A new Medicare pathway promises faster, more predictable coverage for breakthrough medical devices. When participation is voluntary, however, and manufacturers can take other routes, the real test is not who enters the fast lane, but what Medicare learns about those who don’t.
A medical device can be designated “Breakthrough” by the Food and Drug Administration (FDA) long before the most important clinical questions about it have been answered. This designation is an invitation to faster development and review, rather than a confirmation that the technology has been shown to improve patient outcomes.
Medicare is now proposing a new fast lane for certain Breakthrough-designated devices that aims to close the gap between FDA marketing authorization and coverage for medical devices. The Regulatory Alignment for Predictable and Immediate Device (RAPID) coverage pathway would bring the Centers for Medicare & Medicaid Services (CMS) into device development before the pivotal study is finalized, so FDA and CMS can tell manufacturers in advance what evidence each agency will need. If the study succeeds, CMS could publish a proposed national coverage determination (NCD) on the same day as FDA authorization and finalize it roughly 60 to 90 days later. Eligibility is narrow. RAPID is limited to a subset of Breakthrough-designated Class II and III devices at a particular stage of development, with additional requirements around Total Product Life Cycle Advisory Program (TAP) participation, investigational device exemption (IDE) studies, Medicare-relevant populations, clinical outcomes, and Medicare payment. In vitro diagnostics are excluded altogether.
The FDA and CMS make different decisions about the same device. The FDA controls market authorization, while CMS determines whether a technology meets Medicare’s coverage requirements. By statute, the agencies must apply different standards, so evidence sufficient for authorization may still leave important unanswered questions about benefits for Medicare beneficiaries. RAPID aims to bring Medicare’s evidence requirements into study design before market authorization, when manufacturers can still adapt their pivotal trials, using the prospect of national coverage as an incentive. Its ability to encourage additional evidence generation depends partly on how valuable that prospect is compared with the coverage routes available outside the program. Earlier coordination can clarify what CMS wants, but manufacturers’ incentives to produce that evidence depend on the consequences of leaving those questions unanswered.
Most of the debate so far has concerned the persuasiveness of Breakthrough evidence and eligibility for RAPID. Two questions have received much less attention: which manufacturers will voluntarily choose a pathway that asks them to answer Medicare’s coverage questions before market entry, and what happens to the devices that do not enter or leave the pathway before CMS reaches a national decision?
The Fast Lane is Optional
RAPID’s strongest feature is also what makes the incentive question interesting. CMS would engage before the pivotal study is finalized and ask for evidence relevant to Medicare beneficiaries’ clinical health outcomes. Depending on the technology, that may mean a more representative population, clinically meaningful endpoints, longer follow-up, or comparison with contemporary care. Those are exactly the questions a coverage agency should want answered.
Those requirements may increase the cost of a study, even if Medicare’s questions are not intrinsically harder than the FDA’s. A well-conducted study may also find little or no improvement over existing care. A sponsor confident of a favorable result may consider the cost worthwhile for the prospect of a faster, more predictable route to national coverage after FDA authorization. A less confident sponsor faces a different calculation.
RAPID leaves that calculation to the sponsor. “Participation in the RAPID coverage pathway is voluntary.” A manufacturer can instead pursue FDA authorization and then seek Medicare payment through existing routes, including local coverage and claim-by-claim adjudication. A manufacturer may also predict that Medicare coverage is not critical to commercial success, for example, if its target population is predominantly younger and commercially insured.
Seen economically, RAPID offers manufacturers early guidance and, if the evidence supports coverage, a faster and more predictable route to a national decision. Manufacturers bear the cost of producing that evidence, and CMS can decline coverage if it is inadequate. The question is whether RAPID prompts research that manufacturers would not otherwise conduct. Medicare Administrative Contractors (MACs) apply the same statutory reasonable-and-necessary standard as CMS, but coverage outside RAPID may be decided through local policies or individual claims. If a device can obtain Medicare payment outside RAPID without resolving the same evidence gaps, the incentive to enter the pathway is weaker.
That creates a selection problem and potentially adverse selection. The devices most willing to enter a pathway built around stronger evidence may be those whose sponsors are already most confident that stronger evidence will favor them. Technologies for which independent scrutiny would be most informative may have the greatest incentive to choose another route.
This does not imply bad faith by manufacturers. It is a basic feature of voluntary programs: participation is itself an outcome that must be studied. If the sponsors choosing the intervention differ systematically from those avoiding it, results among participants tell only part of the story.
The Exit Ramp Matters as Much as the Fast Lane
CMS’s own notice makes the problem unusually visible. Manufacturers may withdraw before a proposed NCD if data are “incomplete or insufficient”; CMS also notes that a sponsor may decide local coverage is “more advantageous.”
Both are understandable reasons to leave. But they raise an awkward policy possibility. A device could spend years in a process designed to identify the evidence Medicare needs, encounter unresolved problems, withdraw, and then return to the local coverage system. There, the same evidence is weighed jurisdiction by jurisdiction and inconsistent coverage has long been documented. The national review stops; the reimbursement question does not.
The right to withdraw also changes what entering RAPID means. Joining is not cost-free: a sponsor must build CMS’s study expectations into its pivotal trial to remain in the pathway. But the option to leave lowers the stakes. A sponsor can take CMS’s advice, run the study, and, if the evidence is insufficient for national coverage, fall back on the local route. Selection may therefore show up less at the entrance than at the exit, which is why withdrawals, and what happens after them, may tell us more than the number of devices that sign up.
We have written previously in these pages about Category III Current Procedural Terminology (CPT) codes, estimating that fewer than one in five ever become permanent Category I codes, and about the routes through which emerging procedures can become billable while evidence remains incomplete. Requiring further research after authorization does not guarantee that those gaps will be closed. A recent analysis found dozens of FDA-tracked postmarket device studies behind schedule, including one involving an authorized Breakthrough device. RAPID adds another version of the same systems problem. Raising the evidence bar inside one pathway does not necessarily raise it across the system if alternative routes remain available outside it. CMS itself appears to accept the point. In August 2026, it closed the alternative pathway to devices receiving Breakthrough designation after September 30, 2026. However, devices designated by that date may still apply through fiscal year 2029 for extra Medicare payments without demonstrating substantial clinical improvement over existing care, provided they meet the other requirements.
That distinction matters because RAPID could look successful by conventional program metrics. Participating devices might obtain faster NCDs. Their studies might be better aligned with Medicare’s needs. Sponsors might report high satisfaction. All of that could be true while the program has little effect on the devices that avoid it.
Count the Denominator
CMS initially estimated that about 40 devices would qualify, with perhaps 20 more potentially eligible. Numerically, that is a narrow pathway relative to the Breakthrough Device program. But device count is a poor measure of importance. 40 low-cost technologies used by a few thousand people and 40 implanted technologies used across the Medicare population have very different consequences for patients and public spending. RAPID’s significance therefore depends not simply on how many devices enter, but which devices enter, how widely they are used, and how much Medicare ultimately spends on them. The relevant unit is not simply the device count, but the healthcare system’s exposure to those devices: how much they cost, how many patients receive them, and how widely they diffuse.
CMS should not evaluate the pathway only by asking how many participating devices reach national coverage or how quickly they get there. It also needs to know how many technologies meet the eligibility criteria, how many eligible sponsors choose to enter, how many withdraw, at what stage they leave, and crucially, what happens next. Devices excluded by the eligibility rules should be reported separately from those whose sponsors choose not to participate. Once authorized, do eligible nonparticipating or withdrawn devices later obtain local coverage, claim-by-claim payment, or another route to reimbursement? Where CMS has identified unresolved evidence questions, are those questions eventually answered, or do they disappear with the application?
Some premarket information will legitimately remain confidential. That need not prevent aggregate reporting of participation, withdrawal, and subsequent coverage routes, or publication of reasons once they can lawfully be disclosed. CMS already proposes transparency about devices accepted into RAPID. To evaluate a voluntary program, the missing half is visibility into who does not remain in it.
The same principle should apply to evidence. If a manufacturer withdraws after CMS has identified an important unresolved coverage question, that issue should not simply be reset to zero when the technology appears later before a Medicare contractor. The appropriate disclosure mechanism will depend on confidentiality and statutory constraints, but the policy objective is straightforward: changing venue should not erase what the system has already learned.
Experience abroad supports judging RAPID by what it achieves rather than by how smoothly the agencies cooperate. Evaluations of joint advice from European regulators and health-technology-assessment bodies show that it can change development programs, including the choice of comparators and primary endpoints. The UK relaunched its Innovative Licensing and Access Pathway after stakeholders reported that the offer was unclear, entry criteria too permissive, and direct National Health Service (NHS) involvement insufficient to support timely adoption, and has piloted a device counterpart, the Innovative Devices Access Pathway. Early alignment between agencies can change the evidence that gets generated. However, it does not by itself show that patients gain faster access or that the wider evidence system improves. Alignment is an intermediate outcome, not the endpoint.
RAPID could be a success for devices that use it and still fall short as evidence policy if the devices most in need of scrutiny routinely take another route. Five years from now, the pathway should not be judged only by how quickly its successful participants reached national Medicare coverage. The harder test is whether Medicare knows more about the new devices it pays for, including the ones that chose another road.
