Class IIb and Class III devices sit at the top of the MDR clinical evidence ladder. Article 61 still governs the evaluation, and Article 61(11) still requires proportionality — but at this end of the scale proportionate means substantial. For implantable and Class III devices, Article 61(4) sets a default rule that clinical investigations shall be performed, with only four narrow exemption routes in Article 61(4)-(6) clarified by MDCG 2023-7. For Class IIb devices the default is not a mandated investigation, but the Notified Body scrutiny of the clinical evaluation report is intense, equivalence under MDCG 2020-5 is harder to close, and the PMCF expectations are continuous. Getting the evidence strategy right at Class IIb and III is where founders either ship a real device or burn two years and several million euros on work that was either too little or too much.

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

TL;DR

  • MDR Article 61(1) requires clinical evaluation for every device, and Article 61(11) governs proportionality — at Class IIb and III, proportionate means a substantially higher evidence bar than at Class I or IIa.
  • For implantable and Class III devices, Article 61(4) sets a default rule that clinical investigations shall be performed, with only the four exemption routes in Article 61(4), 61(5), and 61(6) available, clarified by MDCG 2023-7 (December 2023).
  • Equivalence under MDCG 2020-5 remains theoretically available at Class IIb and III, but the "sufficient level of access to data" requirement and the full technical-biological-clinical demonstration make it a pathway that closes in practice far more often than it opens.
  • PMCF under Annex XIV Part B is continuous, and for Class III and implantable devices the summary of safety and clinical performance referred to in Article 32 must be updated at least annually with PMCF data per Article 61(11).
  • Notified Body scrutiny at this level is intense — clinical evaluation reports are read page by page, equivalence arguments are stress-tested, and investigation plans under EN ISO 14155:2020+A11:2024 are expected to match the specific evidence gaps rather than generic templates.

Why the bar is higher and what that actually means

Class IIb and III are the classes where the device can cause serious harm or death if it fails, and where the clinical consequences of an inadequate evidence base land directly on patients. The Regulation responds to that with a higher default evidence bar and with specific provisions — the mandatory clinical investigation rule for implantable and Class III devices, the tight exemption list in Article 61(4)-(6), the annual summary of safety and clinical performance update for Class III and implantable devices — that do not apply lower down the scale.

The mistake founders make at this end is usually the opposite of the mistake they make at Class I. At Class I the temptation is to under-scope because "it's just Class I." At Class IIb and III the temptation is to over-scope reactively — commission a large confirmatory trial by default, write a 400-page clinical evaluation report because the Notified Body is going to be strict, buy every template the consultant offers. The default cost balloons into the millions and the timeline stretches past the point where the company can still ship.

Proportionate at Class IIb and III does not mean "light." It means matched to the specific device, the specific claims, and the specific evidence gaps that remain after literature, equivalence where available, standards, and prior clinical experience have been worked through. A novel Class III implant with no predicate and an active mechanism of action requires a substantial clinical investigation programme — the Regulation will not accept literature alone and the Notified Body will not either. A Class IIb device that is an incremental improvement on a widely-used predicate in the same manufacturer's portfolio may have a much lighter investigation footprint because the existing evidence covers most of the claims. Same classes, different evidence bars. The Evidence Pass exists to find the real bar rather than the imagined one.

Proportionate at the higher bar — what the Regulation actually requires

MDR Article 61(1) applies identically across all classes. Confirmation of conformity with the relevant general safety and performance requirements set out in Annex I, under the normal conditions of the intended use of the device, and the evaluation of the undesirable side-effects and of the acceptability of the benefit-risk ratio referred to in Sections 1 and 8 of Annex I, shall be based on clinical data providing sufficient clinical evidence, including where applicable relevant data from post-market surveillance under Annex III and the PMCF plan required under Annex XIV Part B.

Article 61(11) sets the proportionality principle and then adds a specific obligation at the top end: the clinical evaluation and its documentation shall be updated throughout the life cycle of the device concerned with clinical data obtained from the implementation of the manufacturer's PMCF plan in accordance with Part B of Annex XIV and the post-market surveillance plan referred to in Article 84, and for Class III devices and implantable devices, the summary of safety and clinical performance referred to in Article 32 shall be updated at least annually with such data. That annual update obligation is a Class III and implantable-specific requirement. It shapes what the PMCF plan has to be capable of producing, not just what it has to collect.

Annex XIV Part A specifies the process and content of the clinical evaluation. At Class IIb and III the structure is the same as at lower classes — scope definition, identification of clinical data, appraisal of clinical data, analysis of clinical data, and conclusion tied back to the GSPRs — but the depth of each step rises. Scope definition is expected to walk through every claim the manufacturer intends to make. Data identification is expected to cover a comprehensive literature base. Appraisal is expected to apply a documented methodology that a trained reviewer can reproduce. Analysis is expected to weigh the data against the specific GSPRs at a level of rigour that matches the risk the device carries.

The mandatory clinical investigation rule and the four exemption routes

Article 61(4) sets out the default rule. In the case of implantable devices and Class III devices, clinical investigations shall be performed, except if it is duly justified to rely on existing clinical data. That word "shall" is doing work. For implantable and Class III devices, the starting position is that a clinical investigation is mandatory, and the exemption has to be earned.

Article 61(4), 61(5), and 61(6) together define the narrow set of conditions under which the exemption can be claimed. MDCG 2023-7 (December 2023) is the guidance that clarifies exactly how these conditions are interpreted in practice. The document identifies four exemption cases.

The first exemption case, under Article 61(4), applies where the device has been designed by modifications of a device already marketed by the same manufacturer, the manufacturer has demonstrated that the modified device is equivalent to the marketed device in accordance with Annex XIV Section 3, and this demonstration has been endorsed by the Notified Body, and the clinical evaluation of the marketed device is sufficient to demonstrate conformity of the modified device with the relevant safety and performance requirements.

The second exemption case, under Article 61(5), applies where a device that has been modified is placed on the market or put into service by the same manufacturer, the manufacturer demonstrates to the Notified Body that the modifications do not adversely affect the benefit-risk ratio, and existing clinical data is sufficient — this is a narrower sibling of the first case that covers modifications rather than equivalent new designs.

The third case is equivalence to a device marketed by another manufacturer, which Article 61(5) makes available only where a contract is in place that allows the second manufacturer to have full ongoing access to the technical documentation of the first manufacturer, and the original clinical evaluation has been performed in compliance with the requirements of the MDR. The "sufficient level of access to data" language in MDCG 2023-7 comes from this provision, and in practice this exemption is rarely achievable for a startup equivalence claim against a competitor product because the required contractual access simply does not exist.

The fourth case, under Article 61(6), applies where the device has been lawfully placed on the market or put into service in accordance with Directive 90/385/EEC or Directive 93/42/EEC and the clinical evaluation is based on sufficient clinical data and is in compliance with the relevant product-specific common specifications for clinical evaluation of that kind of device, where such common specifications are available. This exemption covers legacy Directive-era devices transitioning into the MDR and does not apply to genuinely new Class III or implantable devices.

The honest read for a startup building a new Class III or implantable device is that the exemption routes are structurally narrow. Unless the device is a modification of an existing product from the same manufacturer, or an equivalent to another product with full contractual access to the underlying technical documentation, or a legacy Directive-era device, the default mandatory clinical investigation rule under Article 61(4) applies. Planning the clinical evaluation on the assumption that an exemption will materialise is not a plan. Planning it on the assumption that a targeted investigation will be needed, and designing that investigation to close the specific evidence gaps, is.

Equivalence under tighter conditions at the top end

MDCG 2020-5 (April 2020) governs equivalence across all classes, and at Class IIb and III its requirements bite harder. Equivalence still has to be demonstrated across technical, biological, and clinical characteristics, and the three columns still all have to survive, but at Class IIb and III the level of detail the Notified Body expects in the equivalence table is higher and the tolerance for unresolved differences is lower.

For implantable and Class III devices the equivalence pathway carries an additional constraint: Article 61(5) requires the "sufficient level of access to data" condition for equivalence claims against another manufacturer's device, and MDCG 2023-7 spells out what that means. Contractual access to the technical documentation of the predicate device is required, continuous, and demonstrable to the Notified Body. For most startup equivalence claims against competitor devices, this access does not exist and cannot be engineered, which closes the equivalence route in practice even where the technical and clinical columns would otherwise have survived.

Where equivalence does work at Class IIb and III is in the same-manufacturer case — an incremental product in the same company's portfolio, where the equivalence table is built against a predicate whose technical file the manufacturer already owns and whose clinical data the manufacturer already controls. This is the first Article 61(4) exemption case in action, and it is the realistic route for companies iterating on their own platform rather than trying to claim equivalence against someone else's product.

Ongoing PMCF expectations

Annex XIV Part B requires a PMCF plan for every device. At Class IIb and III the plan is expected to be active rather than passive — real prospective data collection, registry participation where appropriate, structured follow-up of implanted patients for the expected lifetime of the device, and for Class III and implantable devices the annual summary of safety and clinical performance update under Article 61(11) and Article 32.

The practical consequence is that the PMCF plan must be designed to generate data at the frequency and quality the annual update requires. A PMCF plan built around sporadic literature surveillance will not support an annual update to a Class III summary of safety and clinical performance. A PMCF plan built around prospective patient follow-up, registry enrolment, or structured post-market studies will. The design of the PMCF at this level is not a box-ticking exercise — it is the mechanism by which the clinical evaluation stays current over the lifetime of the device, and the Notified Body reviews it with that expectation in mind.

Notified Body scrutiny patterns at Class IIb and III

Notified Body review of the clinical evaluation report at Class IIb and III is intense. Reviewers read the report page by page, check the literature search against the intended purpose, test the appraisal methodology for reproducibility, and stress the equivalence argument on every dimension where the table claims equivalence. For Class III and implantable devices the clinical evaluation assessment report from the Notified Body is a substantial document in its own right, and under the clinical evaluation consultation procedure for certain Class III implantable and Class IIb active devices intended to administer or remove a medicinal product, an expert panel may also review the evaluation at the top of the scale.

The patterns we see most often in NB findings at this level are familiar. Overclaiming in the intended purpose beyond what the clinical data supports. Literature searches that cannot be reproduced from the documentation. Equivalence tables with unresolved differences in the biological or technical column. PMCF plans that do not match the evidence gaps the clinical evaluation identified. Clinical investigation plans that are generic templates rather than designed around the specific questions the investigation needs to answer. None of these findings are exotic. They are the same findings at higher stakes — and the stakes at Class IIb and III make the cost of a major non-conformity finding substantially higher, because re-doing clinical evidence work under review pressure is the most expensive way to do it.

Common mistakes at the top end

  • Treating Article 61(4) as if the exemption routes are broadly available when MDCG 2023-7 defines them narrowly. Planning on an exemption that does not actually apply is a plan that collapses on NB review.
  • Commissioning a generic confirmatory clinical investigation by default because "Class III needs one" without first defining the specific evidence gap the investigation needs to close. Oversized investigations cost the same whether they were needed or not.
  • Trying to build an equivalence argument against a competitor device without the contractual access to the technical documentation that Article 61(5) and MDCG 2023-7 require. The technical and clinical columns may look fine in draft; the access requirement will collapse the pathway at NB review.
  • Writing a PMCF plan around literature surveillance and complaint monitoring for a Class III device that needs an annual summary of safety and clinical performance update. The plan will not generate data the update can draw on.
  • Designing a clinical investigation under a generic template rather than under EN ISO 14155:2020+A11:2024 for good clinical practice and scoped to the specific evidence gaps the clinical evaluation identified.
  • Assuming that the scale of the clinical evaluation report is the measure of its quality. At Class IIb and III the report is judged on the quality of the argument, the reproducibility of the literature work, and the honesty of the gap analysis — not on page count.
  • Leaving the summary of safety and clinical performance obligation for Class III and implantable devices until after CE marking. The SSCP is a pre-market deliverable that becomes a post-market obligation under Article 32 and Article 61(11).

The Subtract to Ship angle — the Evidence Pass at the top of the scale

The Evidence Pass (the third pass in the Subtract to Ship framework for MDR — see post 65) asks the same question at every class. What is the cheapest legitimate source of evidence that satisfies the relevant GSPR for this specific claim about this specific device? At Class IIb and III the answer is more often "a targeted clinical investigation" than it is at lower classes, but even at the top of the scale it is almost never "every clinical investigation the default plan suggested." Subtraction at Class IIb and III means cutting the parts of the investigation programme that do not match a specific evidence gap, cutting equivalence arguments that cannot survive the access requirement, cutting PMCF activities that will not generate data the annual update can use, and cutting clinical evaluation report volume that is decoration rather than argument. It does not mean cutting required investigation work. The obligation under Article 61 is absolute. The scale of work to meet it is proportionate — and at Class IIb and III, proportionate is substantial but not unlimited.

Reality Check — Where do you stand?

  1. For a Class III or implantable device, have you run Article 61(4)-(6) and MDCG 2023-7 against your specific situation and written down which exemption case, if any, you can actually claim?
  2. If you are planning to rely on equivalence, do you have the contractual access to the predicate device's technical documentation that Article 61(5) and MDCG 2023-7 require, or are you hoping that equivalence will work without it?
  3. Does your clinical evaluation report follow the Annex XIV Part A structure, and can a Notified Body reviewer reproduce your literature search from the documentation as written?
  4. If a clinical investigation is required, is the investigation plan designed under EN ISO 14155:2020+A11:2024 and scoped to the specific evidence gaps the clinical evaluation identified — or is it a generic template?
  5. For a Class III or implantable device, is your PMCF plan capable of generating data at the frequency and quality the annual summary of safety and clinical performance update under Article 61(11) and Article 32 requires?
  6. Have you tested the benefit-risk argument in your clinical evaluation against Section 1 and Section 8 of Annex I, or only against the GSPRs that were easy to address?
  7. When was the last time you cut scope from the clinical evidence plan rather than adding to it, and what was the reason you cut it?

Frequently Asked Questions

Are clinical investigations always required for Class III and implantable devices under MDR? Article 61(4) sets a default rule that clinical investigations shall be performed for implantable and Class III devices, with exemptions available only under the four narrow cases in Article 61(4), 61(5), and 61(6) as clarified by MDCG 2023-7. For a genuinely new Class III or implantable device without a same-manufacturer predicate, without contractual access to another manufacturer's technical documentation, and without legacy Directive-era status, the default applies and a targeted clinical investigation is required.

Can equivalence work for a Class IIb or Class III device? It can, but the conditions are strict. MDCG 2020-5 requires demonstration across technical, biological, and clinical characteristics, and for implantable and Class III devices Article 61(5) adds the requirement of sufficient contractual access to the predicate device's technical documentation. In practice equivalence works at this level mainly in the same-manufacturer case, where the predicate is in the manufacturer's own portfolio. Equivalence against a competitor device at Class IIb or Class III is possible in principle and rare in practice.

What does "proportionate" mean for a Class IIb clinical evaluation? Article 61(11) requires the clinical evaluation and its documentation to match the nature, classification, intended purpose and risks of the device. For Class IIb this means a higher evidence bar than Class IIa and substantial Notified Body scrutiny of the clinical evaluation report, but without the Article 61(4) mandatory investigation default that applies to implantable and Class III devices. A Class IIb device with a strong literature base and good equivalence coverage may not need a new investigation; a Class IIb device with novel claims may.

What is the annual summary of safety and clinical performance update obligation? For Class III devices and implantable devices, Article 61(11) requires the summary of safety and clinical performance referred to in Article 32 to be updated at least annually with clinical data obtained from the PMCF plan and the post-market surveillance plan. This is a Class III and implantable-specific obligation that shapes how the PMCF plan must be designed — it has to generate data at the frequency and quality the annual update can draw on.

How do I avoid over-scoping a Class IIb or III clinical investigation? Run the Evidence Pass honestly before commissioning the investigation. Work the literature, equivalence where available under the tight conditions, compliance with harmonised standards, and prior clinical experience for everything they legitimately cover. Scope the clinical investigation to the specific evidence gaps that remain, under EN ISO 14155:2020+A11:2024, with an investigation plan designed around those gaps rather than around a generic template. A targeted investigation that closes the real gaps will pass Notified Body review; an oversized investigation will cost more without adding anything a reviewer will credit.

Sources

  1. Regulation (EU) 2017/745 of the European Parliament and of the Council of 5 April 2017 on medical devices, Article 61 (clinical evaluation), Article 61(4), 61(5), 61(6) (clinical investigation rule and exemptions for implantable and Class III devices), Article 61(11) (proportionality and annual update for Class III and implantable devices), Annex XIV Part A (clinical evaluation) and Part B (PMCF). Official Journal L 117, 5.5.2017.
  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. EN ISO 14155:2020 + A11:2024 — Clinical investigation of medical devices for human subjects — Good clinical practice.

This post is part of the Clinical Evaluation & Clinical Investigations series in the Subtract to Ship: MDR blog. Authored by Felix Lenhard and Tibor Zechmeister. At Class IIb and Class III, proportionate stops being a way to save money and becomes the specific discipline of matching substantial evidence work to substantial risk — the Evidence Pass exists to make sure that substantial does not quietly become unlimited.