Clinical Evaluation Report: The Document MDR Made Harder

A CER is the document arguing, from clinical data, that a device achieves its intended performance and that its benefits outweigh its risks. It sits in the technical documentation and a notified body reads it closely.

MEDDEV 2.7/1 rev 4 still describes the structure most CERs follow, and the MDR requirements now sit above it.

The five stages

1. Scope and clinical evaluation plan. What the device is, its intended purpose, the claims to be substantiated, and the state of the art it will be compared against.
2. Identification of data. Literature, clinical investigations, and post-market data on the device or an equivalent one.
3. Appraisal. Each dataset assessed for methodological quality and relevance. Weak studies are not excluded silently; they are appraised and weighted.
4. Analysis. Do the data, taken together, demonstrate conformity with the relevant safety and performance requirements.
5. The report. The argument, with its limitations stated.

The plan is not optional and it comes first. A CER written without one reads as a literature review assembled to support a conclusion already reached, which is how notified bodies read it too.

Equivalence got much harder

Under the directive, claiming equivalence to another manufacturer's device was a common route to avoiding your own clinical investigation.

MDR requires equivalence on three characteristics: technical, biological and clinical. And for implantable and Class III devices, claiming equivalence to a device you do not make requires a contract giving you full access to that manufacturer's technical documentation on an ongoing basis.

Competitors do not sign those. In practice that closed the route for most higher-class devices, and manufacturers relying on it under the MDD had to generate their own data or exit.

State of the art

A CER must position the device against the current state of the art, which is a moving target. A report concluding a device is acceptable against 2015 practice is not answering the question when the alternatives have improved since.

This is also why a CER is a living document rather than a one-off.

PMCF closes the loop

Post-market clinical follow-up is the planned, proactive collection of clinical data on the marketed device. It is not the same as complaint handling, which is reactive.

The PMCF plan states what questions remain open, how data will answer them, and on what schedule. The PMCF evaluation report feeds back into the CER, which is updated on a defined frequency: at least annually for Class III and implantables, and at intervals justified by risk for lower classes.

A CER whose PMCF section says data will be collected, with no plan naming the questions, is the gap that gets found.