CE marking is a starting point, not a finish line
Under the EU Medical Device Regulation, 2017/745, obtaining CE marking demonstrates that a device meets the applicable safety and performance requirements at the time of assessment. It does not close the file on clinical evidence. The regulation requires manufacturers to maintain a Post-Market Clinical Follow-up, PMCF, process for as long as the device remains on the market, as part of an ongoing clinical evaluation. This is a change in framing that some manufacturers, understandably, are still adjusting to: clinical evidence generation under the MDR is a continuous activity, not a milestone reached once and then filed away.
What PMCF actually asks for
PMCF is defined in the regulation as a continuous process to update the clinical evaluation, using clinical data collected from the actual use of a device that already carries the CE mark. Its purposes include confirming the safety and performance of the device throughout its expected lifetime, identifying previously unknown side effects, monitoring identified risks, and identifying emerging risks based on factual evidence. In practice, this means a PMCF plan needs pre-specified objectives, a methodology proportionate to the risk class of the device, and statistically sound criteria for what would trigger a reassessment of the risk-benefit profile.
Where studies fall short
A recurring weakness, discussed here in general and educational terms rather than for any specific device or file, is a PMCF plan built around data collection without a clear analysis strategy behind it. Registries are established, questionnaires are distributed, and data accumulate, but without pre-specified statistical objectives, comparable to what a clinical trial would define in advance, it becomes difficult to state afterwards whether the accumulated evidence actually supports the conclusions being drawn from it. A second common gap is a PMCF plan that is proportionate to the regulatory requirement on paper, but not to the actual risk class or novelty of the device, resulting in either an unnecessarily heavy study for a low-risk device or, more concerning, insufficient evidence generation for a higher-risk one.
What a properly planned PMCF study looks like
A well structured PMCF plan defines its clinical questions in advance, in the same way a study protocol would: what safety or performance signal is being monitored, what data will be collected to monitor it, what sample size or duration is needed to detect a meaningful change, and what threshold would trigger an update to the clinical evaluation report. Where a PMCF activity is not a full clinical investigation but a lower-intensity one, for example a registry or a literature and complaint-data review, the same discipline still applies: the objectives and the statistical basis for any conclusion should be defined before the data are collected, not fitted to the data afterwards.
A continuous obligation, not a single event
The clinical evaluation report is a living document under the MDR, expected to be updated as PMCF data accumulate, and the frequency of that update is itself tied to the risk classification of the device. Treating PMCF as a statistical programme that runs for the life of the device on the market, rather than as a study conducted once shortly after CE marking, is the shift in mindset the regulation is asking for.
Key points
- CE marking does not end the clinical evidence obligation under the EU MDR; PMCF continues it for the life of the device on the market.
- PMCF plans need pre-specified statistical objectives, not only a data collection mechanism.
- The scope of PMCF should be proportionate to the device's risk classification and novelty, not to what is administratively convenient.
- The clinical evaluation report should be updated as PMCF evidence accumulates, on a frequency tied to risk class.