Clinical Evaluation & the CER
Clinical evaluation is the systematic, ongoing process of generating, appraising, and analysing the clinical data on a medical device to verify its safety and performance, and the Clinical Evaluation Report (CER) is its documented output. Under the EU Medical Device Regulation it is a core obligation for essentially every device, and it trips up manufacturers for a structural reason: the guidance everyone uses to *write* the CER and the regulation that *requires* it are two different documents from two different eras. This page covers what clinical evaluation is and how those pieces fit; the post-market data that feeds it is the [complaint handling & vigilance](/topics/complaint-handling-and-vigilance) explainer.
What an explainer is not
A topic explainer is SPEQ’s synthesis of what a practice involves, cited to the standards that govern it. It does not reproduce their text, and it does not determine which of them apply to your product or process.
[ POSITION IN THE FRAMEWORK ]
7 DIMENSIONS · 20 LINKSClinical evaluation is the continuous, appraised argument that a device is safe and performs — written to MEDDEV 2.7/1 rev 4's method while meeting the stricter EU MDR Annex XIV, and never finished.
06 · QUALITY MATURITY — CLINICAL EVALUATION & THE CER, REACTIVE TO ADAPTIVE
The CER is a one-time literature dump assembled for CE-marking and then left to age untouched.
A CER procedure exists, but it is written to the old Directive and updated only when a certificate renews.
The CER is structured to MEDDEV 2.7/1 rev 4 and satisfies MDR Annex XIV, with appraised evidence and justified equivalence.
PMS and PMCF feed the CER on a risk-linked cadence; emerging market data updates the benefit-risk before regulators ask.
Clinical evaluation, PMS, and PMCF run as one living loop that continuously demonstrates the device remains safe and effective.
SPEQ’s shared five-stage progression, labelled synthesis — not the FDA QMM rating scale. Where does your organization sit? Score your quality system →
07 · REGULATORY & EVIDENCE
GOVERNING STANDARDS · 3
Derived from the 3 standards SPEQ maps to this subject, across 2 regulatory bodies: ISO, EC.
RECORDS & OBJECTIVE EVIDENCE
- A clinical evaluation plan and CER documenting data identification, appraisal, analysis, and conclusion
- Justification of any equivalence claim, with access to the equivalent device's technical documentation
- A PMCF plan and reports generating new clinical data across the lifecycle
- Risk-linked CER update records tied to the post-market surveillance system
- A benefit-risk determination consistent with the ISO 14971 risk file
COMMON INSPECTION FINDINGS
- A CER written to the Medical Device Directive rather than the EU MDR
- CER static since CE-marking while years of market data accumulated
- Equivalence claimed without access to the comparator's technical documentation
- The CER not coupled to post-market surveillance or PMCF
- Clinical claims unsupported by appraised evidence in the report
What clinical evaluation is
Clinical evaluation is not a single study — it is a *methodological process* of assembling all the relevant clinical evidence for a device and appraising whether it demonstrates that the device is safe, performs as intended, and that its benefits outweigh its risks for its intended purpose. The evidence can come from several routes: clinical investigations of the device itself, clinical data from an equivalent device (where equivalence is rigorously justified), and the published scientific literature. The evaluator gathers this body of evidence, critically appraises the quality and relevance of each source, and analyses whether, taken together, it supports the clinical claims made for the device.
The output is the **Clinical Evaluation Report**, which documents the evaluation plan, the data identified, the appraisal, the analysis, and the conclusion — the evidence trail that a notified body (or the FDA, in its own framework) reviews to judge whether the clinical claims are substantiated. A CER is therefore an argument backed by appraised evidence, not a literature dump: the appraisal step, judging how much weight each data source genuinely carries for *this* device and *this* claim, is where a credible CER is distinguished from a padded one.
Two documents, two eras — the structural confusion
The recurring source of trouble is that the how-to and the legal requirement come from different places. **MEDDEV 2.7/1 Revision 4** (2016) is the detailed guidance manufacturers use to *structure and conduct* a clinical evaluation — the stages, the appraisal method, what a good CER contains — but it was published under the older Medical Device Directive. The legal *requirement* now sits in the **EU MDR (Regulation 2017/745), Annex XIV Part A**, which sets what clinical evaluation and the CER must satisfy, alongside stricter expectations on clinical evidence, equivalence, and post-market updating. The bridging guidance from the MDCG (notably MDCG 2020-6) maps the two together.
The practical effect is that a compliant CER must be written to MEDDEV 2.7/1 rev 4’s method while satisfying the MDR’s (higher) requirements — using the older document for *how* and the regulation for *what*. Two MDR-specific tightenings matter most: the bar for claiming **equivalence** to another device (to borrow its clinical data) is now much harder to meet, often requiring contractual access to the equivalent device’s technical documentation; and clinical evidence expectations rose across the board, so evidence that sufficed under the Directive may not suffice under the Regulation. Writing a CER as if the Directive still governed is a common and costly error.
The CER is never finished
A defining feature of clinical evaluation under the MDR is that it is **continuous, not a one-time pre-market exercise**. The CER must be actively updated throughout the device’s lifecycle with data from the market, and it is tightly coupled to the post-market system: **post-market surveillance (PMS)** feeds real-world safety and performance data in, and **post-market clinical follow-up (PMCF)** proactively generates new clinical data to keep the evidence current and close gaps the pre-market evaluation could not. The CER, the PMS plan, and the PMCF plan form a loop, each feeding the others.
The update cadence is risk-linked: higher-risk devices (and implantables) require more frequent CER updates and are subject to periodic safety update reporting, while lower-risk devices update on a longer cycle — but "update when convenient" is not an option for any of them. A CER that was excellent at CE-marking and then sat static while years of market data accumulated is a finding, because the regulation treats clinical evaluation as a living demonstration that the device *remains* safe and effective in real use, not a historical claim made once at launch.
Clinical evaluation in the device quality system
Clinical evaluation is woven through the device quality system rather than being a standalone regulatory deliverable. Its conclusions depend on and feed the ISO 14971 risk management file (the benefit-risk determination is shared), it draws its post-market inputs from vigilance and PMS, and it informs — and is informed by — the design and its intended-purpose claims. A change to the device, its indications, or its risk profile can require the clinical evaluation to be revisited, which ties it into change control.
The through-line is that the CER is where a manufacturer’s clinical claims stop being assertions and become an appraised, evidence-backed, continuously maintained argument. Its two most common failure modes are structural — writing it to the Directive rather than the Regulation, and treating it as a pre-market document that is then left to age — and both stem from misreading clinical evaluation as a one-time hurdle rather than the lifecycle evidence discipline the MDR makes it. Getting the two-document structure and the continuous-update obligation right is most of what separates a compliant clinical evaluation from a vulnerable one.
FREQUENTLY ASKED
What is a Clinical Evaluation Report (CER)?
The documented output of clinical evaluation — the systematic process of generating, appraising, and analysing the clinical data on a device to verify its safety and performance and that its benefits outweigh its risks. The CER documents the evaluation plan, the data (from clinical investigations, an equivalent device, or the literature), the critical appraisal, the analysis, and the conclusion. It is an evidence-backed argument, not a literature dump.
Why do a MEDDEV guidance and the EU MDR both apply to a CER?
Because the how-to and the legal requirement come from different documents. MEDDEV 2.7/1 Revision 4 (2016, under the old Medical Device Directive) is the detailed guidance for structuring and conducting the evaluation; the legal requirement now sits in EU MDR Annex XIV Part A, with stricter expectations on evidence, equivalence, and updating. A compliant CER is written to the MEDDEV method while satisfying the MDR’s higher requirements, with MDCG 2020-6 bridging them.
Is a CER a one-time document?
No. Under the MDR clinical evaluation is continuous: the CER must be actively updated throughout the device’s life with market data, coupled to post-market surveillance (which feeds real-world data in) and post-market clinical follow-up (which proactively generates new clinical data). Update frequency is risk-linked — more frequent for higher-risk and implantable devices — but a CER left static after CE-marking is a finding.
Why is claiming equivalence harder under the MDR?
The MDR raised the bar for claiming equivalence to another device in order to borrow its clinical data — the technical, biological, and clinical characteristics must match closely, and it often requires contractual access to the equivalent device’s technical documentation. Evidence and equivalence claims that sufficed under the Directive frequently do not under the Regulation, which is why writing a CER as if the Directive still governed is a costly error.