Regulation (EU) 2017/745 (MDR) lets a manufacturer build its clinical evaluation partly on clinical data from another device, if equivalence between the two can be demonstrated. Done well, it saves a clinical investigation or shortens one. Done loosely, it is one of the most common reasons a notified body rejects a clinical evaluation. The bar is higher than under the Directives, and the burden of proof sits entirely with the manufacturer.
Where equivalence appears in the MDR
Mostly in clinical evaluation. Article 61 and Annex XIV Part A, section 3 allow clinical data on an equivalent device; Annex VII and Annex IX ask the notified body to check the claim. Equivalence also reaches further than the clinical evaluation report: it can justify the exemptions from clinical investigation in Article 61(4), feeds the post-market clinical follow-up (PMCF) plan, and appears in the health institution exemption of Article 5(5), where an in-house device must be justified against equivalent devices on the market. The MDR also expects your quality management system to say how you handle equivalence.
The guidance
MDCG 2020-5 (April 2020) is the guide for manufacturers and notified bodies on equivalence. It still refers to MEDDEV 2.7/1 rev. 4 where that document is consistent with the MDR. Since December 2023, MDCG 2023-7 adds two things the original article did not cover: how the exemptions from clinical investigation in Article 61(4) to (6) work, and what “sufficient levels of access” to the other device’s data means. Where guidance and Regulation differ, the Regulation wins.
Technical, biological and clinical characteristics
All three must be shown, and the MDR is stricter than MEDDEV on each.
Technical. The device must be of similar design and used under similar conditions of use, with similar specifications and properties, similar deployment methods, and similar principles of operation and critical performance requirements. The MDR names software algorithms explicitly. For software, what is compared is the functional principle, the clinical performance and the intended purpose, not the code itself. Software without a medical purpose does not have to be similar.
Biological. The same materials or substances in contact with the same human tissues or body fluids, for a similar kind and duration of contact, with similar release characteristics. MEDDEV allowed some exceptions here; the MDR does not.
Clinical. The same clinical condition or purpose, including similar severity and stage of disease, at the same site in the body, in a similar population, for the same kind of user, with similar critical performance for the expected clinical effect.
Compare all three, document the differences, and justify scientifically why none of them is clinically significant for safety and clinical performance. A notified body reads the differences first. An equivalence table that lists only similarities tells it you have not looked.
Access to the other device’s data
The MDR requires the manufacturer to have sufficient access to the data on the equivalent device to justify the claim. For class III and implantable devices, the bar is explicit: where the equivalent device belongs to another manufacturer, the two must have a contract that gives full access to the technical documentation on an ongoing basis (Article 61(5)). MDCG 2023-7 reads the general requirement for other classes as not, in itself, demanding a contract, but you still have to show that the access you have is enough. Published literature on a competitor’s device rarely is.
Where to document it
In the clinical evaluation report. MDCG 2020-5 provides a template equivalence table in its Annex I. Identify each device claimed as equivalent, compare it characteristic by characteristic, and give the justification for each difference.
When equivalence fails
Data on similar devices, meaning devices in the same generic group, is still useful. It feeds risk management (hazards and clinical risks), the state of the art (natural course of disease, alternative treatments), the design of a clinical investigation or PMCF study, the choice of clinical outcome parameters, and the list of adverse events to watch.
What it means for your product
Decide early whether equivalence is a real route or a hope. For a class III or implantable device without a contract with the other manufacturer, it is usually not a route at all. For other classes, test the claim against all three characteristics and against the data you can actually access before you plan around it. If it fails, plan the clinical investigation or a stronger PMCF now rather than after the notified body’s first round of questions.
We write clinical evaluation plans and reports under MDR Annex XIV: Evidence and technical documentation.
Related briefing: What does Regulation (EU) 2026/977 change for notified bodies?