Clinical evaluation is one of the MDR requirements that most surprises manufacturers approaching the transition without support. Not because it is new in principle — the directive already required it. But because the MDR fundamentally changes its nature: from a one-off documentary formality, it becomes a continuous, structured process that is substantially more demanding in terms of data.
What Article 61 actually requires
Article 61 defines clinical evaluation as a systematic and planned process for the generation, collection, analysis and assessment of the clinical data relating to a device, in order to verify the safety and clinical performance of the device when used as intended.
Three words in this definition deserve to be highlighted.
“Process”. Not a report. Not a document. A process — that is, a continuous, planned, documented activity, with inputs, outputs, responsibilities and milestones.
“Continuous”. Article 61(11) specifies that the manufacturer must update the clinical evaluation throughout the life cycle of the device. For implantable and class III devices, an update at least annually is expected. For others, the frequency must be justified in the clinical evaluation plan.
“Substantial”. The level of evidence expected has increased significantly compared with the directive. Clinical data must be relevant, representative of the target population, and of documented methodological quality.
To whom clinical evaluation applies
Article 61 applies to all medical devices, with no exemption by class. A class I device is not exempt from clinical evaluation. The difference with higher classes lies in the level of evidence required and the update frequency — not in the existence of the obligation.
This is one of the most common misunderstandings: “we are class I, so we don’t need a clinical evaluation”. That is incorrect. A class I device must have a documented clinical evaluation. It may be based on a literature review proportionate to the risk, but it must exist.
The two pathways for collecting clinical data
Pathway 1: clinical data specific to the device. Clinical investigations carried out on the device itself (Article 62 MDR), retrospective data from post-market surveillance, registry data, data from systematised collection under the PMCF.
Pathway 2: clinical data from an equivalent device. The manufacturer demonstrates equivalence with an existing device according to the three cumulative criteria of Article 61(5), then relies on the clinical data of that equivalent device. For a device belonging to another manufacturer, contractual access to the technical data is required — which makes this pathway practically inaccessible with a competitor.
What “insufficient clinical data” means for a file
A notified body that concludes that the clinical data are insufficient is not rejecting the file on a point of form. It is pointing to a substantive gap: the available evidence does not allow the conclusion that the clinical benefits of the device outweigh the residual risks under real-world conditions of use. The only way to fill this gap is to produce additional data — which takes time and costs money.
Anticipating clinical gaps from the very design of the file — and building the PMCF plan accordingly — is the smartest way to approach MDR clinical evaluation.
Regulatory source: Article 61 and Annex XIV of Regulation (EU) 2017/745 — EUR-Lex