CER Deficiencies That Trigger Notified Body Questions

CER deficiencies are one of the main reasons Clinical Evaluation Reports receive Notified Body questions under EU MDR. Many CERs fail review not because clinical data is missing, but because the evidence story is incomplete, inconsistent, poorly justified, or disconnected from the technical documentation.

Does the clinical evidence actually support the device’s intended purpose, claims, risks, and benefit-risk conclusion?

This is where many CERs become weak.

Under EU MDR, a CER is not just a literature summary. It is a structured clinical evidence argument. It should clearly connect the device, intended purpose, clinical claims, clinical risks, available evidence, State of the Art, PMS, PMCF, and benefit-risk conclusion.

A CER should support the device’s intended purpose, clinical safety, clinical performance, claimed clinical benefits, relevant GSPRs, IFU, benefit-risk conclusions, and ongoing post-market evidence strategy.

Below are 10 common CER deficiencies that can trigger Notified Body questions.

Intended Purpose Is Not Consistent

The intended purpose is the foundation of the CER.

If the IFU says one thing, the CER says another, and the website or brochure says something broader, the clinical evidence becomes difficult to defend.

For example, “temporary haemostasis during surgery” is not the same as “reduces postoperative bleeding and complications.” The second statement introduces stronger clinical outcomes and may require stronger evidence.

What manufacturers should do:
Create one controlled intended-purpose statement and align it across the IFU, CER, risk management file, PMS, PMCF, labels, website, and marketing materials.

Clinical Claims Are Stronger Than the Evidence

Many CER findings happen because the claims are too strong for the available evidence.

A device may have bench testing that supports technical performance. But that does not automatically prove reduced complications, faster recovery, lower infection risk, or improved quality of life.

Each claim needs the right level of evidence.

A technical claim may be supported by bench data.
A performance claim may need clinical evidence.
A comparative claim may need comparative data.
A clinical benefit claim may need outcome-based evidence.

What manufacturers should do:
Build a Clinical Claims Matrix and map every major claim to required evidence, available evidence, and remaining gaps.

State of the Art Is Too Weak

A weak State of the Art section often reads like a general disease background.

That is not enough.

A strong SOTA should define current clinical practice, available treatment options, alternative technologies, accepted endpoints, expected outcomes, complication rates, safety benchmarks, performance benchmarks, and unmet clinical needs.

Without benchmarks, it is difficult to justify whether device safety and performance are acceptable.

What manufacturers should do:
Use SOTA to establish clinical benchmarks, then compare device-specific evidence against those benchmarks.

Literature Search Is Not Reproducible

A CER cannot simply say, “PubMed and Google Scholar were searched.”

Notified Bodies expect a clear and reproducible literature search strategy.

The CER should explain which databases were searched, which keywords were used, what time period was covered, how articles were screened, what was excluded, and why studies were included.

What manufacturers should do:
Document the search protocol, databases, keywords, inclusion criteria, exclusion criteria, screening process, and evidence selection flow.

Evidence Appraisal Is Too Superficial

Finding papers is not the same as proving clinical evidence.

Every article does not carry the same value.

A randomized controlled trial, retrospective study, case report, expert opinion, registry, and review article all provide different levels of confidence.

Evidence appraisal should assess relevance, methodological quality, and clinical applicability.

A paper may be scientifically strong but not relevant to the device. Another paper may be highly relevant but methodologically weak.

What manufacturers should do:
Appraise each source for relevance to the device, population, indication, claims, and clinical outcomes.

Equivalence Is Claimed Without Proper Justification

Equivalence is often misunderstood.

Saying two devices look similar is not enough.

Under MDR, equivalence should consider technical, biological, and clinical characteristics. This includes design, dimensions, materials, tissue contact, duration of contact, intended purpose, anatomical site, user, population, and clinical performance.

A similarity claim should not be presented as equivalence unless the MDR equivalence criteria are properly demonstrated.

What manufacturers should do:
Prepare a characteristic-by-characteristic equivalence assessment and justify whether differences are clinically significant.

CER and Risk Management Are Not Connected

Clinical evaluation and risk management cannot work separately.

If the risk management file identifies a clinically relevant risk, the CER should evaluate whether clinical evidence supports the acceptability of that residual risk.

For example, if the RMF identifies tissue injury as a risk, the CER should assess reported tissue injury, incidence, severity, comparison with SOTA, and residual risk acceptability.

What manufacturers should do:
Connect hazard, clinical risk, risk control, residual risk, clinical evidence, and benefit-risk conclusion.

Device-Specific Clinical Evidence Is Insufficient

A CER can include many references and still have weak clinical evidence.

The real question is not how many references are included.

The real question is how much evidence directly relates to the device under evaluation.

A CER may contain many SOTA articles and competitor publications, but very little device-specific evidence. That can create a major review concern.

What manufacturers should do:
Separate evidence into device-specific evidence, equivalent-device evidence, similar-device evidence, and SOTA evidence. Do not treat them as equal.

PMCF Is Generic and Does Not Address Evidence Gaps

A generic PMCF plan is a common weakness.

Statements like “we will continue to monitor clinical performance through literature and feedback” may not be enough if there are specific unresolved clinical questions.

PMCF should be driven by the evidence gap.

If the gap is long-term performance, then the PMCF activity should collect long-term data. If the gap is rare adverse events, then the method should be capable of detecting them.

What manufacturers should do:
Start with the remaining uncertainty, then select the PMCF activity that can answer it.

CER Conclusion Overstates the Evidence

The CER conclusion is not just a closing paragraph.

It is where the full clinical evidence argument should come together.

A weak conclusion says:

“The device is safe and effective.”

A stronger conclusion explains which claims are supported, what evidence supports safety and performance, how the device compares with SOTA, whether benefits outweigh residual risks, what uncertainty remains, and how PMCF will address it.

A strong CER conclusion does not claim that everything is proven. It clearly states what the available evidence supports and what will continue to be monitored.

The Hidden Problem: CER Deficiencies Are Connected

CER deficiencies rarely happen in isolation.

A broad intended purpose can create broad claims.
Broad claims can make evidence look insufficient.
Insufficient evidence can lead to stretched equivalence.
Weak equivalence can weaken benefit-risk justification.
Then PMCF becomes unclear, and the CER conclusion becomes difficult to defend.

This is why simply rewriting the CER conclusion may not solve the problem.

Manufacturers may need to revisit the complete evidence structure.

A Better Way to Build a CER

A stronger CER should start before writing begins.

The process should begin with device understanding, intended purpose, clinical claims, clinically relevant risks, evidence requirements, available evidence, evidence gaps, and evidence generation needs.

Only after this should the CER be written.

This makes the CER stronger, clearer, and easier to defend during Notified Body review.Final Thoughts

A strong CER is not defined by the number of pages.

It is defined by whether a reviewer can clearly trace:

Intended Purpose → Clinical Claims → Clinical Risks → Evidence Requirements → Available Evidence → SOTA Benchmarks → Benefit-Risk → Evidence Gaps → PMCF

When these elements are connected, the CER becomes a defensible clinical evidence argument.

When they are disconnected, Notified Body questions become much more likely.

Clinical evidence, technical documentation, QMS, post-market obligations, and local representative requirements may differ from one market to another.

How Bioexcel Can Help

Bioexcel supports medical device manufacturers with CER preparation, CER updates, MDR remediation, systematic literature review, SOTA analysis, claims-to-evidence mapping, equivalence assessment, risk-management alignment, PMCF planning, and Notified Body response support.

Bioexcel can also perform a CER Clinical Evidence Gap and Notified Body Readiness Assessment to identify critical, major, and minor gaps before submission.

Request a CER Evidence Gap Assessment and strengthen your CER before your next Notified Body review.