Substantial equivalence is the legal test at the heart of the FDA 510(k) pathway: a new device is cleared when the applicant shows it has the same intended use as a legally marketed predicate device and either the same technological characteristics or different characteristics that do not raise different questions of safety and effectiveness. For EU founders, the concept looks superficially similar to equivalence under MDR Article 61 and MDCG 2020-5, but the two are different mechanisms solving different problems under different laws. Predicate selection is the single most consequential decision in a 510(k), and most of the time it is where EU founders try to reason from MDR instincts and get the answer wrong. This post frames the FDA concept at the orientation level and contrasts it with the MDR equivalence framework the same founders already know.

By Tibor Zechmeister and Felix Lenhard. Last updated 10 April 2026.

TL;DR

  • Substantial equivalence is the FDA 510(k) legal test. A device is cleared when the applicant convincingly demonstrates that it is as safe and effective as a legally marketed predicate.
  • The two analytical pillars on the FDA side are intended use (must be the same) and technological characteristics (same, or different in ways that do not raise different questions of safety and effectiveness).
  • Equivalence under MDR Article 61 and Annex XIV Part A Section 3 is a different mechanism with three pillars (technical, biological, clinical) and a data access requirement for implantable and Class III devices set out in Article 61(5) and clarified by MDCG 2023-7.
  • Predicate selection is the highest-leverage strategic decision in the 510(k) pathway. A weak predicate produces rework, delay, and in the worst case a Not Substantially Equivalent finding.
  • EU founders should not reason from MDR equivalence instincts when picking a 510(k) predicate. The logic, the evidence, and the documentation all run on different rails.
  • Specific predicate selection, product code lookup, and the detailed content of substantial equivalence arguments belong in a structured meeting with a US regulatory specialist. This post is the orientation.

Why EU founders need to understand this concept clearly

In Tibor's early conversations with EU founders exploring the US market, the word "equivalence" shows up quickly. The founders have usually spent two or three years inside MDR thinking. They have read MDCG 2020-5. They have written at least a draft clinical evaluation report. They know what a predicate means in their world. When a US regulatory specialist then mentions "substantial equivalence" and "predicate device" during a pathway discussion, the founders nod along — and then quietly assume the two concepts are the same concept with different labels.

They are not. The overlap in vocabulary hides a difference in legal framework, analytical method, evidence expectations, and documentation. A founder who walks into a 510(k) project with MDR equivalence instincts will over-engineer the parts that do not matter on the FDA side and under-engineer the parts that do. The cost of the confusion is measured in wasted months of draft submission work, testing that does not answer the right question, and in the worst cases a Not Substantially Equivalent finding late in review.

This post separates the two concepts deliberately. It explains substantial equivalence on the FDA side at the orientation level, contrasts it with equivalence under MDR Article 61 and MDCG 2020-5, and lays out the shape of a defensible predicate device strategy. It does not turn you into a 510(k) expert — that is not what a blog post does — and it does not replace the US regulatory specialist who owns your actual submission.

What substantial equivalence means at the FDA

On the FDA side, substantial equivalence is the legal test that governs whether a new device can be cleared through the 510(k) pathway. The applicant identifies a legally marketed predicate device and demonstrates that the new device is substantially equivalent to that predicate along two dimensions: intended use and technological characteristics.

If the intended use matches and the technological characteristics are the same, the case is straightforward. If the technological characteristics are different, the applicant must demonstrate that the differences do not raise different questions of safety and effectiveness compared with the predicate. This second branch is where most real 510(k) submissions actually live, because almost every new device differs from its predicate in some technical respect that matters to the engineers, even if not necessarily to the reviewer.

A successful substantial equivalence argument produces a clearance letter from the FDA that lets the device be placed on the US market under the intended use described in the submission. A failed argument produces either additional information requests that prolong the review or, in the worst case, a Not Substantially Equivalent finding that forces the applicant to reconsider the pathway entirely — typically by escalating into the De Novo pathway for genuinely novel low-to-moderate risk devices.

A few things substantial equivalence is not, because the vocabulary is routinely misread by people coming from other regulatory systems. It is not a finding that the device is safe and effective in any absolute sense — it is a comparative finding against a predicate. It is not an approval — it is a clearance, which is a different legal outcome from the Premarket Approval applicable to most Class III devices. And it is not a rubber stamp — the FDA reviews the argument, can ask for more information, and can reject it.

For the broader 510(k) framing see The 510(k) process explained: how to get FDA clearance for your device. For the FDA framework as a whole, see FDA regulation of medical devices: a primer for EU startups.

How it differs from MDR equivalence

Equivalence under the MDR is the superficially similar concept that trips up EU founders. Both allow a new device to lean on an existing device in some way. Both involve comparison tables and careful argument. Both reward clean predicate selection. The similarities stop there.

Under MDR Article 61 and Annex XIV Part A Section 3, equivalence is a mechanism that lets a manufacturer use the clinical data generated for an equivalent device as clinical evidence for its own device during clinical evaluation. The test, clarified by MDCG 2020-5, runs across three independent pillars — technical, biological, and clinical characteristics — all of which must be demonstrated in full. For implantable devices and Class III devices, Article 61(4)–(6) further requires that the manufacturer has sufficient levels of access to the predicate's technical documentation, which MDCG 2023-7 (December 2023) clarifies in practice typically means a contract with the original manufacturer. The whole machinery exists so that a Notified Body can accept someone else's clinical data as the evidence backing your clinical evaluation report.

Substantial equivalence at the FDA does a different job. It is not about inheriting someone else's clinical data. It is the legal test that determines whether the 510(k) pathway is available at all, by asking whether the new device is comparable enough to a legally marketed predicate that a lighter premarket notification is the right regulatory route rather than a Premarket Approval or a De Novo. The evidence supporting substantial equivalence is typically bench performance data, biocompatibility where relevant, electrical safety and electromagnetic compatibility for electrical devices, software verification and validation for software-containing devices, and in some device categories clinical data — but the clinical data, when it is included, supports the safety and effectiveness argument rather than being "inherited" from the predicate.

The practical implication for an EU founder is that the two frameworks are solving different problems and the mental model from one does not carry over. A device can be genuinely equivalent to a competitor under MDR and have no viable predicate for a 510(k), because the US regulatory history or product code landscape simply does not line up. A device can have a strong 510(k) predicate and still need its own clinical investigation for MDR, because the intended purpose under MDR Article 2(12) locks the clinical population in ways an FDA intended use does not. Mapping one to the other is not a translation exercise. It is two separate analyses.

See Equivalence under MDR: demonstrating a predicate device for the MDR side in depth, and Equivalence under MDR vs substantial equivalence under 510(k) for the direct side-by-side.

Predicate device selection

Picking the predicate is the highest-leverage strategic decision in a 510(k) project and the one EU founders most consistently get wrong when they try to run the pathway without US-specific help. A strong predicate strategy treats the decision as a screening and ranking exercise rather than a lookup.

A valid predicate is a device that has been legally marketed in the United States and that your new device can credibly be compared against along the two substantial equivalence dimensions. Strong predicate screening considers at least the following:

  • Intended use fit. The predicate's cleared intended use is close enough to your intended use that the comparison is not forced.
  • Technological proximity. The underlying technology is close enough that performance testing to bridge any differences is feasible with a reasonable budget.
  • Product code alignment. The device sits in the same FDA product code as the predicate, or there is a defensible argument for the product code the submission will use.
  • Review history. The predicate's own clearance history and any related FDA communications do not signal that the category is under increased scrutiny or mid-reclassification.
  • Currency. The predicate is still legally marketed and has not been withdrawn or recalled in ways that would weaken the comparison.

Founders sometimes assume that any cleared device in a vaguely similar space is a valid predicate. It is not. A specialist will typically screen several candidates, rank them, and pick the one that produces the cleanest comparison with the lowest testing burden. Doing this well is the difference between a 510(k) that moves through review in a predictable window and one that cycles through additional information rounds.

For classification logic on the US side, see FDA device classification for EU startups.

Intended use matching

Intended use is the single most common place where 510(k) arguments quietly fail. The new device must have the same intended use as the predicate. Not overlapping. Not similar. The same.

Intended use on the FDA side plays a role comparable to intended purpose under MDR Article 2(12): it is the anchor that defines what the device is, what it is for, and who is meant to use it. Drift in intended use between subject and predicate is the primary reason a 510(k) runs into trouble with a review division. A difference in patient population, a difference in the body site, a claim expansion relative to the predicate, a shift in the user profile — any of these can turn a promising predicate into a broken argument.

The discipline for EU founders is to draft the indications for use statement early and lock it before the submission drafting begins. Drift between the intended purpose used for CE marking and the indications for use in a 510(k) is common in dual-market projects and is a recurring source of rework. The two statements do not have to be identical word for word, but they must be internally consistent, and the differences must be defensible in both directions.

Technological characteristics matching

The second substantial equivalence dimension is technological characteristics. The question is whether the new device has the same technology as the predicate, or whether any differences raise different questions of safety and effectiveness. The second branch is where the real analytical work lives, and it is where performance testing is built.

A credible technological characteristics analysis lays out the key features of the new device and the predicate side by side, identifies every difference that matters, and for each difference explains why it does not raise a new safety or effectiveness question — with evidence, not with assertion. The evidence is typically the performance testing that the submission includes: bench performance data, biocompatibility where relevant, electrical safety and EMC for electrical devices, sterility and shelf life where relevant, and software verification and validation for software-containing devices.

The temptation for engineering-led teams is to highlight every differentiator the new device offers, because that is how engineers think about product. The discipline for a 510(k) is the opposite: the new device's differences need to be neutralised from a safety and effectiveness standpoint, not celebrated. Differentiators are for the commercial deck. The submission is about the comparison, not the differentiation.

Performance data

Performance data is what turns a substantial equivalence argument from narrative into something a reviewer can assess. At the orientation level, the performance data in a 510(k) submission typically covers bench testing of the device's key functions, biocompatibility for materials in patient contact, electrical safety and electromagnetic compatibility for electrical devices, sterility and shelf life where relevant, and software verification and validation for software-containing devices. Clinical data is required in some device categories and not others — the category-specific answer belongs to a US regulatory specialist.

Two points matter for EU founders from day one. First, much of the underlying performance testing overlaps with the testing already planned or completed for CE marking. Test plans written with both regulators in mind can often serve both submissions with minimal duplication. Second, the testing must be designed to answer the specific substantial equivalence questions the predicate raises, not to look comprehensive. Testing that does not target a real substantial equivalence question is waste, even when it looks diligent.

Common misunderstandings

  • "Substantial equivalence is the same as MDR equivalence." It is not. Different legal basis, different purpose, different evidence expectations.
  • "Any cleared device is a valid predicate." It is not. Intended use, product code, technological proximity, and currency all matter.
  • "The new device's innovations will impress the FDA reviewer." They will not, at least not in this submission. The reviewer is assessing comparability, not novelty.
  • "If the MDR technical documentation is complete, the 510(k) writes itself." It does not. The underlying data is often reusable; the submission has to be rebuilt around substantial equivalence to a predicate.
  • "The intended purpose statement from the MDR file can be dropped into the 510(k) unchanged." Sometimes yes, often no. Both statements have to be drafted with both regulators in mind, and drift between them is a recurring source of rework.
  • "A weak predicate can be strengthened by adding more testing." Only when the testing is answering the actual substantial equivalence question. Piling on unrelated testing does not rescue a bad predicate choice.

The Subtract to Ship angle on substantial equivalence

The Subtract to Ship discipline applies to 510(k) work the same way it applies to MDR work: strip every activity that does not trace to a specific regulatory obligation. On the substantial equivalence side, the most common waste to subtract is speculative drafting against a predicate that has not been validated by a US specialist, performance testing designed to look thorough rather than to answer a specific substantial equivalence question, and narrative elaboration that does not move the reviewer's comparison forward.

The underlying work — predicate screening, intended use drafting, technological characteristics comparison, performance testing aligned to the comparison, and submission assembly — still has to happen. Subtract to Ship does not cut regulatory work. It cuts regulatory theatre that is disguised as regulatory work, and on the 510(k) side that theatre most often shows up as effort spent propping up a weak predicate choice instead of picking a better one. See The Subtract to Ship framework for MDR for the methodology.

Reality Check — Is your predicate strategy defensible?

  1. Have you identified a specific predicate by manufacturer, device name, product code, and clearance number — or are you still thinking in terms of "devices like ours"?
  2. Is your intended use for the 510(k) drafted and locked, and is it internally consistent with your intended purpose under MDR Article 2(12)?
  3. For every technological difference between your device and the predicate, can you explain in plain language why that difference does not raise a different question of safety and effectiveness?
  4. Is your planned performance testing designed to answer the specific substantial equivalence questions your predicate raises, or to look comprehensive?
  5. Have you had your predicate choice screened by a US regulatory specialist, or are you still assuming the predicate is valid because the technology is common?
  6. Do you know the product code your submission will use, and is it the same as the predicate's, or do you have a defensible argument for a different one?
  7. If a reviewer asked you tomorrow why this predicate is the right predicate, could you answer in two clear sentences?

If you cannot answer more than three of these seven cleanly, your predicate strategy is not ready for submission drafting. That is a solvable problem — but the solution is a structured conversation with a US specialist, not another round of internal analysis.

Frequently Asked Questions

Is substantial equivalence under 510(k) the same as equivalence under MDR? No. Substantial equivalence at the FDA is the legal test that determines whether the 510(k) pathway is available for a device, based on intended use and technological characteristics compared against a legally marketed predicate. Equivalence under MDR Article 61 and Annex XIV Part A Section 3 is a mechanism that lets a manufacturer use another device's clinical data as clinical evidence, with three pillars (technical, biological, clinical) as clarified by MDCG 2020-5 and a data access requirement for implantable and Class III devices clarified by MDCG 2023-7. The two concepts share vocabulary and not much else.

What makes a predicate device valid for a 510(k)? A valid predicate is legally marketed in the United States, has a cleared intended use close enough to the new device to support a credible comparison, uses a technology close enough to allow practical bridging, sits in the same or a defensible product code, and is still on the market. A specialist typically screens several candidates and ranks them before the submission strategy is locked.

Can my MDR equivalence predicate also be my 510(k) predicate? Sometimes, but not automatically. The fact that a device is equivalent under MDR does not mean it has an FDA clearance history that makes it usable as a 510(k) predicate, and vice versa. The two analyses have to be done separately and the overlap, when it exists, is a happy outcome rather than a default.

What happens if the FDA finds my device not substantially equivalent? A Not Substantially Equivalent finding does not end the project, but it does close the 510(k) door for that specific submission. The typical next step is to reconsider the pathway, which for a novel low-to-moderate risk device often means the De Novo pathway. The cost of the detour is real, which is why predicate strategy is the single highest-leverage decision in a 510(k) project and why it benefits disproportionately from specialist input early.

How is intended use different from MDR intended purpose? They play comparable roles as the anchor that defines what a device is and what it is for, but they are drafted for different regulators with different expectations and review logic. Intended purpose under MDR Article 2(12) is the statement that drives classification, conformity assessment, and clinical evaluation. The indications for use in a 510(k) is the FDA-facing statement that defines the cleared label language. The two should be internally consistent but are not automatically identical, and drift between them is a recurring source of rework in dual-market projects.

Does reusing MDR bench testing help my 510(k) submission? Yes, often substantially. Risk management files, bench testing, biocompatibility, electrical safety, EMC, and software verification and validation can frequently be reused as supporting data for a 510(k) when the underlying test plans were written with both regulators in mind. The submission document itself, however, has to be rebuilt around substantial equivalence to a predicate — the reuse is at the evidence level, not at the document level.

Where does this primer stop and where should I bring in a US specialist? This post is the orientation for EU founders who want to understand substantial equivalence clearly enough to have a productive first meeting with a US regulatory specialist. For the actual predicate selection, the actual product code, the actual content of the substantial equivalence argument, and the actual performance testing that your submission needs, work with a specialist who practices inside the FDA system daily. Tibor does not hold himself out as an FDA expert, and this blog stays honest about that boundary.

Sources

  1. Regulation (EU) 2017/745 of the European Parliament and of the Council of 5 April 2017 on medical devices (MDR), in particular Article 2(12) (intended purpose), Article 61 (clinical evaluation), Article 61(4)–(6) (clinical investigations and equivalence for implantable and Class III devices), and Annex XIV Part A Section 3 (equivalence). Official Journal L 117, 5.5.2017, as amended.
  2. MDCG 2020-5 — Clinical Evaluation — Equivalence: A guide for manufacturers and notified bodies, April 2020.
  3. MDCG 2023-7 — Guidance on exemptions from the requirement to perform clinical investigations pursuant to Article 61(4)–(6) MDR and on "sufficient levels of access" to data needed to justify claims of equivalence, December 2023.
  4. U.S. Food and Drug Administration — Center for Devices and Radiological Health (CDRH), public guidance on the 510(k) premarket notification pathway, substantial equivalence, and predicate devices. Referenced at the general framework level. For specific predicate lookups, product code assignments, current review division expectations, and the current content requirements for substantial equivalence arguments, consult the FDA directly at https://www.fda.gov/medical-devices and work with a US regulatory specialist. This primer intentionally stays at the orientation level for the FDA-side content.

This post is part of the FDA & International Market Access series in the Subtract to Ship: MDR blog. Authored by Felix Lenhard and Tibor Zechmeister. A note on the limits of our expertise, stated once and clearly: Tibor's authority is in the EU MDR and the Notified Body system, and the FDA framing in this post stays deliberately at the general level. For specific substantial equivalence questions on your device — predicate screening and ranking, product code selection, intended use drafting for the FDA, technological characteristics comparison, and the performance data that will support the argument — work with a US regulatory specialist who practices inside the FDA system daily. This post is the orientation. The specialist is the execution.