EU IVDR Performance Evaluation Under Article 56: Three Pillars and the PER

1. Key takeaways

Performance evaluation is the technical evidence core of the IVDR. Under Article 56(3) it is built on three pillars: scientific validity, analytical performance and clinical performance. Article 56(4) requires a clinical performance study unless it is duly justified to rely on other sources of clinical performance data; Article 56(5) requires the three pillars, their data evaluation and the clinical evidence drawn from them to be documented in the performance evaluation report (PER), and the PER forms part of the technical documentation.

The three pillars are not three documents but one evidence chain: scientific validity establishes that the analyte is associated with the clinical or physiological state, analytical performance establishes that the device measures reliably, and clinical performance establishes that a reliable measurement translates into a clinically correct conclusion. The PER is also a living document: for Class C and Class D devices it must be updated as necessary and at least annually, drawing on post-market performance follow-up data. Building the file around the pillars, and building the update mechanism around the PER, is the structure that notified body reviewers expect to find.

2. Legal basis

The table below lists the primary legal sources on performance evaluation.

3. Expert insights

Insight 1: the three pillars are a progressive evidence chain and cannot substitute for each other.

Article 2(44) defines performance evaluation as the assessment and analysis of data to establish or verify the scientific validity, the analytical performance and, where applicable, the clinical performance of a device. The first pillar answers whether the association of the analyte with the target clinical or physiological state is scientifically supported; the second answers whether the device measures accurately and reliably; the third answers whether an accurate measurement truly leads to a clinically correct conclusion for the intended use. If any pillar is missing, the chain breaks, and no volume of data in the other two pillars repairs it. Notified bodies assess the chain as a whole, so the PER should explicitly show where each pillar is addressed.

Insight 2: Article 56(4) makes the study the default and the absence of a study the exception that must be argued.

The wording of Article 56(4) is that a clinical performance study shall be carried out, unless it is duly justified to rely on other sources of clinical performance data. Usable other sources include literature, experience data from previous similar devices, or studies using leftover samples. The justification must be written into the PER and explain why the intended purpose can be supported without a study. A generic reference to literature is not an argument: the justification must connect the cited sources to the specific claims of the device and show that the data cover the intended users and sample types.

Insight 3: the PER is a living document, updated at least annually for Class C and D.

Article 56(6) requires the performance evaluation and its documentation to be updated throughout the lifecycle with data gained from the PMPF plan, and for Class C and Class D devices the performance evaluation report must be updated as necessary and at least annually. This turns the PER from a pre-market file into a continuing compliance activity. The update mechanism needs owners, an annual trigger and a defined route from PMPF data to revised conclusions; without those, the annual update degenerates into a formality that a notified body will identify as such at the next surveillance audit.

Insight 4: the required strength of the evidence must be argued by the manufacturer first.

The second subparagraph of Article 56(1) requires the manufacturer to specify and justify the level of clinical evidence necessary to demonstrate conformity with the relevant general safety and performance requirements, appropriate to the device characteristics and its intended purpose. In other words, the evidence strategy comes before the evidence, and the justification record itself is an object of notified body review. A well-written level-of-clinical-evidence justification also disciplines the size of the clinical performance work: it prevents both under-evidence for high-risk claims and unnecessary studies for low-risk claims.

4. Practical recommendations

The following actions structure the performance evaluation file and its maintenance.

1. Build separate volumes for the three pillars: scientific validity (literature and evidence on the analyte to clinical state association), analytical performance (precision, accuracy, limit of detection, linearity, interference, cross-reactivity, stability) and clinical performance (clinical sensitivity and specificity, expected values, comparison with reference methods).
2. Issue a level-of-clinical-evidence justification stating the intended purpose, the risks, the evidence level and the reasons (Article 56(1)).
3. Where no clinical performance study is conducted, complete the justification of other sources under Article 56(4).
4. Compile the PER according to the structure of Annex XIII Part A and include it in the technical documentation (Annex II).
5. For Class C and D, establish a mechanism to update the PER at least annually, linked to the PMPF data of Annex XIII Part B.

5. Verification checklist

Confirm the following points on the performance evaluation file.

• □ Evidence for the three pillars (scientific validity, analytical performance, clinical performance) has been compiled in separate volumes
• □ The level-of-clinical-evidence justification has been prepared (Article 56(1))
• □ Where no clinical performance study is conducted, the justification of alternative sources under Article 56(4) has been completed
• □ The PER has been compiled per Annex XIII Part A and included in the technical documentation (Article 56(5))
• □ For Class C and D, a mechanism to update the PER at least annually is in place (Article 56(6))
• □ A route for PMPF data (Annex XIII Part B) to flow back into the PER has been established

Regulatory information verified as of October 29, 2026.

GRZAN supports IVD manufacturers in the EU market with CE certification pathway and classification strategy under Regulation (EU) 2017/746. For performance evaluation we help you build three-pillar evidence files, draft level-of-clinical-evidence justifications, structure PERs per Annex XIII Part A and establish annual update mechanisms linked to PMPF. Contact us at info@grzanglobal.com to discuss your EU IVDR programme.