Why Your IVD Clinical Performance Report Keeps Getting Deficiency Letters

Written by HATEM RABEH, MD, MSc Ing

Your Clinical Evaluation Expert And Partner

in
S

I see this pattern repeat with IVD manufacturers every review cycle. The clinical performance report gets written. It gets submitted. Then the Notified Body or competent authority sends back questions that reveal the same structural problems.

The issue is not usually the volume of evidence. It is how that evidence was evaluated, interpreted, and connected to the specific performance characteristics of the device. Most deficiency letters for IVD clinical performance reports come from the same few root causes.

Understanding these causes changes how you prepare the entire evaluation. Not at the submission stage. At the planning stage.

The Structure That Nobody Explains Clearly

IVDR Article 56 requires manufacturers to demonstrate clinical performance through an evaluation that follows defined methods and produces specific outputs. MDCG 2022-2 provides the framework. But between the regulation and the guidance document, there is a practical gap.

Clinical performance evaluation for IVDs is not a literature review with conclusions. It is a structured assessment of whether the device performs as intended in the target population under defined conditions. That performance must be measurable, verifiable, and clinically relevant.

When I review IVD clinical performance reports during audits, I see the same structural weakness repeated. The report presents data. It describes studies. But it never establishes the performance standard against which the device was evaluated. Without that standard, the entire evaluation collapses under scrutiny.

This deficiency appears because the evaluation was written backwards. The manufacturer collected data, then wrote conclusions. But the evaluation should have started with the definition of what constitutes acceptable clinical performance. That definition should have been established before interpreting any data.

The Performance Claim That Was Never Validated

Many IVD clinical performance reports list performance claims in the intended use section. Sensitivity, specificity, accuracy, precision. The numbers are stated clearly. Then the evidence section presents studies.

But the connection between the claimed performance and the presented evidence is often missing or weak. The report shows that studies were performed. It does not demonstrate that those studies validated the specific performance claims under the intended conditions.

Reviewers ask the same question every time. Where is the evidence that supports this specific claim for this specific population in this specific setting?

This is not about having more studies. It is about demonstrating that the studies you present actually address the performance claims you make. That demonstration requires explicit analysis.

I have seen manufacturers present excellent analytical validation studies, then claim clinical performance based on those studies alone. That does not work. Analytical performance and clinical performance are different evaluations. One does not replace the other.

Clinical performance must show that the analytical performance translates into clinically meaningful results in the target population. That requires clinical data or a very strong equivalence demonstration.

The Equivalence Path That Leads Nowhere

When manufacturers use equivalence to demonstrate clinical performance, the deficiency letters become predictable. The equivalence demonstration is incomplete. The comparator device is not clearly identified. The technical and clinical comparison is superficial. Or the equivalence is claimed but never actually demonstrated.

IVDR allows equivalence for clinical performance evaluation under strict conditions. Those conditions are rarely fully met in the reports I review. The manufacturer states that the device is equivalent to another device, then assumes that the clinical performance of the comparator applies to the new device.

That assumption requires proof. The proof must show that the two devices are equivalent in all characteristics that could affect clinical performance. This includes analytical characteristics, sample type, measurement principle, intended use, and target population.

What happens instead is a partial comparison. The manufacturer shows that the devices measure the same analyte. Then claims equivalence. But equivalence is not about measuring the same thing. It is about measuring the same thing in the same way with the same performance in the same clinical context.

When this deficiency appears, the entire evaluation must be rebuilt. You cannot patch equivalence after the fact. Either the devices are equivalent across all relevant dimensions, or they are not. If they are not, you need direct clinical performance data.

The Literature Review That Proves Nothing

Almost every IVD clinical performance report includes a literature review. The review is often long. It cites many studies. It covers the analyte, the disease, the measurement technologies. Then it concludes that clinical performance is demonstrated.

The problem is that a literature review alone does not demonstrate clinical performance of your device. It demonstrates the state of knowledge about the measurand or the clinical condition. That is relevant context. It is not performance evaluation.

Reviewers look for evidence that directly evaluates the device under review. Generic literature about the biomarker does not substitute for specific data about how your device performs when measuring that biomarker in your intended population.

This distinction gets lost constantly. Manufacturers think that showing extensive knowledge about the clinical area is equivalent to demonstrating device performance. It is not.

The clinical performance evaluation must connect the device characteristics to clinical outcomes or clinical decision-making. Literature that does not involve the device or a demonstrated equivalent device cannot make that connection.

When I see a clinical performance report that is 80% literature review and 20% device evaluation, I know deficiency letters are coming. The balance must be reversed. The evaluation must focus on the device. Literature supports that focus. It does not replace it.

The Missing Link to Clinical Impact

Even when performance data is presented, the clinical performance report often fails to explain why those performance characteristics matter clinically. Numbers are given. Thresholds are met. But the clinical significance remains unclear.

IVDR requires that clinical performance be clinically relevant. This means the performance characteristics must support appropriate clinical decision-making or patient management. The report must explain that connection.

For example, stating that sensitivity is 95% is not enough. The evaluation must address what that sensitivity means for patient outcomes. Does it enable earlier detection? Does it reduce false negatives to a clinically acceptable level? How does that performance translate into clinical benefit or reduced risk?

This part of the evaluation requires clinical reasoning. It cannot be automated or templated. It requires understanding the clinical pathway, the role of the test in that pathway, and the consequences of correct or incorrect results.

Many IVD manufacturers are not accustomed to this type of clinical analysis. They focus on analytical validation because that is their core expertise. But clinical performance evaluation requires clinical thinking. That is where the gaps appear.

When this question arrives, it is often difficult to answer without conducting additional analysis or gathering expert input. That work should have been done during the evaluation, not in response to the deficiency letter.

The Performance Follow-Up That Was Never Planned

Clinical performance evaluation is not a one-time exercise. IVDR requires ongoing performance evaluation through post-market performance follow-up. The initial clinical performance report must include a plan for how performance will be monitored and updated.

Most reports mention PMPF. Few describe a functional plan. The plan must address what data will be collected, how it will be evaluated, and how it will feed back into the clinical performance evaluation. Without that plan, the initial evaluation is incomplete.

Reviewers see this immediately. They ask how the manufacturer will verify that the claimed performance holds in real-world use. They ask how changes in clinical practice or population characteristics will be detected. They ask what triggers will lead to updating the clinical performance evaluation.

If those questions cannot be answered, the clinical performance evaluation is not considered sufficient under IVDR. The evaluation must demonstrate not only current performance but also a system for maintaining that performance assessment over time.

This requires integration between clinical affairs, regulatory affairs, and quality management. That integration must be visible in the clinical performance report. The plan must be specific, resourced, and operationally realistic.

The Way Forward

Deficiency letters for IVD clinical performance reports are not random. They target the same structural weaknesses. Those weaknesses come from treating clinical performance evaluation as a documentation exercise instead of a clinical and scientific assessment.

The way to reduce deficiency letters is to build the evaluation correctly from the start. Define the performance standard. Establish the claims clearly. Gather evidence that directly addresses those claims. Demonstrate equivalence rigorously if you use that path. Connect performance to clinical relevance. Plan for ongoing verification.

None of this is optional. IVDR Article 56 and MDCG 2022-2 establish these requirements explicitly. The problem is not understanding what is required. The problem is executing that understanding in a clinical performance report that withstands regulatory review.

That execution requires clinical competence, regulatory precision, and scientific rigor. When those elements align, the clinical performance report becomes defensible. When they do not, the deficiency letters keep coming.

What remains is to integrate performance evaluation into every stage of device development, not as a final documentation step but as an ongoing clinical assessment that guides design, validation, and post-market surveillance.

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