What your consultant cannot decide for you
I review outsourced clinical evaluation reports every month. The reports arrive complete, formatted, signed by qualified experts. Yet the manufacturer cannot defend them in front of the Notified Body. The consultant did their job. But the manufacturer never owned the choices that mattered.
In This Article
The question comes up often: what parts of clinical evaluation can we outsource, and what must stay in-house?
The instinct is to think about tasks. Who writes the report. Who searches literature. Who builds the PMCF plan.
But the real line is not about tasks. It is about accountability.
Under MDR Article 10, the manufacturer holds full responsibility for the clinical evaluation. Outsourcing the work does not transfer that responsibility. It does not reduce it. The manufacturer must demonstrate that they understand the clinical data, that they made informed decisions, and that they can justify those decisions to the Notified Body.
What I see in audits is this: manufacturers treat clinical evaluation as a document to receive, not a process to own.
Where outsourcing works well
External consultants can bring value in specific areas where expertise is deep and execution is repeatable.
Literature searches benefit from trained methodologists. A well-structured search protocol, executed by someone who understands medical databases, yields better coverage and fewer gaps. Consultants who do this daily are faster and more thorough than an internal team doing it occasionally.
The search itself can be outsourced. The protocol that defines the search cannot.
Why? Because the protocol reflects clinical strategy. It defines what equivalence means. It chooses which outcomes matter. These are not technical questions. They are regulatory positions that the manufacturer must own.
If the Notified Body challenges the search strategy, the consultant cannot answer. The manufacturer must explain why those inclusion criteria were chosen, why that definition of equivalence was adopted, why those endpoints were prioritized.
You can outsource execution of literature searches. You cannot outsource the clinical rationale that defines what to search for and why.
The same applies to appraisal. Extracting data from studies, summarizing methods, documenting outcomes—these are tasks that consultants perform routinely. But deciding whether a study is relevant to your device, whether its data supports your claims, whether its limitations matter for your intended use—those are judgments the manufacturer must make.
A consultant can prepare the analysis. The manufacturer must validate the conclusions.
What cannot leave the building
Certain decisions define the entire clinical evaluation strategy. These must originate inside the manufacturer because they depend on knowledge that external consultants do not have.
The intended purpose and clinical claims come first. These are not just regulatory statements. They are promises about what the device does, for whom, under what conditions. The consultant does not know your market. They do not know your design constraints. They do not know the conversations you had with key opinion leaders or the feedback from early users.
The intended purpose must be written by people who understand the product, the clinical need, and the commercial reality.
If those statements are drafted externally and then accepted without challenge, the entire clinical evaluation is built on someone else’s interpretation of your device.
I see this often. The consultant writes a draft intended purpose based on the IFU or a brief call. The manufacturer approves it. Six months later, the Notified Body asks why certain claims are missing or why certain populations were excluded. The manufacturer has no answer because they never asked those questions.
Manufacturers approve intended purpose and claims drafted by consultants without cross-checking against design history, risk files, and clinical input. The statements are defensible on paper but disconnected from the actual development process.
State of the art determination is another area that cannot be fully outsourced. A consultant can identify comparable devices, summarize published technologies, and describe clinical practices. But only the manufacturer knows which technologies were considered during development, which design choices were made, and why certain alternatives were rejected.
SOTA is not an academic exercise. It is a demonstration that your device reflects current knowledge and practice. The Notified Body expects the manufacturer to show that design decisions were informed by that knowledge.
If the SOTA section was written externally without input from R&D, it becomes a separate narrative. The clinical evaluation report describes one version of the state of the art. The design history file describes another. The Notified Body sees the gap.
The equivalence decision
This is the point where most outsourcing arrangements break down.
Manufacturers assume that the consultant will identify an equivalent device, gather its data, and build the equivalence argument. The consultant delivers a report. The report names an equivalent device and cites clinical studies.
Then the Notified Body asks: why did you choose this device as equivalent? What clinical, technical, and biological characteristics did you compare? Who made the decision?
The manufacturer points to the consultant. The consultant points to the data in the report.
But the regulation does not ask who wrote the report. It asks who made the decision and on what basis.
Under MDCG 2020-5, equivalence is a multi-step determination involving technical, biological, and clinical comparisons. It requires access to design specifications, material compositions, performance data, and risk analysis. Most consultants do not have access to this information. Even if they do, they are not responsible for the device design.
The equivalence claim is a manufacturer decision. It cannot be delegated.
What the consultant can do is document the comparison once the manufacturer has identified the equivalent device and provided the necessary technical data. The consultant structures the argument, applies MDCG 2020-5 framework, checks for gaps. But the starting point—this device is equivalent to ours—must come from the manufacturer.
Equivalence is not identified by consultants. It is determined by the manufacturer and documented by consultants. If you cannot explain why the chosen device is equivalent, the claim was never yours.
Clinical investigation planning
When new clinical data is needed, the manufacturer must decide what questions the study will answer, what endpoints will demonstrate safety and performance, and what risk the study is designed to address.
A consultant can design the protocol. But they cannot define the clinical objective without direction from the manufacturer.
I have reviewed protocols where the study design was technically sound but clinically irrelevant. The endpoints measured something, but not the thing that mattered for the clinical claims. The sample size was justified statistically but inadequate for the intended population.
This happens when protocol design is outsourced without sufficient manufacturer input. The consultant builds a study that works on paper. The manufacturer approves it because it looks professional. The Notified Body asks why the study does not address the main uncertainties in the clinical evaluation.
There is no good answer.
Planning a clinical investigation requires integration across functions. Clinical input to define relevant outcomes. Risk management input to identify what needs to be confirmed. Design input to clarify what performance looks like in practice. The consultant cannot integrate what they do not see.
PMCF planning and implementation
Post-market clinical follow-up is where the line between outsourcing and ownership becomes clearest.
The PMCF plan must describe how the manufacturer will collect real-world evidence, monitor emerging risks, and update the clinical evaluation. It must align with risk management, vigilance, and trend analysis systems that are internal to the manufacturer.
A consultant can structure the PMCF plan. They can propose data sources, suggest timelines, draft protocols for surveys or registries. But they cannot implement the plan. They do not have access to your complaint database. They do not see your sales data. They do not attend your cross-functional meetings.
PMCF execution is an internal process. Outsourcing pieces of it—data analysis, literature updates, report writing—is possible. But the manufacturer must own the integration of that data into decision-making.
What I see often is PMCF plans that describe activities the manufacturer never performs. The plan lists literature surveillance every six months. It does not happen. The plan proposes customer feedback forms. They are not deployed. The plan mentions integration with risk management. No process exists.
The Notified Body asks for evidence that PMCF is active. The manufacturer points to the plan. The plan was written externally. The activities were never internalized.
PMCF plans describe theoretical activities drafted by consultants, not actual processes integrated into the manufacturer’s operations. When asked for PMCF outputs, manufacturers produce nothing because the plan was never implemented.
The report signature trap
Many manufacturers believe that if the clinical evaluation report is signed by a qualified expert, they are compliant. The external consultant has medical credentials, they reviewed the data, they concluded that the benefit-risk is favorable.
But the qualified person is not signing off on behalf of the manufacturer. They are attesting that the report meets methodological standards, that conclusions are supported by data, that the evaluation was conducted per MDR Annex XIV.
They are not attesting that the manufacturer understands the evaluation, agrees with the conclusions, or can defend the choices made.
In audits, Notified Bodies often separate the qualified person from the manufacturer team and ask questions independently. If the manufacturer cannot answer without referring back to the consultant, the deficiency is noted. The regulation requires the manufacturer to demonstrate competence, not to hire competence.
The qualified person signature is necessary. It is not sufficient.
What internal ownership looks like
Ownership does not mean doing everything internally. It means understanding what was done, why it was done, and being able to justify it.
Before outsourcing clinical evaluation work, the manufacturer should have internal clarity on these questions: What is our device intended to do? What are the clinical benefits we claim? What are the residual risks? What evidence do we need to support our claims? What gaps exist in that evidence?
These questions are answered through internal cross-functional work. Clinical affairs, regulatory, R&D, risk management, quality. The consultant enters after these questions are answered, not before.
During the work, the manufacturer reviews intermediate outputs. Not just the final report, but the search protocol, the appraisal tables, the equivalence comparison, the draft conclusions. Each review is an opportunity to verify that external work aligns with internal knowledge.
After the work, the manufacturer validates the report. This is not a signature. It is a structured review where internal stakeholders confirm that the clinical evaluation reflects the actual device, the actual risks, the actual clinical data.
If the report contains statements that surprise the internal team, something went wrong.
Internal ownership is demonstrated through active validation, not passive approval. If your team reads the clinical evaluation report for the first time after it is finalized, you do not own it.
The accountability question
Outsourcing creates efficiency. It gives access to expertise that may not exist in-house. It allows small manufacturers to meet regulatory requirements without building large clinical teams.
But it does not change who is responsible.
When the Notified Body raises concerns about the clinical evaluation, the manufacturer must respond. The consultant may help draft the response. But the decision about how to address the concern, what additional data to provide, whether to revise claims or add studies—these decisions belong to the manufacturer.
If the manufacturer cannot make those decisions without waiting for consultant direction, the dependency is too high.
The goal is not to do everything internally. The goal is to remain the decision-maker, even when execution is external.
What this means in practice
When planning to outsource clinical evaluation work, define the boundaries clearly. What will the consultant do? What will the manufacturer decide?
A well-structured engagement separates tasks from decisions. The consultant executes literature searches per a manufacturer-approved protocol. The consultant documents equivalence based on manufacturer-provided technical comparisons. The consultant drafts the clinical evaluation report following manufacturer input on claims, risks, and data gaps.
At each step, the manufacturer provides direction and validates output.
This requires time. It requires internal competence to review the work critically. It requires knowing enough to ask the right questions.
If the manufacturer does not have that competence, outsourcing the report does not solve the problem. It delays it until the audit.
The alternative is to build internal capability before outsourcing. Train your regulatory and clinical affairs teams to understand clinical evaluation methodology. Give them access to MDCG guidance. Walk them through a few reports so they know what good looks like.
Then outsource execution while retaining judgment.
This is not about distrust. It is about accountability. The regulation places responsibility on the manufacturer. That responsibility cannot be outsourced, even if the work can be.
Outsourcing without internal competence does not create compliance. It creates dependency. Build enough knowledge internally to validate what you receive externally.
Clinical evaluation is not a document. It is a process of gathering, analyzing, and interpreting clinical data to demonstrate that your device is safe and performs as intended. That process can involve external support. But the interpretation, the conclusions, and the decisions must be yours.
If they are not, you are compliant on paper and exposed in practice.
Frequently Asked Questions
What is a Clinical Evaluation Report (CER)?
A CER is a mandatory document under MDR 2017/745 that demonstrates the safety and performance of a medical device through systematic analysis of clinical data. It must be updated throughout the device lifecycle based on PMCF findings.
How often should the CER be updated?
The CER should be updated whenever significant new clinical data becomes available, after PMCF activities, when there are changes to the device or intended purpose, and at minimum during annual reviews as part of post-market surveillance.
What causes CER rejection by Notified Bodies?
Common reasons include inadequate equivalence demonstration, insufficient clinical data for claims, poorly structured SOTA analysis, missing gap analysis, and lack of clear benefit-risk determination. Structure and logical flow are as important as the data itself.
Which MDCG guidance documents are most relevant for clinical evaluation?
Key documents include MDCG 2020-5 (Equivalence), MDCG 2020-6 (Sufficient Clinical Evidence), MDCG 2020-13 (CEAR Template), MDCG 2020-7 (PMCF Plan), and MDCG 2020-8 (PMCF Evaluation Report).
Need Expert Help with Your Clinical Evaluation?
Get personalized guidance on MDR compliance, CER writing, and Notified Body preparation.
✌
Peace, Hatem
Your Clinical Evaluation Partner
Follow me for more insights and practical advice.
– Regulation (EU) 2017/745 Article 10 (Obligations of manufacturers)
– Regulation (EU) 2017/745 Annex XIV (Clinical evaluation)
– MDCG 2020-5 Rev.1 (Clinical Evaluation – Equivalence)
– MDCG 2020-7 Rev.1 (Post-Market Clinical Follow-up)
Deepen Your Knowledge
Read Complete Guide to Clinical Evaluation under EU MDR for a comprehensive overview of clinical evaluation under EU MDR 2017/745.
