACs) to engage, and what evidence is actually in the pipeline. A portfolio company without an NCD is not necessarily a portfolio company without revenue. The evaluation question is whether the company has built a defensible claim-by-claim strategy, or whether it has been waiting for a coverage decision that may never come.
That is a more sophisticated diligence question than the binary “do they have coverage or not” filter. It is also more accurate to how the breakthrough device market actually works.
This series started in Malta last fall with me arguing with Lee about how CMS was killing breakthrough medical device innovation. It is going to take me a long time to fully change my mind on that, but Lee has made me understand the constraints CMS operates under in ways I had not appreciated before and made me hopeful that there are strategies to start generating impactful revenue at the early stages for MedTech startups.
I am going to keep watching. The proposed procedural notice for RAPID has not published yet, and that will tell us a lot more about how narrow this gets in practice. Lee and I have agreed to keep this conversation going, so expect more on this as it develops.
Thanks for sticking with this three-part series. Special thanks to Lee Fleisher and Elly Burke at Rubrum Advising for the back-and-forth that made these articles possible!
Blythe Karow is a strategic management consultant and founder of The Karow Advisory Group. She writes The Device Files, a Substack publication on MedTech strategy, market access, and commercialization.
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.