Many IVD manufacturers arrive at the IVDR with a validation file that served them well under the Directive: precision studies, a method comparison, a few published papers. The notified body then asks where the scientific validity report is, why there is no clinical performance report, and how the benefit-risk conclusion was reached. The data may exist; the structure the IVDR expects does not.
IVDR performance evaluation is not a renamed verification report. Regulation (EU) 2017/746 sets out a specific, continuous process with three separate lines of evidence, a plan, a report and a post-market follow-up. This guide walks through Article 56, Annex XIII and MDCG 2022-2, and the gaps that most often send a file back for another round.
What Article 56 IVDR requires
Article 56(1) states that confirmation of conformity with the general safety and performance requirements in Annex I must be based on scientific validity, analytical and clinical performance data providing sufficient clinical evidence. The manufacturer must specify and justify the level of clinical evidence needed, and that level must be appropriate to the characteristics of the device and its intended purpose.
The rest of the article builds the process around that principle:
- Article 56(2): clinical evidence must support the intended purpose and be based on a continuous process of performance evaluation, following a performance evaluation plan.
- Article 56(3): the performance evaluation must follow a defined and methodologically sound procedure to demonstrate scientific validity, analytical performance and clinical performance.
- Article 56(4): clinical performance studies under Annex XIII Part A Section 2 must be carried out unless it is duly justified to rely on other sources of clinical performance data.
- Article 56(5): the data, their assessment and the resulting clinical evidence are documented in the performance evaluation report, which forms part of the technical documentation.
- Article 56(6): the performance evaluation and its documentation are updated throughout the life cycle with data from the PMPF plan and the post-market surveillance plan. For class C and D devices, the performance evaluation report is updated when necessary, but at least annually.
The three pillars: scientific validity, analytical performance, clinical performance
Article 2 of the IVDR defines each element, and the definitions matter because each pillar answers a different question.
Scientific validity
Article 2(38) defines scientific validity of an analyte as the association of an analyte with a clinical condition or a physiological state. It is about the marker, not your device. Annex XIII Section 1.2.1 lists sources: information on devices measuring the same analyte, peer-reviewed literature, consensus expert opinions, proof-of-concept studies and clinical performance studies. The result is documented in a scientific validity report.
MDCG 2022-2 adds that where the association is well established, existing data such as literature and textbooks can be enough. Where evidence is insufficient or the device is novel, the manufacturer should give a scientific rationale, perform a gap analysis and may need to generate new data.
Analytical performance
Article 2(40) defines analytical performance as the ability of a device to correctly detect or measure a particular analyte. The parameters are those in Annex I Section 9.1(a), including analytical sensitivity and specificity, trueness, precision (repeatability and reproducibility), accuracy, limits of detection and quantitation, measuring range, linearity, cut-off, specimen collection and handling, interference and cross-reactions. Annex XIII Section 1.2.2 says analytical performance should, as a general rule, be demonstrated by analytical performance studies, documented in an analytical performance report.
Clinical performance
Article 2(41) defines clinical performance as the ability of a device to yield results that are correlated with a particular clinical condition or physiological or pathological process or state in accordance with the target population and intended user. Annex I Section 9.1(b) lists parameters such as diagnostic sensitivity, diagnostic specificity, positive and negative predictive values, likelihood ratio and expected values in normal and affected populations. Annex XIII Section 1.2.3 allows clinical performance studies, peer-reviewed literature and published experience from routine diagnostic testing as sources, documented in a clinical performance report.
Not every parameter applies to every device. Where one is not relevant, the performance evaluation should say why rather than leave it silent.
The performance evaluation plan and report under Annex XIII
Annex XIII Part A Section 1.1 sets the minimum content of the performance evaluation plan. It includes the intended purpose, device characteristics, the analyte or marker, metrological traceability, the target groups with indications and limitations, the Annex I requirements needing support, the methods and statistical tools for analytical and clinical performance, a description of the state of the art, benefit-risk acceptance parameters, reference databases for software, a development plan with milestones and acceptance criteria, and PMPF planning. Any element left out must be justified in the plan.
Section 1.3.2 describes the performance evaluation report. It brings together the scientific validity, analytical performance and clinical performance reports with an overall assessment, and covers the justification of the approach, the literature search methodology and report, the technology, the nature and extent of the data, the performance against the state of the art and the conclusions of PMPF reports. Section 1.3.3 requires the clinical evidence and report to be updated through the life cycle and to include both favourable and unfavourable data.
The level of evidence expected rises with risk, so start by confirming your IVDR classification under Annex VIII. Under Article 29, class C and D devices also need a summary of safety and performance that includes a summary of the performance evaluation and relevant PMPF information.
Performance studies under Articles 57 and 58
Article 57 sets general requirements for performance studies: they should be performed in circumstances similar to normal conditions of use, and the rights, safety, dignity and well-being of subjects prevail. Studies using left-over samples remain subject to data protection law.
Article 58(1) applies additional requirements, including Articles 58 to 77 and Annex XIV, where sample-taking is surgically invasive and done only for the study, where the study is an interventional clinical performance study, or where it involves additional invasive procedures or other risks for subjects. Under Article 58(2), studies of companion diagnostics are subject to the same requirements, except those using only left-over samples, which must be notified to the competent authority. MDCG 2025-5 (June 2025) answers common questions on performance studies.
PMPF: post-market performance follow-up
Annex XIII Part B defines PMPF as a continuous process that updates the performance evaluation, addressed in the post-market surveillance plan. Section 5.2 lists the plan content: general methods such as user feedback and literature screening; specific methods such as ring trials, epidemiological studies, registries, genetic databanks and post-market clinical performance studies; a rationale; references to the performance evaluation report and risk management; specific objectives; evaluation of equivalent or similar devices and the state of the art; relevant standards and guidance; and a justified time schedule. Findings go into a PMPF evaluation report (Section 6). Under Section 8, if PMPF is not appropriate, the justification is documented in the performance evaluation report. The structure closely mirrors post-market clinical follow-up under the MDR.
Where this goes wrong
- Scientific validity treated as obvious. The file assumes the analyte–condition link is well known and cites nothing. Even for established markers, Annex XIII expects a documented scientific validity report with a traceable literature search.
- Analytical studies reused for clinical claims. A method comparison against a predicate shows agreement, not diagnostic sensitivity and specificity in the target population. Clinical performance needs its own data or a justification under Article 56(4).
- Claims wider than the evidence. The intended purpose names a population, specimen type or user group that the studies did not include, such as lay users for a self-test.
- No state-of-the-art comparison. The report presents the device’s figures without showing how they compare with current practice, which Section 1.3.2 requires.
- Silent omissions. Parameters from Annex I Section 9.1 are left out with no justification.
- Unfavourable data filtered out. Section 1.3.3 requires both favourable and unfavourable data; a literature review listing only supportive papers will be challenged.
- A PMPF plan without objectives. “Monitor complaints” is not a plan. Tie each activity to an open question from the performance evaluation.
Frequently asked questions
What are the three components of IVDR performance evaluation?
Article 56(3) IVDR requires the performance evaluation to demonstrate scientific validity, analytical performance and clinical performance. Scientific validity concerns the link between the analyte and a clinical condition or physiological state. Analytical performance is the device’s ability to detect or measure the analyte correctly. Clinical performance is its ability to yield results correlated with a clinical condition in the target population and intended user.
Are clinical performance studies always required under the IVDR?
No. Article 56(4) requires clinical performance studies unless it is duly justified to rely on other sources of clinical performance data. Annex XIII Section 1.2.3 lists those sources as peer-reviewed literature and published experience from routine diagnostic testing, alongside studies. The justification must be documented, and it becomes harder to make for novel analytes or intended purposes with little published evidence.
How often must the performance evaluation report be updated?
Article 56(6) IVDR requires the performance evaluation report for class C and D devices to be updated when necessary, but at least annually, using PMPF and post-market surveillance data. MDCG 2022-2 indicates that for class A and B devices it is updated as necessary and as defined in the post-market surveillance plan. Plan the update cycle alongside the PSUR for class C and D.
Is MDCG 2022-2 legally binding?
No. MDCG 2022-2, published in January 2022, states that it is not a European Commission document and is not legally binding. The binding requirements are in Article 56 and Annex XIII of Regulation (EU) 2017/746. The guidance explains how notified bodies and competent authorities expect those requirements to be applied, so departing from it should be deliberate and justified.
Can PMPF be skipped for an established IVD?
Possibly, but only with a documented justification. Annex XIII Part B Section 8 states that where PMPF is not deemed appropriate for a specific device, a justification must be provided and documented in the performance evaluation report. MDCG 2022-2 confirms that PMPF can sometimes be justified as not required. Post-market surveillance itself remains mandatory.
If your IVD file needs restructuring into scientific validity, analytical and clinical performance reports, or you want a gap review before notified body submission, Medfins has supported 175+ regulatory projects for 100+ clients. See our EU IVDR CE marking support or contact our team.
