MDR clinical evaluation: what Article 61 really requires

Clinical evaluation is one of the most structuring requirements of the MDR 2017/745. It concerns all manufacturers, whatever the device class, and constitutes a living document — not a one-off exercise. Here is what Article 61 concretely requires, and how to approach it without getting lost.

Why MDR clinical evaluation differs from the directive

Under Directive 93/42/EEC, clinical evaluation was often reduced to a cursory literature review, sometimes delegated to a consultant who produced a generic report in a few days. The MDR changed the game on three points:

1. Clinical equivalence is very restrictive — you must demonstrate that the equivalent device is identical or similar in its clinical, technical and biological characteristics, and have contractual access to the equivalent manufacturer's data (except for class I devices).
2. The clinical evaluation report (CER) must be updated at least once a year for class III and implantable IIb devices, and according to PMCF data for other classes.
3. PMCF (Post-Market Clinical Follow-up) is mandatory — not optional. If you choose not to conduct a PMCF study, you must justify that decision in a documented way.

What Article 61 says, word for word

Article 61 of the MDR sets out several key obligations:

• The clinical evaluation must follow a defined and methodologically sound procedure (Article 61(1))
• It must be based on sufficient clinical data to assess the safety and performance of the device (Article 61(3))
• For class III devices and certain implantable IIb devices, a clinical investigation is mandatory unless existing clinical data suffices (Article 61(4))
• Equivalence data must be justified according to the criteria of Annex XIV, Part A (Article 61(5))

The European Commission published MDCG 2020-5 (guidance on clinical evaluation) and MDCG 2020-6 (PMCF guidance) to clarify these requirements.

The three routes to building clinical data

Route 1: Device-specific clinical data

This is the most robust route. It includes:

• Results of clinical investigations conducted specifically on the device
• Data from post-market surveillance (PMS)
• Documented clinical experience feedback

Route 2: Clinical equivalence

You demonstrate that your device is equivalent to another for which clinical data exists. The three dimensions to demonstrate simultaneously:

Dimension	What must be demonstrated
Clinical	Same indication, same population, same anatomical site, same conditions of use
Technical	Same operating principle, same design characteristics (materials, dimensions, surface properties)
Biological	Same type and intensity of contact with tissues, same substances released

⚠️ Critical point: if the equivalent device belongs to another manufacturer, you must have a contractual agreement giving you access to their technical and clinical data. In practice this is nearly impossible except in the context of mergers and acquisitions — which pushes most manufacturers towards route 1 or 3.

Route 3: Scientific literature data

A literature review remains valid, but it must be systematic (documented search protocol, databases queried, inclusion/exclusion criteria for studies, assessment of methodological quality). Copy-pasting a few PubMed abstracts is no longer enough.

The structure of an MDR-compliant CER

A clinical evaluation report (CER) compliant with the MDR and EN ISO 14155 typically includes:

1. Device scope — description, indication, target population, intended users
2. Clinical evaluation plan — methodology, data sources, acceptability criteria
3. Review of safety and performance data — preclinical data, clinical data (own or equivalence), literature
4. Equivalence analysis (if applicable) — justification of the three dimensions
5. Benefit/risk balance assessment — linked to the ISO 14971 risk analysis
6. Conclusions — sufficiency of clinical data, residual uncertainties
7. PMCF plan — or justification for the absence of a PMCF investigation
8. References — exhaustive list of sources

PMCF: continuous clinical evaluation after placing on the market

PMCF is not optional. Annex XIV Part B of the MDR defines the acceptable methods:

• PMCF studies (registries, cohort studies, case-control studies)
• PMCF surveys (clinician user questionnaires)
• Reviews of complaint/PMS data
• Ongoing post-market literature review
• Disease or procedure registries

PMCF results must feed the PSUR (Periodic Safety Update Report) for class IIa, IIb and III devices, and the annual safety report for class I devices.

The link with the SSCP

For class III and implantable IIb devices, the SSCP (Summary of Safety and Clinical Performance) is mandatory. It is a public document, accessible via EUDAMED, that summarises in accessible language:

• The clinical performance of the device
• Adverse incidents and corrective measures
• The conclusions of the latest clinical evaluation

It must be updated at least once a year for class III, every two years for implantable IIb. Its drafting flows directly from the CER.

Common mistakes that notified bodies penalise

Based on published audit reports and field feedback:

1. CER not updated — treated as a frozen document when it must evolve with PMCF data and market surveillance
2. Equivalence not justified across all three dimensions — the biological dimension is often neglected
3. No documented PMCF plan — or a PMCF plan without a schedule or designated owner
4. Literature data without a search protocol — the included studies are not justified
5. CER not signed by a qualified clinical evaluator — the MDR requires evaluators to have adequate qualifications (Annex XIV, section 1.1)

What MedQara monitors for you

Regulatory monitoring around clinical evaluation is particularly active right now:

• MDCG 2023-x — updates to PMCF and CER guidance
• SCENIHR and new harmonised standards for clinical evaluation by specialty
• Notified body decisions published on EUDAMED
• MDR amendments concerning transition deadlines for legacy devices

The platform automatically classifies these publications by criticality and provides a contextual analysis for each update — without you having to manually monitor the 12 official sources.

---

Need structured monitoring on clinical evaluation and PMCF? Try MedQara for free — 5 alerts per week, no credit card.