Complete Guide to Clinical Evaluation under EU MDR 2017/745
The definitive resource covering every aspect of clinical evaluation for medical device manufacturers. From Clinical Evaluation Plan to CER, from SOTA analysis to PMCF. everything you need to achieve and maintain CE marking under the Medical Device Regulation.
Table of Contents
- What Is Clinical Evaluation?
- Regulatory Framework
- Clinical Evaluation Plan (CEP)
- State of the Art (SOTA)
- Literature Review Methodology
- Clinical Evaluation Report (CER)
- Equivalence Assessment
- Benefit-Risk Determination
- Post-Market Clinical Follow-up (PMCF)
- CER Update Cycle
- Common Notified Body Deficiencies
- Frequently Asked Questions
1. What Is Clinical Evaluation under EU MDR?
Clinical evaluation is a systematic and planned process to continuously generate, collect, analyse, and assess the clinical data pertaining to a medical device. Its purpose is to verify the safety and performance of the device, including its clinical benefits, when used as intended by the manufacturer.
Under EU MDR 2017/745, clinical evaluation is defined in Article 2(44) and its requirements are set out in Article 61 and Annex XIV. Unlike the previous Medical Device Directive (MDD/93/42/EEC), the MDR introduces significantly more demanding requirements for clinical evidence, particularly regarding:
- State-of-the-art analysis demonstrating the device performs at least as well as current alternatives
- Equivalence claims stricter criteria under Article 61(4)-(5) with contract requirements
- Benefit-risk quantification quantitative comparison against defined acceptance criteria
- Continuous updating clinical evaluation must be ongoing throughout the device lifecycle
- PMCF integration mandatory post-market clinical follow-up to confirm pre-market conclusions
Who Must Perform Clinical Evaluation?
Every manufacturer placing a medical device on the EU market must perform clinical evaluation, regardless of device class. There are no exemptions under MDR. Even Class I devices under Article 61(10) must demonstrate through clinical evaluation that they conform to the applicable general safety and performance requirements, although the depth of evidence may be proportionate to the risk profile.
2. Regulatory Framework
The clinical evaluation requirements under MDR are distributed across several regulatory instruments:
MDR Article 61. Clinical Evaluation
Article 61 is the primary legal basis for clinical evaluation. It establishes that manufacturers shall plan, conduct, and document a clinical evaluation based on clinical data that provides sufficient clinical evidence. The article addresses:
- Article 61(1) General requirement for clinical evaluation for all devices
- Article 61(3) Clinical evaluation must follow a defined and justified procedure
- Article 61(4)-(5) Conditions for using equivalent device data (Class III, implantables)
- Article 61(6) Conditions for using equivalent device data (other classes)
- Article 61(10) Simplified provisions for Class I and specific devices
- Article 61(11) Requirement for PMCF as part of clinical evaluation
Annex XIV. Clinical Evaluation and PMCF
Part A covers the Clinical Evaluation Plan (CEP) and Clinical Evaluation Report (CER). Part B covers Post-Market Clinical Follow-up (PMCF), including the PMCF plan and PMCF evaluation report.
Key MDCG Guidance Documents
| Guidance | Title | Focus |
|---|---|---|
| MDCG 2020-5 | Clinical Evaluation – Equivalence | Demonstrating equivalence under Article 61(4)-(6) |
| MDCG 2020-6 | Sufficient Clinical Evidence | What constitutes sufficient clinical evidence for MDR |
| MDCG 2020-7 | PMCF Plan Template | Structure and content of the PMCF plan |
| MDCG 2020-8 | PMCF Evaluation Report Template | Structure and content of the PMCF evaluation report |
| MDCG 2020-13 | Clinical Evaluation Assessment Report | How Notified Bodies assess clinical evaluations (CEAR) |
| MDCG 2020-1 | Clinical Evaluation – Guidance | General guidance on clinical evaluation under MDR |
3. Clinical Evaluation Plan (CEP)
The Clinical Evaluation Plan is the strategic foundation document that defines the scope, methodology, and acceptance criteria for the entire clinical evaluation process. Required by MDR Annex XIV Part A, the CEP must be established before clinical data is collected or analysed.
A well-structured CEP defines what clinical questions need to be answered, what evidence is required, and what benchmarks will be used to determine if the device meets its intended purpose safely and effectively.
What the CEP Must Include
- Device description, intended purpose, and target population
- Clinical claims and performance parameters to be evaluated
- Literature search protocol (databases, search strings, inclusion/exclusion criteria)
- Equivalence route justification (if applicable)
- State-of-the-art methodology and acceptance criteria
- PMCF strategy overview
- Timeline for CER updates
4. State of the Art (SOTA)
The State of the Art analysis is arguably the most misunderstood component of clinical evaluation under MDR. SOTA is not a literature review. It is a structured analysis that defines the current knowledge about the clinical condition, available treatment alternatives, and accepted benefit-risk parameters.
The purpose of SOTA is to establish the benchmarks against which your device’s performance will be judged. Without a proper SOTA analysis, your benefit-risk determination has no reference point, and your CER’s conclusions cannot be scientifically justified.
SOTA vs Literature Review
| Aspect | State of the Art | Literature Review |
|---|---|---|
| Purpose | Define benchmarks and acceptance criteria | Collect clinical evidence about the device |
| Scope | Disease area, all treatment alternatives | Specific device and equivalents |
| Output | Benefit-risk parameters, acceptance thresholds | Appraised clinical data for analysis |
| Sources | Guidelines, registries, meta-analyses, consensus | Device-specific clinical studies, PubMed, Embase |
5. Literature Review Methodology
The systematic literature review is a critical component of the CER. Under MDR, the literature search must be systematic, reproducible, and documented. Notified Body reviewers routinely repeat your searches to verify completeness.
PICO Framework
The recommended approach uses the PICO framework to structure search queries:
- P (Population). target patient population, disease/condition
- I (Intervention). the device under evaluation
- C (Comparison). alternative treatments, equivalent devices
- O (Outcome). safety endpoints, performance endpoints, clinical benefit measures
Database Requirements
MDR does not prescribe specific databases, but best practice requires searching at least two independent databases. The standard combination is:
- PubMed/MEDLINE biomedical literature
- Embase broader coverage of European and pharmaceutical literature
- Cochrane Library systematic reviews and clinical trials
- Device-specific registries and regulatory databases (MAUDE, BfArM, MHRA)
6. Clinical Evaluation Report (CER)
The Clinical Evaluation Report is the single most scrutinised document in your MDR technical documentation. It presents the findings of the clinical evaluation process, demonstrating that the device achieves its intended purpose with an acceptable benefit-risk profile based on the current state of the art.
The CER must include: device description, clinical background, SOTA analysis, equivalence assessment (if applicable), literature review findings, clinical data analysis, benefit-risk determination, and conclusions on conformity with relevant General Safety and Performance Requirements (GSPRs).
MDCG 2020-13 Alignment
Notified Bodies use the MDCG 2020-13 checklist when reviewing CERs as part of the Clinical Evaluation Assessment Report (CEAR). This guidance defines exactly what reviewers check, making it the de facto standard for CER structure and content.
7. Equivalence Assessment
Claiming equivalence allows manufacturers to use clinical data from a similar existing device (the “equivalent device”) instead of conducting their own clinical investigation. However, MDR has made equivalence significantly harder to establish compared to MDD.
Three Pillars of Equivalence
Under Article 61(4)-(5), equivalence must be demonstrated across all three dimensions:
- Technical equivalence same design, specifications, materials, surfaces, manufacturing
- Biological equivalence same materials in contact with the same tissues/body fluids
- Clinical equivalence same intended purpose, clinical condition, site in the body, target population, clinical effects, duration of use
Contract Requirement
For Class III and implantable devices (Article 61(5)), the manufacturer must have a contract with the equivalent device manufacturer granting sufficient access to the technical documentation. This is a strict requirement that effectively prevents most “predicate device” approaches commonly used under MDD.
8. Benefit-Risk Determination
The benefit-risk determination is the central conclusion of the CER. It must demonstrate that the residual risks associated with the device are acceptable when weighed against the clinical benefits, and that the device’s performance is at least comparable to the current state of the art.
Quantification Requirement
MDR requires benefit-risk to be quantified, not merely stated qualitatively. This means:
- Defining measurable benefit parameters (e.g, complication rate, healing time, diagnostic accuracy)
- Extracting benchmark values from the SOTA analysis
- Comparing device data against these benchmarks using statistical methods where possible
- Demonstrating the device meets or exceeds defined acceptance criteria
9. Post-Market Clinical Follow-up (PMCF)
PMCF is a continuous, proactive process to collect and evaluate clinical data from the use of an already CE-marked device. Required by Article 61(11) and defined in Annex XIV Part B, PMCF confirms the long-term safety, performance, and benefit-risk profile established during pre-market clinical evaluation.
PMCF Plan vs PMCF Report
The PMCF Plan defines what data will be collected, through which methods, with what endpoints, and over what timeframe. The PMCF Evaluation Report presents the results of those activities and their impact on the clinical evaluation conclusions.
Common PMCF Activities
- Clinical registries and multi-centre observational studies
- Post-market clinical studies (interventional or non-interventional)
- Surveys and questionnaires to users and patients
- Ongoing literature monitoring and surveillance
- Analysis of complaint data, vigilance reports, and FSCA trends
10. CER Update Cycle
MDR requires the CER to be updated at regular intervals throughout the device lifecycle. The update frequency depends on the device class and risk profile:
- Class III and implantable devices at least annually
- Class IIa and IIb devices at least every 2–5 years, depending on the risk profile and PMCF findings
- Class I devices when significant new data becomes available
An update is also triggered by significant changes: new PMCF data contradicting pre-market conclusions, safety signals from vigilance data, new publications challenging the state of the art, or device modifications affecting clinical performance.
11. Common Notified Body Deficiencies
Based on published MDCG guidance and real-world Notified Body feedback, these are the most frequently cited deficiencies in clinical evaluation submissions:
Weak SOTA Analysis
Treating SOTA as a literature dump instead of structured analysis that defines benefit-risk parameters and acceptance criteria.
Unsupported Equivalence
Claiming equivalence without demonstrating all three pillars (technical, biological, clinical) or without contract access to the equivalent device data.
Incomplete Literature Search
Using only one database, not documenting the search protocol, or failing to capture key publications that reviewers find independently.
No Benefit-Risk Quantification
Stating benefits qualitatively without quantifying them against benchmarks or comparing with alternative treatments.
Risk-CER Disconnect
Failing to link adverse events identified in literature to the risk management file, or not cross-referencing ISO 14971 outputs.
Missing PMCF Integration
PMCF plan that looks like generic PMS without device-specific endpoints, sample size justification, or link to CER conclusions.
Outdated Clinical Data
CER not updated with recent literature, PMCF results, or vigilance data despite regulatory requirement for continuous evaluation.
Vague CEP Scope
Acceptance criteria too broad or unmeasurable, making it impossible to objectively determine if the device meets performance expectations.
12. Frequently Asked Questions
What is clinical evaluation under EU MDR 2017/745?
Clinical evaluation is a systematic and planned process to continuously generate, collect, analyse, and assess clinical data pertaining to a medical device. Under MDR, it is required for all device classes and must demonstrate safety, performance, and an acceptable benefit-risk profile throughout the device lifecycle.
Is clinical evaluation required for all medical device classes?
Yes. MDR Article 61 requires clinical evaluation for all device classes, including Class I. Article 61(10) provides simplified provisions for certain lower-risk devices, but does not exempt them from clinical evaluation entirely.
What is the difference between clinical evaluation and clinical investigation?
Clinical evaluation is the overarching process of assessing all available clinical data (from literature, equivalent devices, clinical investigations, PMCF, etc.). Clinical investigation is one possible source of clinical data. a prospective study conducted on human subjects according to ISO 14155. Not all devices require a clinical investigation, but all require clinical evaluation.
How is State of the Art (SOTA) different from a literature review?
SOTA defines the current knowledge about the clinical condition and treatment alternatives, establishing benefit-risk parameters and acceptance criteria. The literature review is a systematic search for device-specific clinical evidence. SOTA sets the benchmarks; the literature review provides the data to measure against those benchmarks.
When is equivalence acceptable under MDR?
Equivalence is acceptable when the manufacturer can demonstrate technical, biological, and clinical similarity between their device and an existing device. For Class III and implantable devices (Article 61(5)), a contract with the equivalent device manufacturer granting access to technical documentation is mandatory.
What does MDCG 2020-13 require?
MDCG 2020-13 defines how Notified Bodies should assess clinical evaluations through the Clinical Evaluation Assessment Report (CEAR). It provides a structured checklist covering SOTA adequacy, literature search completeness, equivalence validity, benefit-risk determination, and PMCF plan appropriateness.
How often must the CER be updated?
For Class III and implantable devices, the CER must be updated at least annually. For other classes, updates should occur at appropriate intervals determined by the manufacturer based on risk profile, PMCF data, and any new clinical evidence. Updates are also triggered by safety signals, significant design changes, or new MDCG guidance.
Is PMCF mandatory for Class I devices?
Article 61(11) requires PMCF for all devices. For Class I devices, the PMCF plan should be proportionate to the risk profile. Exemption from clinical investigation does not mean exemption from PMCF. The manufacturer must still monitor long-term safety and performance in the field.
Can I use clinical data from outside Europe?
Yes, but with conditions. Clinical data from non-EU studies must be assessed for relevance and transferability to the EU population. Factors such as ethnicity, standard of care, healthcare system differences, and regulatory context must be considered. FDA 510(k) clearance alone does not constitute sufficient clinical evidence for MDR.
What happens if my clinical evaluation has gaps?
Clinical evidence gaps must be identified in the CER and addressed through the PMCF plan. The gap analysis chapter should clearly state what evidence is missing, why, and how it will be generated. If gaps are critical (affecting benefit-risk conclusions), a clinical investigation may be required before CE marking can be achieved.
About the Author
Start Your Clinical Evaluation Journey
Join the Clinical Evaluation Navigator community to get weekly expert guidance on CER writing, SOTA analysis, PMCF planning, equivalence assessment, and MDR compliance. Ask questions directly to Dr. Hatem Rabeh and connect with regulatory professionals worldwide.
