For about two decades, medical device startups followed a well-worn route. Get a CE mark in Europe first, where approval was faster and cheaper. Start selling, gather clinical data, then take that evidence to the US Food and Drug Administration. Europe was the proving ground. America was the prize.
That route has largely closed. The EU Medical Device Regulation, Regulation (EU) 2017/745, raised the bar for clinical evidence and put notified bodies, the private organisations that certify devices, under far tighter control. The transition proved so difficult that the EU had to extend it. Under Regulation (EU) 2023/607, many legacy devices can stay on the market until the end of 2027 or 2028, depending on risk class, provided their manufacturers had applied to a notified body by 2024.
The usual story is that Europe got stricter and the US stayed flexible. That is partly true. But it misses the more useful point for anyone building a device company: the major regulatory systems do not just differ in how strict they are. They differ in who does the reviewing, what they are paid to do, and what kind of company each system quietly favours.
The United States: comparison, not proof
Most moderate-risk devices in the US reach the market through the 510(k) pathway. A manufacturer shows that its device is “substantially equivalent” to a predicate, a device already legally on the market. It does not have to independently prove safety and effectiveness in the way a high-risk device going through premarket approval must.
This makes the US pathway fast and predictable for incremental innovation. A slightly better catheter, a new version of an imaging system or a refined surgical instrument can often find a predicate. The review is done by FDA staff, funded partly by user fees, under published performance goals.
The weakness is also well known. Predicates can chain back through decades of small modifications, so a device may be cleared by comparison with something quite different from itself. For genuinely novel devices without a predicate, the De Novo route exists, and for devices that address serious conditions, the Breakthrough Devices Program offers more interaction with FDA reviewers. According to FDA figures reported at the start of 2026, the programme had granted 1,246 breakthrough designations by the end of 2025, and 185 of those devices had gone on to receive marketing authorisation. Designation is a signal of promise, not a guarantee of approval.
Europe: private reviewers, public rules
The EU does not have a central device regulator comparable to the FDA. Instead, manufacturers contract with a notified body, a private organisation designated by a member state, which assesses conformity and issues the certificate. As of October 2025, the European Commission’s database listed 51 notified bodies designated under the MDR.
That structure creates tensions the US system does not have. Notified bodies are paid by the companies they assess. Under the old directives, critics argued that this encouraged manufacturers to shop for lenient reviewers. The MDR tightened oversight in response, but the result, at least during the transition, was a capacity squeeze: fewer designated bodies, more work per device and longer queues. Small companies with a single product have found themselves at the back of those queues.
The EU system also requires more ongoing clinical evaluation and post-market surveillance than before. That is defensible on safety grounds. It also means the cost of keeping a device on the European market is now higher, which falls hardest on low-volume and niche products.
Who pays the reviewer, and what that does
The deepest difference between the systems is not the rulebook. It is the economics of review. At the FDA, a public agency reviews submissions, funded by a mix of appropriations and user fees negotiated with industry, with published performance goals for how quickly it responds. The reviewer is not chosen by the company and does not compete for its business. The pressure on the FDA is political and procedural: to meet goals, to avoid both recalls and accusations of slowness.
In Europe, the company chooses and pays its reviewer, and notified bodies compete, within the rules, for clients. That is not inherently corrupt; auditors in many industries work the same way. But it creates different pressures. A notified body with limited capacity rationally prioritises large clients with many products and predictable files over a start-up with one novel device and an unfamiliar evidence package. The queue is not only a function of rules. It is a function of who is worth serving first.
For a founder, that has a concrete implication. In the US, the quality of the submission largely determines the experience. In Europe, the relationship with the notified body, and when you secured a slot, can matter nearly as much.
The 510(k) critique, and where it lands
The US pathway has its own serious critics. In 2011, an Institute of Medicine committee asked by the FDA to review the 510(k) process concluded that it could not reliably determine safety and effectiveness, precisely because it rests on comparison rather than independent evidence, and recommended that the FDA develop a new framework. That did not happen in the form the committee proposed. The FDA has instead tightened guidance and encouraged the use of more modern predicates, while leaving the basic structure intact.
That critique partly holds, and it matters for founders as well as patients. A device cleared quickly on a thin comparison may still face payers who want outcome data, hospital value-analysis committees that ask for evidence, and product liability exposure if something goes wrong. Speed at the regulator can be followed by slowness everywhere else. The cheapest route to clearance is not always the cheapest route to a business.
The quiet convergence
For all the talk of divergence, there is also real convergence, mainly on quality systems. The Medical Device Single Audit Program lets a manufacturer undergo one audit of its quality management system that is recognised by regulators in Australia, Brazil, Canada, Japan and the United States. It does not replace product review, but it removes one of the most tedious duplications in global market entry.
Convergence on quality systems and divergence on clinical evidence is a sensible division. Factories and processes can be judged by common standards. What counts as enough evidence of benefit for a particular patient population is a harder, more political question.
Europe tries to fix itself
Europe has not ignored the damage. In December 2025, the European Commission presented a proposal to amend the MDR and the in vitro diagnostics regulation. Among its stated aims are simpler rules, fewer re-certification obligations, timelines for completing conformity assessments, participation in the Medical Device Single Audit Program, and new pathways for orphan and breakthrough devices. The proposal now has to pass through the European Parliament and the Council, which will take time and may change it substantially.
For anyone planning a launch in the next two years, the sensible assumption is that the current MDR is the system you will face, and that any relief will arrive late and in amended form. But the direction of travel is informative. Europe’s own institutions have concluded that the balance tipped too far toward cost and delay, particularly for small companies and niche products. That makes the political risk of a European strategy lower than it looked two years ago, even if the practical burden has not yet changed.
The trade-off founders do not always see
The choice of first market is usually framed as a question of speed. It is really a question of strategy.
A US-first route suits companies making incremental improvements that can find a predicate, and who have access to US investors and a US reimbursement plan. A route through Europe, under the MDR, now demands more clinical evidence up front, which suits companies that need to generate that evidence anyway. Other markets, including those in the MDSAP group, can follow once a quality system and core evidence are in place.
What no longer works is treating Europe as a cheap testing ground. And what has never worked is choosing a market for regulatory speed without a plan for who will pay for the device once it is approved. Clearance is not adoption.
There is a public-health cost here too. When regulatory costs rise, companies drop low-margin products and small markets first. The products most exposed are those serving small patient groups, such as paediatric devices, where sales volumes may never justify the cost of recertification. Stricter regulation protects patients from bad devices; it can also leave them without good ones.
A founder’s sequence, sketched
Consider two hypothetical companies. The first makes an improved orthopaedic implant with clear predicates in the US market and an investor base in Boston. For it, US clearance first is the natural route, with Europe following once it has clinical data from American use that can support an MDR file. The second makes a novel diagnostic device for a rare paediatric condition, with no predicate and a small market. For it, a De Novo request in the US, possibly with breakthrough designation, may make sense, but only if it has a realistic reimbursement plan; in Europe, it would be wise to track whether the proposed orphan-device pathway survives the legislative process before committing to the full cost of certification.
Neither company is choosing on speed alone. Each is matching its evidence, its buyers and its capital to the system that rewards what it already has. That is the sense in which a regulator is a business model. Get the order wrong and a good device can spend years and much of its funding waiting in the wrong queue, while a competitor with an inferior product but a better-matched route reaches patients and payers first.
Beyond the founders
Regulatory choices ripple outward. Hospital procurement teams in Europe have already seen some products withdrawn because manufacturers judged recertification not worth the cost, and clinicians have had to find substitutes, sometimes for devices they had used safely for years. Procurement and clinical leaders can reduce that risk by asking suppliers early whether a product’s certification is secure, and by flagging to regulators the low-volume devices their patients cannot do without.
Investors, for their part, should price regulatory route into valuations more carefully than they often do. A company with a clear predicate and a US reimbursement code is a different asset from one depending on a novel pathway and an unproven notified-body slot, even if their technology is equally impressive.
And for regulators, the lesson from the past decade is that the costs of a system fall unevenly. Rules that large manufacturers absorb as overhead can remove small companies and niche products from the market altogether. The things to watch over the next year or two are whether the European reform keeps its promised timelines and orphan-device pathway through the legislative process, and whether the FDA’s breakthrough programme turns more of its designations into authorisations and, crucially, into paid-for products.
The takeaway
Comparing device regulators by speed alone misses the point. The US rewards incremental improvement through comparison. Europe now rewards companies that can afford sustained clinical evidence and patient notified-body queues. Global harmonisation is advancing on factories, not on evidence. Founders who treat regulatory strategy as business strategy will choose better markets, and in the right order.
Has the MDR changed which market your organisation launches in first, or which products it keeps? Tell us in the comments.
