Is CMS Killing Breakthrough MedTech? A View from the Other Side

This article is part one of a three-part interview series with Dr. Lee Fleisher, former Chief Medical Officer of CMS and now part of Rubrum Advising, on what the breakthrough device coverage problem looks like from CMS’s point of view.

I met Lee Fleisher at MedTech Malta. Someone who had read this article:

thought the two of us should talk, because as I was complaining about how CMS killed the MCIT program, and Lee was pretty happy they had removed it and replaced it with TCET. He served as Chief Medical Officer and Director of the Center for Clinical Standards and Quality (CCSQ) at CMS from July 2020 to July 2023, a tenure that spanned both the Trump and Biden administrations. We had a fascinating first conversation.

We have run into each other at a few conferences since then, and I kept bugging him to let me do an interview for The Device Files because getting CMS’s point of view on all of this is quite honestly fascinating to me!

Let’s start with a teaser of some of our first convo in Malta:

I was actually a bit worried when we first met because he hadn’t seen my article and it was really harsh on CMS. I told him not to read it before we spoke the next day, but of course when I ran into him the next morning he started off with, “So I read your article.”

Yikes.

But Lee is a really good sport, and honestly fun to debate and think big thoughts with.

The exchange that stuck with me the most from Malta was when Lee described a meeting with industry advocates that were pushing to keep the Medicare Coverage of Innovative Technology (MCIT) program. A lobbying group had made the case that without faster Medicare coverage, breakthrough medical device companies would go out of business. The argument did not move him. The argument that would have landed, in his view, was one centered on patients. Implied, which I don’t think Lee would phrase this way but it’s how it landed, was that big business is always thinking about the bottom line, but we at CMS keep patients at the heart of everything we do. (That sounds harsh, but I think a lot of industry people have the same thoughts in the other direction: “What the heck is wrong with CMS? We’re just trying to get better devices into the hands of patients and they’re stopping it by not thinking through the real world consequences!”)

My response, which honestly got a thoughtful look in return, was that if the company goes under, the patient never gets the technology. Patient access and business viability are not separable in a system where private capital has to fund the bridge from breakthrough designation to Medicare coverage. If the company can’t get funded, the pivotal trial does not run, the device does not get to market, and the patient gets nothing. I won’t rehash the prior article, but you saw the link above and can go back if you haven’t read it yet.

This type of thoughtful back and forth is exactly why I wanted to have this interview with Lee. So let’s dive into it further.

The first thing Lee corrected me on was the assumption that CMS has broad discretion over what it covers. The Social Security Act defines the categories of items and services that CMS is statutorily authorized to pay for and what’s excluded. For example, hearing aids are not a covered category. Screening tests for asymptomatic patients are largely not covered.

In order to get new types of things covered (we’re talking coverage areas, like breast cancer screening, not a specific test from one company), you generally need the U.S. Preventive Services Task Force (USPSTF). The U.S. Preventive Services Task Force is an entity that lives within HHS and can recommend screening coverage through its A or B ratings.

To me this sounds like a great vehicle to GSD (Get Shit Done), but it turns out they haven’t been called to meet under the current Health and Human Services Secretary. The carve-out mechanism that exists in statute is not actively producing decisions right now. It can also only do so much. It’s pretty limited to things related to preventative health and screenings.

I hear the explanation. It makes sense. But it also doesn’t. What does this even have to do with most of the BDD devices?

Why does any of this matter for the original article? Because the MCIT program, which the first Trump administration created by executive order to provide automatic Medicare coverage for breakthrough devices, did not address the issue of benefit categories.

When Congress created Medicare in 1965, it didn’t tell CMS to pay for “whatever doctors think is medically useful.” It wrote a list. The Social Security Act, in Sections 1861 and 1862, lays out specific buckets of things Medicare is allowed to cover. If something doesn’t fall into one of those buckets, CMS legally cannot pay for it, no matter how clinically useful or cost-effective it is. Those buckets are called benefit categories.

The main Medicare Part A and Part B benefit categories include things like inpatient hospital services, physician services, outpatient hospital services, diagnostic tests, durable medical equipment, prosthetics and orthotics, ambulance services, and a defined list of preventive services. Part D added prescription drugs in 2003. There are others, but those are the big ones.

Share

What’s important is what’s not on the list. Hearing aids are not a benefit category. Routine eye exams and eyeglasses are not. Routine dental care is not. Long-term custodial care is not. Most screening tests for asymptomatic patients are not, with specific exceptions Congress has carved out one at a time over the years (mammography in 1990, colorectal cancer screening in 1997, lung cancer screening through USPSTF in 2015, multi-cancer detection more recently). Each of those carve-outs required either a new act of Congress or a specific administrative mechanism Congress had pre-authorized.

So when CMS says it cannot cover something, sometimes what they mean is “we evaluated it and decided it isn’t reasonable and necessary,” which is a judgment call within their authority. And sometimes what they mean is “this isn’t a benefit category at all, we have no legal authority to pay for it regardless of how good it is.” Those are very different statements, but they are often voiced with the same shrug.

So, according to Lee, the law was structurally flawed in part because it did not address benefit categories. It said, in effect, pay for things that in a lot of cases Congress had not authorized CMS to pay for.

The way Lee described it, the political team in the first Trump administration insisted on inclusivity. They wanted the rule to cover everything, every breakthrough designation, regardless of whether the device was intended for a Medicare population or whether it fell into a Medicare benefit category. The problem is that congressional laws trump executive orders (no pun intended). The President cannot, by executive order, expand the statutory authority Congress has given a federal agency. So with their arms tied by the benefit category constraint, CMS lawyers could not figure out how to legally structure the rule so that it followed both the executive order and the existing law.

OK, that makes sense. And back in the first Trump admin this had all been done via executive order instead of a bill. But then why fight the current bill trying to work through Congress? If Congress passed it, that would fix the problem, right?

When Lee and I first met I thought he told me he was lobbying to try and kill the current bill, but in our follow up conversation he corrected me. He hasn’t been trying to kill it, he’s actually been on Capitol Hill working with various congressional offices to help shape it into something workable.

Very interesting. More on that in part 3.

Another thing I complained about in the How CMS is Killing Breakthrough Devices article was the time and cost of the clinical evidence burden imposed by CMS to get to meaningful coverage.

The argument is this:

The evidence required to get CMS coverage is high. It is usually more than one study, its often not the same evidence the FDA wanted, and they’re usually looking for published, peer-reviewed, RCT (Randomized Controlled Trial) studies.

For smaller MedTech companies without top-name KOLs at major academic medical centers, the entire process of getting published in top tier journals is its own robust effort on top of running the studies themselves. Clinical evidence can take years and cost millions of dollars to gather. A single pivotal trial can run $1 to 3 million and have a run time of a year or more. If CMS expects multiple trials of meaningful size to grant coverage on a wearable or De Novo Class II device, that is a major factor in whether that company gets investment in the first place. The amount of money and time it takes to get from here to there is often too much.

Once again, Lee made some pretty good points on how I may have misunderstood.

Statutorily, CMS coverage decisions require a clear, public, and transparent evidence base. CMS has interpreted that to mean published evidence, but the definition of published is much broader than founders typically think. Acceptable formats might include preprint servers (Lee specifically mentioned Research Square), in-press articles that have been accepted but not yet appeared in the journal, and open-access journals, which sometimes have higher acceptance rates. CMS will accept evidence on ClinicalTrials.gov as well, with caveats.

That changes the strategic picture for companies that have generated solid clinical data but cannot get into the most prestigious journals on a first submission. The publication requirement is not the wall it is sometimes presented as.

Lee also pushed back on the size and structure of the studies themselves. His view is that you do not always need a large, randomized, pharma-scale trial to get CMS to engage. A smaller study (50 to 100 patients), with a synthetic comparator built from claims data or electronic health record data with propensity matching, can do real work if the indication is narrowed to the population where the data shows benefit.

I still agree that things need to significantly change at CMS to help support innovative MedTech. Now, after conversations with Lee, I understand that some of the things that looked like obstruction could just be CMS showing constraint and respect for existing law. The issues outlined here are real and the explanation of why MCIT could not work in the form it was written makes sense.

And yet, when I sit with all of it, part of me still wonders whether the explanation is the whole story or whether it is the kind of reasoning an institution reaches for when it does not want to do something.

I love the fact that Lee is still involved in trying to shape what’s next – maybe I can visit Capitol Hill with him at some point to show how two sides of the table can work together to form a cohesive solution that works!

Next week we’ll take a deeper look at the TCET program that replaced MCIT, and the CMS thought process that went into its design.

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.