What CMS's FY2027 IPPS Rule Actually Changed for Breakthrough Devices and When

Most coverage of the FY2027 Inpatient Prospective Payment System final rule has compressed a sequenced policy change into a single sentence: CMS killed the Breakthrough Device shortcut to Medicare add-on payment. That is directionally right and specifically wrong in two ways that matter if you are planning around it. This is a closer read of what the rule says, when it takes effect, and what it leaves standing.

Background

CMS released the FY2027 IPPS and Long-Term Care Hospital final rule (CMS-1849-F) on July 31, 2026. It was published in the Federal Register on August 4, 2026 and is effective October 1, 2026. The rule sets a 2.3% payment update for hospitals that report quality data and are meaningful EHR users, reflecting a 3.2% market basket increase reduced by a 0.9 percentage point productivity adjustment. CMS estimates the finalized IPPS changes increase hospital payments by roughly $2.1 billion in FY2027, and separately estimates that additional payments for cases involving new medical technologies rise by approximately $779 million.

The New Technology Add-on Payment provisions are where the change for device and software developers sits.

Prior Alternative Pathway

NTAP has existed since 2001 to keep DRG-based payment from discouraging hospitals from adopting costly new technology. Under the traditional pathway at 42 C.F.R. § 412.87(b), a technology qualifies if it satisfies three criteria:

1. Newness: not substantially similar to an existing technology, and CMS has not yet recalibrated the relevant MS-DRGs to reflect its cost.
2. Cost: charges exceed a threshold set as the lesser of 75% of the standardized amount, or 75% of one standard deviation beyond the geometric mean standardized charge for all cases in the applicable DRG.
3. Substantial clinical improvement: the technology substantially improves diagnosis or treatment relative to what was previously available.

In 2019, CMS added an alternative pathway for FDA-designated Breakthrough Devices. Here is the part that is routinely misreported: the alternative pathway did not waive all three criteria. It deemed Breakthrough Devices "new" and not substantially similar by virtue of their program participation, and it excused them from demonstrating substantial clinical improvement. The cost criterion still applied. So did the requirement to hold FDA marketing authorization before May 1 preceding the fiscal year in question.

In practice, that made the alternative pathway a reduction in evidentiary burden, not a bypass.

Now

Beginning with FY2028 applications, CMS eliminates the alternative NTAP pathways for Breakthrough Devices and for Qualified Infectious Disease Products. From that point, applicants must satisfy the standard eligibility criteria, including substantial clinical improvement, unless they qualify for a limited grandfathering transition.

Nothing in this changes FY2027. The rule was finalized in the same document that awarded FY2027 NTAPs under the old rules. CMS explained the change by pointing to concerns it developed with what it described as the limited evaluation process for alternative pathway applications, and its view that technologies receiving NTAP should represent a substantial clinical improvement for Medicare beneficiaries.

FY2027 Numbers

• CMS approved 19 new NTAP applications for FY2027: 3 under the traditional pathway, 16 under the alternative pathway. All 16 held Breakthrough Device designation. CMS considered no QIDP applications and none under the Limited Population Pathway for Antibacterial and Antifungal Drugs.
• CMS continued NTAP for 41 existing technologies.
• CMS discontinued NTAP for 13 technologies whose three-year anniversary of U.S. market entry falls before April 1, 2027 (or, for one technology first approved in FY2026, before October 1, 2026).
• Application volume tells its own story. CMS received 47 NTAP applications for FY2027: 32 alternative, 15 traditional. For FY2020, the last year before the alternative pathway took effect, it received 18 in total.

Eighty-four percent of this year's approvals came through the pathway being retired. That is the scale of what changes for FY2028 filers.

As a sector illustration, the American College of Cardiology reported that of 11 cardiovascular-related alternative pathway applications this cycle, seven were approved, three were withdrawn, and one was denied.

Grandfathering

Technologies already approved under the alternative pathway, including all 16 approved in this rule, remain eligible for continued NTAP under that pathway for as long as they otherwise satisfy continuation requirements. The repeal is prospective, applying to applications rather than to existing awards.

Two timing mechanics are worth knowing if you are modelling revenue:

• For technologies first approved before FY2025, CMS uses the midpoint of the upcoming fiscal year (April 1) to judge whether the newness period has expired. For technologies first approved in FY2025 or later, CMS extends eligibility by an additional year when the three-year anniversary of market entry falls on or after October 1 of that fiscal year.
• A documented delay in commercial availability extends the newness period only until the NTAP becomes effective. If the technology is still unavailable then, the newness period is treated as beginning September 30 before the NTAP start date.

On the Other Side

Reading the NTAP repeal alone produces a misleading impression of the policy direction. On April 23, 2026, CMS and FDA announced the Regulatory Alignment for Predictable and Immediate Device (RAPID) coverage pathway, designed to compress the gap between FDA marketing authorization and national Medicare coverage.

Under RAPID as announced, CMS would issue a proposed national coverage determination the same day a participating device receives marketing authorization, triggering a 30-day comment period, with a final NCD potentially following within roughly two months, against the year or more that national coverage typically takes today.

Eligibility as announced is narrow. A device must hold Breakthrough Device designation; be either a Class III device or a Class II device participating in FDA's Total Product Life Cycle Advisory Program; and be the subject of an IDE study that enrolls Medicare beneficiaries and studies clinical health outcomes agreed between FDA and CMS. As of the announcement, 122 devices had enrolled in the TAP pilot since its 2022 launch.

RAPID is not yet in force. CMS and FDA anticipated publishing a formal proposal in the Federal Register, followed by a 60-day comment period and a final notice, with launch tied to that final notice. No launch date has been given.

Taken together, the two moves read as a reallocation rather than a retreat: a higher evidentiary bar for supplemental payment, paired with a proposed acceleration of coverage for a narrower set of devices.

Clinical AI

Breakthrough Device designation retains its FDA-side value, prioritized review and greater access to agency interaction during premarket review. What it stops carrying, for FY2028 applications onward, is a reduced evidentiary burden at the payment stage.

The practical consequence is that substantial clinical improvement becomes the document you have to write. It is worth knowing that CMS defines it more broadly than "we ran an RCT." Under 42 C.F.R. § 412.87(b)(1)(ii), a technology can satisfy the criterion by offering a treatment option for patients unresponsive to or ineligible for current treatments; by diagnosing a condition that is currently undetectable, or diagnosing it earlier than existing methods with evidence that the earlier diagnosis affects patient management; by significantly improving clinical outcomes relative to existing technologies; or where the totality of the information otherwise demonstrates substantial improvement.

For diagnostic and triage software, the second route is the one that fits most naturally, and it contains the requirement most often skipped in AI evidence packages: it is not enough to detect earlier. You need evidence that detecting earlier changed what clinicians did.

That is a study design decision, and for anyone intending to file for FY2028, it is a decision with roughly one application cycle left to make.

Open Questions

• Whether raising the bar improves what reaches the bedside, or mainly advantages developers who can fund larger studies, is contested, and reasonable people in device policy land on both sides.
• The counter-argument to CMS's position is that reduced NTAP access slows adoption, which in turn slows the accumulation of the real-world evidence everyone says they want. The counter-argument to that is that supplemental Medicare payment is a poor instrument for subsidizing evidence generation.
• Congress may yet weigh in. The Ensuring Patient Access to Critical Breakthrough Products Act (H.R. 5343 / S. 1717) would provide four years of transitional Medicare coverage on Breakthrough Device designation. It advanced out of House Ways and Means 37-3 in September 2025, and in April 2026 a bipartisan group of 82 lawmakers urged HHS and CMS to use it as a model for regulation. Similar bills have been introduced repeatedly since 2016 without reaching a floor vote.

Sources.

CMS FY2027 IPPS/LTCH final rule (CMS-1849-F) and accompanying fact sheet, July 31, 2026; Federal Register document 2026-15833, published August 4, 2026; 42 C.F.R. § 412.87(b) and (c); 84 Fed. Reg. 42,044 (August 16, 2019), establishing the alternative pathway; CMS and FDA RAPID coverage pathway announcement, April 23, 2026. Application counts, pathway breakdown and grandfathering detail as reported in published analyses of the final rule by Holland & Knight (August 10, 2026), McDermott+ (August 4, 2026), and the American College of Cardiology (August 5, 2026).