Introduction
In our previous article, EU MDR – Where It Applies and Who Needs Representation, we looked at where the EU MDR actually applies and who must appoint an Authorised or Legal Representative. A natural next question follows for any manufacturer planning to generate clinical evidence in Europe: what does the EU MDR actually require when you run a study with your device?
It is a common, and costly, misconception that any study involving a medical device automatically becomes a clinical investigation under the EU MDR, with the full weight of competent authority authorisation behind it. In reality the picture is more nuanced. Whether your study falls under the MDR – and, if so, under which provisions – depends on what you are investigating, whether your device is CE-marked, and whether it is being used within its intended purpose. Get this wrong and you risk either over-engineering a simple study or, more seriously, running an unauthorised investigation.
This article explains when a study counts as a clinical investigation under the EU MDR, the main regulatory pathways (Articles 62, 74 and 82), and why you must always check national requirements on top of the Regulation.
What the EU MDR means by a “clinical investigation”
The Regulation defines a clinical investigation precisely. Under Article 2(45), a clinical investigation is “any systematic investigation involving one or more human subjects, undertaken to assess the safety or performance of a device.” Three elements must all be present: it must be systematic; it must involve human subjects; and its purpose must be to assess the safety or performance of a device.
That last element is the one most often misunderstood. The deciding factor is not whether a device is present in the study, but whether the safety or performance of that device is what is being assessed.
Not every study with a device is a clinical investigation
MDCG 2021-6 Rev.1, the Commission’s question-and-answer guidance on clinical investigations, makes this distinction explicit. It gives the example of a clinical trial of a medicinal product that happens to measure blood pressure or oxygen saturation. Devices are used to take those measurements, but because the safety or performance of the blood-pressure or oxygen-saturation device is not what is being assessed, the study is not a clinical investigation under the MDR, and those devices are not “investigational devices.”
There is an important caveat. If any endpoint or objective of the study relates to assessing the safety or performance of the device itself, the study does meet the Article 2(45) definition and becomes a clinical investigation. A study can therefore move in or out of scope depending purely on what its objectives say about the device.
Two further points matter here. First, even when a study is not a clinical investigation, any device used in it must still be lawfully on the market or put into service in compliance with the MDR (i.e. CE-marked, under Article 5). The MDR does not disappear simply because the study is not an investigation. Second, other EU and national legislation continue to apply regardless – including the General Data Protection Regulation (GDPR), national ethical-review rules, and national or local rules on research involving human subjects.
When it is a clinical investigation: the main pathways
Once you have established that your study is a clinical investigation, the next question is which provisions of the MDR apply. The answer turns on two things: whether the device is CE-marked, and whether it is used within its CE-marked intended purpose. The Regulation creates several distinct pathways.
Article 62 – premarket investigations for conformity assessment
Article 62 governs clinical investigations conducted as part of the clinical evaluation to demonstrate conformity – in other words, to generate the clinical evidence needed to CE-mark a device. These are the studies most people picture when they think of a “device trial”: first-in-human and pivotal studies of a device that does not yet bear the CE marking.
Under Article 62(1), such investigations must be designed, authorised, conducted, recorded and reported in accordance with Articles 62 to 80 of the MDR. In practice this means a formal application to the competent authority of each Member State where the study will run – which validates and then assesses the application within the timelines set out in the MDR.
For lower-risk devices (class I, and non-invasive class IIa or IIb), Article 70(7)(a) allows the investigation to begin once the application is validated, unless national law provides otherwise – and national requirements can be stricter. For other devices, Article 70(7)(b) requires express authorisation from the competent authority before you start. This is the most demanding and time-consuming pathway.
Article 74(1) – PMCF investigations on CE-marked devices
Article 74(1) covers post-market clinical follow-up (PMCF) investigations: studies of a device that already bears the CE marking, used within its intended purpose, to further assess it in a post market setting. The key trigger is whether the study requires participants to undergo procedures additional to those performed under normal conditions of use, and whether those additional procedures are invasive or burdensome.
Where that is the case, Article 74(1) requires the sponsor to notify the competent authority of each Member State concerned at least 30 days before the investigation begins, and a defined subset of the MDR’s premarket requirements (such as safety reporting) still applies. MDCG 2021-6 describes “burdensome” procedures broadly – those that may cause pain, discomfort, fear, risk, or disruption to a person’s life – and “invasive” procedures as those penetrating the body. If you are unsure whether your additional procedures cross that threshold, the guidance encourages you to ask the relevant national authority before you start.
If a PMCF study does not involve additional invasive or burdensome procedures, Article 74(1) does not apply; it instead falls under Article 82 (below), together with any national requirements.
Article 74(2) – CE-marked devices used outside their intended purpose
What if the device is CE-marked, but your study uses it outside its approved intended purpose: for a new indication, population, or anatomical site not covered by its CE marking? Article 74(2) makes clear that this is not treated as a light-touch PMCF study. Instead, the full premarket regime applies – Article 74(2) states that “Articles 62 to 81 shall apply” – exactly as if the device were not CE-marked for that use. (That is why the figure differs from the “Articles 62 to 80” cited for the Article 62 pathway above: the two provisions cross-refer to different blocks of the Regulation; both are correct in their own context.) A CE mark does not give you a free pass to study the device for something it was not approved to do.
Determining whether your planned use is “within” the intended purpose is itself a careful exercise; MDCG 2021-6 sets out how to compare the planned study use against the instructions for use, declaration of conformity, labelling and clinical evaluation.
Article 82 – “other” clinical investigations
Finally, Article 82 is the catch-all for clinical investigations that are not conducted for conformity-assessment purposes under Article 62(1) and are not notified PMCF studies under Article 74(1). It also captures studies of a CE-marked device used outside its intended purpose where the investigation is not run for conformity-assessment purposes – the non-conformity branch of Article 74(2). These could be investigator-initiated or academic studies that nonetheless assess a device’s safety or performance, for example independent research on a CE-marked device within its intended purpose.
For these, the MDR sets only a baseline of minimum requirements (a defined subset of Article 62) and then expressly leaves the detail to the Member States. This is the pathway where national law does the heavy lifting: you must check and follow the applicable national provisions in every Member State where the study will run.
At a glance: which article applies?
| Scenario | Device status | Within intended purpose? | Pathway | What you do |
| Generate evidence to CE-mark a device (first-in-human, pivotal) | Not CE-marked | n/a | Article 62 | Apply to the competent authority for authorisation, plus an ethics committee opinion (Articles 62 to 80) |
| Further assess a CE-marked device with extra invasive or burdensome procedures (PMCF) | CE-marked | Yes | Article 74(1) | Notify the competent authority at least 30 days before start; a subset of premarket rules applies |
| Assess a CE-marked device for a use it is not approved for | CE-marked | No | Article 74(2) | Full premarket regime applies (Articles 62 to 81, per Art. 74(2)) |
| Other investigations (e.g. investigator-initiated), not for conformity assessment and not 74(1) | Either | Typically yes | Article 82 | Meet the MDR minimum requirements plus national provisions in each Member State |
| Study uses a device but does not assess its safety or performance | Either | n/a | Not a clinical investigation under the MDR | Device must still be MDR-compliant; GDPR and national requirements still apply |
Why you still have to check national requirements
A point that catches many manufacturers out: national requirements apply both when your study sits squarely under the MDR and when it falls outside the EU MDR entirely. The MDR deliberately allows, and in some places requires, Member States to maintain their own national legislation – for example on ethics-committee review, the language of submission, fees, insurance and indemnity, and the detail of Article 82 studies. And where a study is not a clinical investigation under the MDR at all, national clinical-research, ethics and data-protection rules can still apply.
The practical consequence is that the same study can require different submissions in different countries. The European Commission maintains a consolidated overview, “National requirements for clinical studies” (the latest version, 1.1, was published in April 2026), which links to each Member State’s national websites and competent-authority contact points. It is the best single starting point for mapping country-by-country obligations.
Where the country-by-country differences bite
Study start-up timelines, in particular, can vary widely with national procedure. For an Article 62 study, some competent authorities require the ethics committee opinion to be in place before you can submit to the competent authority, while others allow the ethics committee and competent authority reviews to run in parallel – a difference that can add or save weeks or even months. Ethics review itself is structured differently from country to country: some operate a single central or regional ethics committee, others rely on local ethics committees at each participating site, and some sites additionally require their own local institutional approval on top of the ethics opinion.
Some countries also bring additional bodies into the process beyond the competent authority and ethics committee. In Germany, for example, the Federal Office for Radiation Protection (BfS) must be involved where the study uses ionising radiation beyond standard of care; in France, involvement of the data-protection authority (CNIL) and the national medical council (CNOM) may be required. Each of these carries its own requirements and timelines.
Getting it right from the start
Mapping a study to the correct regulatory pathway is one of the highest-leverage decisions when you are planning your clinical investigation. It determines whether you notify or apply for approval, how long approval takes, what documentation you need, and which national or local bodies you deal with. The distinctions between assessing a device and merely using one, between within and outside intended purpose, and between notification and authorisation, are precisely where you could lose time and money.
At Franklyn Health, we help manufacturers and sponsors make these determinations with confidence, design studies that fit the right pathway, and manage the competent-authority and ethics submissions that follow, across the EU and beyond. Whether you are running your first feasibility study or planning PMCF on a marketed device, we can help you find the most efficient compliant route.
Contact our team to discuss your clinical investigation strategy under the EU MDR.
References
Regulation (EU) 2017/745 – EU Medical Device Regulation (EUR-Lex)
MDCG 2021-6 Rev.1 – Questions & Answers regarding clinical investigation (European Commission)
National requirements for clinical studies (European Commission)
Franklyn Health – EU MDR: Where It Applies and Who Needs Representation
BfS – Applications of ionising radiation in medicine (Germany)
CNIL – Reference methodology MR-001 for health research (France)
CNOM – Conventions with industry under the anti-gift law (France)
This article is provided for general information only and does not constitute legal or regulatory advice. Always confirm requirements against the current text of Regulation (EU) 2017/745, the applicable MDCG guidance, and national provisions.
