The Clinical Evaluation Report (CER) is the documented summary of the clinical evaluation of a device. It gathers all available clinical data, analyses them critically and in a structured manner, and concludes on the device’s compliance with the clinical requirements of the MDR — in particular on the benefit/risk ratio and compliance with the applicable GSPR.
It is the central clinical document of the technical documentation. And it is often the document that generates the most exchanges with notified bodies during the assessment.
The structure of the CER according to MEDDEV 2.7/1 rev 4
The guidance MEDDEV 2.7/1 rev 4, published in 2016, remains the most detailed methodological reference for drafting the CER, even under the MDR. It proposes a five-stage structure.
Stage 1 — Scope. Definition of the scope of the evaluation: device concerned, intended use, target population, inclusion and exclusion criteria for clinical data.
Stage 2 — Identification of clinical data. Systematic literature review according to a documented and reproducible strategy. Identification of available post-market surveillance data. Clinical investigation data where these exist.
Stage 3 — Appraisal of data quality. Each data source is appraised for its methodological quality: level of evidence, risk of bias, relevance to the device under evaluation.
Stage 4 — Analysis and synthesis. Synthesis of safety and performance data. Identification of residual clinical gaps. Conclusion on the benefit/risk ratio.
Stage 5 — Conclusion and PMCF plan. Formal conclusion on compliance with the applicable GSPR. PMCF plan to address the identified gaps.
The expected level of evidence by class
Class I: a documented literature review covering the safety and performance data available for comparable devices. The demonstration that the device meets the GSPR may rely mainly on preclinical data and literature data, where the risk profile justifies it.
Class IIa: substantial clinical data, derived from the literature or from device-specific data. Demonstration of equivalence is possible if the three criteria of Article 61.5 are rigorously documented. Notified bodies examine the methodological quality of the studies selected.
Class IIb: direct clinical data are generally necessary, or a very robust demonstration of equivalence with a device from the same manufacturer. Notified bodies examine the CER in greater depth for class IIb than for IIa.
Class III: device-specific clinical data are in most cases indispensable. A literature review alone is rarely sufficient.
The lifespan of the CER
The CER has no fixed expiry date. It must be updated whenever new data become available — post-market data, new clinical publications on comparable devices, new guidelines — and at least at the frequency defined in the CEP.
For implantable and class III devices: annual update as a minimum. For classes IIa and IIb: every two to three years depending on the risk profile and the stability of the data.
A CER that has not been updated for several years while post-market data exist is a warning sign for notified bodies. It signals a clinical evaluation process on standby, not a continuous process.