CER Conclusions: What Makes Them Convincing

Written by HATEM RABEH, MD, MSc Ing

Your Clinical Evaluation Expert And Partner

in
S

The conclusion section of a Clinical Evaluation Report is not a summary. It is not a recap of what was already written. It is the place where the manufacturer demonstrates clinical acceptability based on the totality of evidence.

When this section is weak, the entire CER becomes questionable. Reviewers begin to doubt whether the evaluation was rigorous, whether the data was properly interpreted, and whether the manufacturer understands the regulatory burden of proof.

This is not about writing style. It is about regulatory logic.

What the Conclusion Section Must Deliver

According to MDR Annex XIV, the clinical evaluation must confirm that the device meets the general safety and performance requirements under normal conditions of use. The conclusion section is where this confirmation is made explicit.

It must address three core elements:

First, the benefit-risk determination. This is not a statement that benefits exist and risks exist. It is a demonstration that the clinical benefits outweigh the residual risks for the intended patient population under the intended conditions of use.

Second, the justification of residual risks. Every medical device has risks that cannot be eliminated. The conclusion must show why these risks are acceptable in relation to the clinical benefit, and why no safer alternative exists that provides equivalent benefit.

Third, the confirmation that clinical data supports the claims. This includes performance claims, safety claims, and any claims made in the instructions for use. If the data does not fully support a claim, the conclusion must acknowledge this and explain how it will be addressed.

When these elements are missing or weak, the CER loses its regulatory value. The Notified Body cannot assess whether the manufacturer has met the requirements. The result is a deficiency letter, often late in the process, and often requiring substantial rework.

Why Most Conclusion Sections Fail

The first failure mode is descriptive writing. The manufacturer lists the studies, lists the findings, and stops. No determination is made. No position is taken.

I have reviewed CERs where the conclusion section read like an executive summary. It described what was evaluated. It mentioned that data was reviewed. But it never stated whether the device was clinically acceptable.

The second failure mode is unsupported claims. The manufacturer states that the benefit-risk profile is favorable, but does not explain how this determination was made. No reference to specific data. No reasoning process. Just an assertion.

Reviewers do not accept assertions. They expect to see the logic that connects the evidence to the determination.

The third failure mode is the avoidance of uncertainty. Many manufacturers write conclusions as if all questions have been answered. But clinical evaluation often reveals gaps. Data may be limited. Long-term outcomes may be unknown. Certain subpopulations may not have been studied.

When these gaps are not acknowledged in the conclusion, the CER appears incomplete or dishonest. Reviewers notice. They ask why the manufacturer did not address obvious limitations.

These failures do not happen because the manufacturer lacks data. They happen because the conclusion section was treated as an afterthought. It was written last, often quickly, without understanding its regulatory function.

How to Structure a Convincing Conclusion

The conclusion should begin with a clear statement of clinical acceptability. This is the determination. It should be explicit.

For example: “Based on the clinical evaluation of data from six clinical studies, post-market surveillance data covering 12,000 patient-years, and published literature on equivalent devices, the device meets the general safety and performance requirements under MDR Annex I for the intended purpose and patient population.”

This statement is direct. It references the data sources. It connects the evaluation to the regulatory requirement. It leaves no ambiguity about the manufacturer’s position.

After the initial determination, the conclusion must address the benefit-risk profile. This requires specific reasoning.

What are the key clinical benefits? How were they demonstrated? What are the residual risks? How were they characterized? Why do the benefits outweigh the risks?

This is not about listing benefits and risks. It is about showing the reasoning that led to the determination. Reviewers need to follow the logic. If the logic is not visible, the determination is not credible.

Next, the conclusion must address data gaps and uncertainties. Every evaluation has limitations. Acknowledge them. Explain how they will be managed.

If long-term data is limited, state this. Explain how PMCF will address it. If certain patient subgroups were not studied, acknowledge this. Explain the rationale for the intended patient population.

Reviewers do not expect perfect data. They expect honest assessment and a plan to address limitations.

Finally, the conclusion must confirm that the claims are supported. This includes performance claims, safety claims, and any indications or contraindications stated in the IFU.

If a claim is partially supported, say so. Explain what additional data will be collected. If a claim cannot be supported, remove it.

This level of transparency builds credibility. It shows that the evaluation was rigorous and that the manufacturer understands the evidence base.

What Reviewers Look For

Notified Body reviewers and competent authorities do not read the CER sequentially. They often start with the conclusion. They want to know the manufacturer’s position before they evaluate the evidence.

If the conclusion is vague or unconvincing, they approach the rest of the document with skepticism. They look for problems. They question the interpretation.

If the conclusion is clear and well-reasoned, they look for consistency. They check whether the data supports the determination. They verify that the logic holds.

The conclusion sets the tone. It signals whether the manufacturer has done the work.

Reviewers also look for alignment between the conclusion and the risk management file. If the conclusion states that residual risks are acceptable, but the risk management file identifies unmitigated risks, there is a contradiction. This triggers questions.

The conclusion must reflect the current state of the risk management process. It must be consistent with the final risk-benefit determination documented in the risk management report.

Another element reviewers check is the link to the State of the Art. The conclusion must confirm that the device performs at least as well as equivalent devices, or that any differences in performance are justified by other benefits.

If the SOTA analysis identified equivalent devices with better safety profiles, the conclusion must address this. Why is the device still acceptable? What is the justification?

This is where many manufacturers struggle. They do not want to acknowledge that competitors have advantages. But if the SOTA analysis shows this, ignoring it in the conclusion is not an option.

Common Mistakes That Weaken Conclusions

One common mistake is using passive voice and vague language. “It has been determined that the device is safe.” Who determined this? Based on what criteria? This type of writing creates distance between the manufacturer and the determination.

The manufacturer must own the conclusion. Use active voice. “We have determined that the benefit-risk profile is favorable based on the following reasoning.”

Another mistake is over-reliance on statistical significance. Manufacturers often cite p-values as if they prove clinical acceptability. Statistical significance does not equal clinical relevance.

The conclusion must address whether the effect size is clinically meaningful, whether it translates into patient benefit, and whether it justifies the residual risk.

A third mistake is treating the conclusion as optional. I have seen CERs with one-paragraph conclusions that simply state compliance with the MDR. This is not acceptable.

The conclusion is a required element under Annex XIV. It must be substantive. It must demonstrate clinical acceptability. A perfunctory statement does not fulfill the requirement.

Finally, many manufacturers fail to update the conclusion when the CER is revised. The conclusion was written based on an earlier version of the data analysis or risk assessment, and it was never revisited.

The conclusion must reflect the final state of the evaluation. Every time the CER is updated, the conclusion should be reviewed and revised if necessary.

How to Test Whether Your Conclusion Is Convincing

Before submitting the CER, read only the conclusion section. Then ask:

Can someone who has not read the rest of the report understand what determination was made and why?

Is the reasoning visible? Are the key data points referenced?

Are the residual risks acknowledged and justified?

Are the limitations addressed with a plan?

If the answer to any of these questions is no, the conclusion needs work.

Another test is to compare the conclusion with the summary of safety and clinical performance (SSCP) if one exists. The SSCP is a public document that must be consistent with the CER conclusion. If there are contradictions, both documents are called into question.

A third test is to have someone outside the project team read the conclusion. If they cannot follow the reasoning or if they have questions that are not answered, the conclusion is not complete.

These tests are simple, but they reveal weaknesses that would otherwise only be identified during Notified Body review.

Final Considerations

The conclusion section is not about making the CER look good. It is about making a defensible regulatory determination based on the evidence.

When this determination is clear, well-reasoned, and supported by data, the CER becomes a credible document. Reviewers can assess it. They can agree or disagree with specific points, but they cannot dismiss it.

When the determination is vague, unsupported, or incomplete, the CER loses its function. It becomes a document that describes an evaluation without making a judgment.

That is not acceptable under the MDR. The manufacturer must take a position. The conclusion is where that position is documented.

Most deficiencies in the conclusion section are avoidable. They happen because the conclusion was not given the attention it requires. It was treated as a formality instead of a regulatory deliverable.

If you invest time in writing a strong conclusion, it changes how the entire CER is received. It signals competence. It shows that the manufacturer understands the regulatory framework and has done the work to demonstrate compliance.

That is what makes a conclusion convincing.

Peace,
Hatem
Clinical Evaluation Expert for Medical Devices
Follow me for more insights and practical advice.