They don’t read your CER from page one. Here’s what they check first.

Written by HATEM RABEH, MD, MSc Ing

Your Clinical Evaluation Expert And Partner

in
S

This is Part 2 of the Notified Body Deep Dive series. In Part 1, we looked at how reviewers are trained and what drives their decisions. Now we examine how they actually assess your technical documentation—specifically the clinical evaluation sections they prioritize first.

Understanding this sequence matters. Not because it changes what you must provide, but because it changes how you prepare your documentation. When you know what gets scrutinized first, you stop wasting effort on secondary details before the foundations are solid.

The Reality of Technical Documentation Assessment

MDR Annex IX defines the conformity assessment procedures. It requires Notified Bodies to review technical documentation for completeness, consistency, and adequacy. MDCG 2020-6 provides detailed guidance on what clinical evaluation must contain.

But guidance documents describe requirements. They don’t explain the sequence of review or the psychology of the reviewer sitting across from your submission.

Here’s what actually happens. The reviewer receives your file. They have 90 days for initial assessment under most circumstances. They have ten other files this month. They open yours.

They skip the table of contents. They skip the device description for now. They go straight to the clinical evaluation report. And within that report, they don’t start reading from page one.

The Three Sections Reviewed First

Every experienced reviewer follows a pattern. The pattern is not written in guidance documents. It emerges from practice, from efficiency, and from pattern recognition developed over hundreds of assessments.

They check three things first. If these three sections raise concerns, the reviewer already knows this will be a difficult file. If these sections are solid, they can proceed with more confidence that the rest might hold up.

Section One: The Equivalence Claim

If you claim equivalence to another device, this is the first thing they read. Not your executive summary. Not your literature search strategy. Your equivalence demonstration.

Why? Because equivalence is the highest-risk claim in clinical evaluation. It determines whether your entire clinical data strategy is valid. If the equivalence claim fails, your clinical evaluation collapses.

The reviewer looks for three elements immediately:

First, they check whether you actually demonstrated technical equivalence according to MDCG 2020-5. Not similarity. Not comparability. Technical equivalence. Same materials, same specifications, same design and manufacturing, same biological and mechanical characteristics.

Most manufacturers describe similarity and call it equivalence. The reviewer sees this in the first two paragraphs. Red flag.

Second, they check whether you demonstrated clinical equivalence. Same clinical performance, same indications, same target population, same duration of use, same severity of condition treated.

If your equivalent device was used in acute settings and yours is intended for chronic use, you don’t have clinical equivalence. If your equivalent was used under direct medical supervision and yours is for home use, you don’t have clinical equivalence.

The reviewer knows this. You should too.

Third, they look at your access to the equivalent device’s clinical data. Do you have comprehensive technical documentation from the equivalent manufacturer? Do you have authorization to reference their clinical data? Or are you relying on public literature about a device you don’t actually have full access to?

If any of these three elements are weak, the reviewer already knows your clinical evaluation strategy is compromised. They will proceed with heightened skepticism.

Section Two: The Literature Search Strategy

After checking equivalence—or if you have no equivalence claim—the reviewer goes to your literature search protocol.

They don’t read the hundreds of articles you appraised. Not yet. They read how you searched. They evaluate whether your search could have reasonably captured the relevant clinical data.

They check your databases. If you only searched PubMed, that’s a problem. Where are Embase, Cochrane, Scopus, other relevant databases for your device type?

They check your search strings. Were the search terms appropriate for your device technology and intended purpose? Did you use medical subject headings correctly? Did you combine device-specific terms with clinical outcome terms?

They check your date range. If you stopped your search two years ago and you’re submitting now, they want to know why. If you only searched back five years for a device technology that has existed for fifteen years, they want to know your justification.

They check your exclusion criteria. What did you decide not to include and why? If your exclusion criteria are vague or overly broad, the reviewer suspects you filtered out inconvenient data.

If the search strategy has obvious gaps, the reviewer already doubts whether your clinical data set is complete. They will scrutinize your appraisals with suspicion that relevant studies were missed—intentionally or through negligence.

Section Three: The State of the Art Analysis

Third, they read your state of the art section. This is often the weakest section in clinical evaluation reports. It’s also one of the most revealing.

The state of the art (SOTA) is not a literature review summary. It’s not a market analysis. It’s not a description of competing devices.

SOTA is a demonstration that you understand the current knowledge concerning your device type, the clinical condition it addresses, and the alternative treatment options. It establishes the clinical context in which your device will be used.

The reviewer looks for whether you identified relevant alternatives—not just other medical devices, but surgical procedures, drug therapies, non-interventional management. They look for whether you discussed performance benchmarks, known risks, and clinical outcome expectations for devices in this category.

Most manufacturers write SOTA as if it’s a formality. They copy device descriptions from competitors’ websites. They summarize a few guidelines. They call it complete.

The reviewer sees through this immediately. A superficial SOTA signals that the manufacturer doesn’t deeply understand the clinical environment their device will enter. If you don’t understand current clinical standards, how can you demonstrate your device meets those standards?

When SOTA is weak, the reviewer questions whether the benefit-risk evaluation later in the report is grounded in reality. How can you claim acceptable benefit-risk if you haven’t established what current clinical standards actually are?

Why This Sequence Matters

These three sections are reviewed first because they determine the validity of everything that follows.

Equivalence determines whether your data strategy is legitimate. Literature search determines whether your evidence base is complete. SOTA determines whether your analysis is contextually grounded.

If these three sections are solid, the reviewer can proceed through the rest of your CER with reasonable confidence. They will still scrutinize appraisals, check benefit-risk analysis, evaluate PMCF planning. But they move forward knowing the foundations are present.

If any of these three sections fail, the rest of the CER is contaminated. Your appraisals might be excellent, but if your search was inadequate, how does the reviewer know those appraisals cover all relevant data? Your benefit-risk conclusion might be reasonable, but if your SOTA is superficial, how does the reviewer know your conclusion is calibrated to actual clinical standards?

This is not about checklist compliance. It’s about structural validity.

What This Means for Documentation Preparation

Don’t write your clinical evaluation report from beginning to end and hope for the best. Build it strategically.

Start with your equivalence demonstration if you’re claiming equivalence. Invest the time to properly demonstrate technical and clinical equivalence according to MDCG 2020-5. If you can’t demonstrate it rigorously, don’t claim it. An invalid equivalence claim is worse than no equivalence claim.

Build your literature search protocol carefully. Document your decisions. Explain your database selections, your search strings, your date ranges, your exclusion criteria. Make it clear that you searched for all relevant data, not just data that supports your device.

Write your SOTA as a clinical analysis, not a market overview. Show that you understand the clinical condition, the current treatment landscape, the performance benchmarks, and the known risks in your device category. Demonstrate that you know what clinical standards your device must meet.

These three sections are not formalities. They are the structural foundations that make the rest of your clinical evaluation defensible.

The Fourth Section They Check (But Later)

After those three, the reviewer moves to your benefit-risk determination. But by the time they reach it, they already have a hypothesis about whether your conclusion will be credible.

If equivalence was weak, if the literature search was narrow, if SOTA was superficial—they approach your benefit-risk section expecting problems. Your conclusion might be correct, but the reviewer has already lost confidence in your judgment.

If those first three sections were solid, the reviewer approaches benefit-risk expecting competence. They still scrutinize the analysis, but they do so looking for confirmation rather than hunting for failures.

Perception drives review intensity. And perception is formed in the first thirty minutes of document assessment.

What About the Rest of the CER?

Everything else—device description, literature appraisals, clinical data analysis, detailed safety evaluation, PMCF planning—is reviewed after these initial sections establish confidence or doubt.

Those sections are absolutely required. They must be complete and compliant. But they are assessed in the context created by what the reviewer read first.

This is not stated in guidance documents. This is observed in practice. Reviewers are human. They form working hypotheses. Those hypotheses guide how skeptically they read everything that follows.

Understand the sequence. Use it to structure your documentation preparation. Don’t spend three months perfecting appraisals before you’ve validated that your equivalence claim or literature search will survive first contact with a reviewer.

Final Thought

The technical documentation assessment is not a linear reading process. It’s a strategic evaluation process. Reviewers prioritize what matters most for structural validity.

You can’t control the review process. But you can control how well you address what reviewers check first. Build your CER knowing where the scrutiny will be most intense.

In Part 3 of this series, we’ll examine what triggers major nonconformities during document review. Not minor findings. The issues that stop certification.

Because understanding what reviewers check first is only useful if you know what makes them write a major NC instead of a clarification request.