Clinical evaluation demonstrates, with clinical data, that your device is safe and performs as intended for its intended purpose (MDR Article 61 and Annex XIV). The data may come from your own clinical investigations, from scientific literature or from an equivalent device. The report is not a one-off: it is updated regularly with data from post-market surveillance.
How is a clinical evaluation built?
Everything starts with writing the intended purpose clearly: what the device does, in which patient group, and which clinical benefit you claim. Those claims are tied to measurable acceptance criteria. The clinical evaluation plan (CEP) defines up front which data will be collected, by which method, and how it will be appraised.
Data is then gathered: scientific literature, your own clinical investigations, data on an equivalent device and post-market experience. Each source is appraised separately for relevance, methodological quality and contribution to the claim. The resulting report has to base the benefit-risk balance on evidence.
What it covers
- Clinical evaluation plan (CEP) and report (CER)
- A reproducible literature search and appraisal of the data
- Equivalence analysis (MDCG 2020-5)
- PMS plan, PMS report (class I) and PSUR (classes IIa, IIb and III)
- PMCF plan, surveys and registry studies, PMCF evaluation report
- Summary of safety and clinical performance (SSCP)
How often is it updated?
Clinical evaluation, PMS and PMCF feed each other. The PSUR is updated at least annually for class IIb and III devices and at least every two years for class IIa devices. For class III and implantable devices the PMCF evaluation report and the SSCP are also updated at least annually. The clinical evaluation report is reviewed together with that data.
Link with post-market surveillance
Clinical evaluation does not stand alone. The post-market surveillance plan defines which data is collected and how often: complaints, service records, literature, safety information on similar devices. That data flows in two directions — it updates the clinical evaluation and it tests the assumptions in the risk management file.
Post-market clinical follow-up (PMCF) collects the clinical part of that data in a planned way, for example through surveys, registry studies or follow-up investigations. If PMCF is not needed, the justification has to be written down; it is one of the first things a notified body asks about.
When is a clinical investigation required?
For implantable and class III devices the rule is that clinical investigations are performed (MDR Article 61(4)). The exceptions are narrow: if the device is a modification of a device already marketed by the same manufacturer and the notified body has endorsed the demonstration of equivalence, an investigation may not be required.
The second exception is the list of well-established technologies in Article 61(6). It began with sutures, staples, dental fillings, dental braces, tooth crowns, screws, wedges, plates, wires, pins, clips and connectors; Commission Delegated Regulation (EU) 2026/1451 added products such as cannulas, feeding tubes, suture pledgets and bone wax. Being on the list does not remove the obligation to carry out a clinical evaluation — only the obligation to investigate.
Where an investigation is required, it is designed and run to ISO 14155. In Türkiye the Medical Device Clinical Investigations Regulation (Official Gazette, 8 July 2022, No. 31890) requires TİTCK authorisation and ethics committee approval; the protocol, investigator’s brochure, case report form and insurance certificate are inseparable parts of the application. The investigation timetable is often the longest item in the project and should be planned together with the classification decision.
Where demonstrating conformity on the basis of clinical data is not deemed appropriate (Article 61(10)), the justification is written on the basis of the risk management output, the specifics of the interaction between the device and the human body, the intended clinical performance and the manufacturer’s claims. That route is narrow and it does not remove the investigation requirement for devices covered by Article 61(4).
Completing clinical data for legacy devices
For devices certified under the MDD, the most frequent problem is that the available data does not reach the level the MDR expects. MDCG 2020-6 explains what counts as sufficient clinical evidence for these devices and ranks the sources of data from strongest to weakest: well-designed clinical investigations at the top, individual case reports and expert opinion at the bottom.
The route can be summed up in three questions: what do you have, what does the MDR ask for, and how is the gap closed. Missing clinical evidence is usually closed with a planned PMCF activity — a survey, a registry study or a follow-up investigation. The earlier the gap analysis is done, the fewer surprises there are in the certification timetable.
Clinical data also feeds the risk management file: adverse effects seen in clinical data have to be reflected in the risk file, and harms in the risk file have to be discussed in the clinical evaluation.
Who writes the report?
The competence of the people performing the clinical evaluation is part of the file. MEDDEV 2.7/1 rev. 4 asks for the evaluators’ curriculum vitae and declaration of interest: knowledge of the relevant medical field, of research methodology and of the regulatory framework. Where one person does not cover all three, a team is formed and it is recorded who appraised what.
The manufacturer’s own staff may do the work; there is no requirement for independence, but there has to be evidence that the appraisal was carried out objectively. The most frequent omissions are a missing declaration of interest and CVs that were never filed.
Common problems
A literature search that cannot be reproduced is the most frequent finding. If the database, search terms, date range and the reason each paper was excluded are not recorded, the search is not accepted.
The second frequent problem is an equivalence claim that is not sufficiently demonstrated: technical, biological and clinical characteristics all have to be compared, and differences discussed in terms of their clinical effect. The third is leaving the claim that PMCF is not needed without justification.
