Post-market clinical follow-up under MDR is the continuous, proactive collection and evaluation of clinical data from a CE marked device in real-world use, in order to confirm safety and clinical performance throughout the expected lifetime of the device, identify previously unknown side-effects, monitor known risks, and keep the clinical evaluation current. The obligation is set out in Annex XIV Part B of Regulation (EU) 2017/745 and tied to the clinical evaluation update cycle by Article 61(11). PMCF is not optional unless the manufacturer writes and justifies a specific non-applicability statement inside the post-market surveillance plan.

By Tibor Zechmeister and Felix Lenhard. Last updated 10 April 2026.

TL;DR

  • PMCF is the clinical arm of the post-market surveillance system. It lives inside the PMS plan required by Articles 83 and 84, and it is documented in its own PMCF plan under Annex XIV Part B.
  • The outputs are a PMCF plan (written before CE marking) and a PMCF evaluation report (produced as the plan runs). The PMCF evaluation report feeds directly into updates of the clinical evaluation report under Article 61(11).
  • PMCF methods include post-market studies, registries, user surveys, structured real-world data collection, focused literature surveillance, and follow-up of equivalent or similar devices. The choice depends on the device, the risk class, the novelty, and the resources of the manufacturer.
  • PMCF may be declared not applicable for a specific device, but only with a written justification that addresses the Annex XIV Part B criteria explicitly. "We are a startup and cannot afford it" is not a justification.
  • For higher-class devices, PMCF findings are explicitly named in Article 86 as a required element of the Periodic Safety Update Report, and for Class III and implantable devices they feed the Summary of Safety and Clinical Performance under Article 32.
  • MDCG 2025-10 (December 2025) is the current operational guidance on post-market surveillance and how PMCF sits inside the broader PMS system.

Why PMCF exists — the arm strap that passed every lab test

A small MedTech company we worked with built a sleep-monitoring device worn on the upper arm. Biocompatibility testing was done before market. Everything passed. The notified body accepted the file. The product launched. And then, weeks into real-world use, the complaints started. Skin irritations. Mostly mild, a few not mild. The pattern was consistent: extended overnight wear, perspiration, and the specific textile-polymer interface that had never been tested against weeks of continuous contact in real conditions. Lab testing simulated. Real users slept.

The story shows up in the post on post-market surveillance under MDR because it is the cleanest example of why PMS exists at all. It belongs here for a different reason. The signal that caught the problem was a clinical signal — a cluster of adverse skin reactions in people using the device as intended. A purely technical complaint handling system might have logged "device complaint, skin" and closed it. The PMCF lens is what asks the next question. Is this a known side-effect or a new one? Is the frequency consistent with what the clinical evaluation predicted? Does the benefit-risk ratio still hold? Does the clinical evaluation report need to be updated? Does the risk file need to be updated? Does the labelling need to change?

That chain of questions is exactly what PMCF is designed to answer. Not "did we get a complaint" but "what does real-world clinical use tell us about the device that pre-market evidence could not." On the arm strap, PMCF discipline is what turned a cluster of complaints into a material specification change, an updated IFU, and an updated clinical evaluation — before the cluster became a safety action. That is the value of PMCF stated without marketing language.

What PMCF is under Annex XIV Part B

Annex XIV Part B of the MDR is the definitive source on PMCF. It is short, dense, and everything in this post traces back to it.

Annex XIV Part B defines PMCF as a continuous process that updates the clinical evaluation referred to in Article 61 and Annex XIV Part A, and is specifically addressed in the manufacturer's post-market surveillance plan. When performing PMCF, the manufacturer proactively collects and evaluates clinical data from the use in or on humans of a device which bears the CE marking and is placed on the market or put into service within its intended purpose, with the aim of confirming the safety and performance throughout the expected lifetime of the device, of ensuring the continued acceptability of identified risks, and of detecting emerging risks on the basis of factual evidence.

Read that paragraph carefully, because every word matters. Continuous — not episodic, not reactive. Proactive — not waiting for complaints to arrive, but actively seeking data. Clinical data — not service records and not production yields, but data on real-world clinical use in humans. Throughout the expected lifetime — not only the first year after launch. Confirming safety and performance — which implies a comparison against what the clinical evaluation predicted. Detecting emerging risks on the basis of factual evidence — which implies a method that can actually detect them.

Annex XIV Part B also sets out the specific objectives PMCF must aim to achieve. These include confirming the safety and performance of the device throughout its expected lifetime, identifying previously unknown side-effects and monitoring the identified side-effects and contraindications, identifying and analysing emergent risks on the basis of factual evidence, ensuring the continued acceptability of the benefit-risk ratio referred to in Sections 1 and 9 of Annex I, and identifying possible systematic misuse or off-label use of the device with a view to verifying that the intended purpose is correct.

These objectives are the test PMCF activities must pass. A PMCF activity that does not serve at least one of them is not really PMCF. It is busywork with a PMCF label on it.

The PMCF plan — structure and content

Annex XIV Part B requires the manufacturer to specify in the PMCF plan the methods and procedures for proactively collecting and evaluating clinical data. The plan is a document, part of the technical documentation, written before CE marking, and kept current during the lifecycle of the device.

The plan has to address, at a minimum, the following elements. The general methods and procedures of the PMCF to be applied, such as the gathering of clinical experience gained, feedback from users, screening of scientific literature and of other sources of clinical data. The specific methods and procedures of PMCF to be applied, such as evaluation of suitable registers or PMCF studies. A rationale for the appropriateness of the methods and procedures applied. A reference to the relevant parts of the clinical evaluation report referred to in Section 4 of Annex XIV Part A and to the risk management referred to in Section 3 of Annex I. The specific objectives to be addressed by the PMCF. An evaluation of the clinical data relating to equivalent or similar devices. Reference to any relevant common specifications, harmonised standards when used by the manufacturer, and relevant guidance on PMCF. And a detailed and adequately justified time schedule for PMCF activities, such as analysis of PMCF data and reporting, which the manufacturer is to implement.

That list is the anatomy of a real PMCF plan. The common startup mistake is to write a PMCF plan that names one generic method — "user surveys quarterly" — and stops there. A notified body auditor reading such a plan will ask where the rationale is, where the link to the CER is, where the link to the risk file is, which specific objectives each activity addresses, and what the time schedule is. If the plan cannot answer those questions, the plan is incomplete against Annex XIV Part B.

For the template-level walkthrough of the plan section by section, see the PMCF plan structure under Annex XIV Part B.

PMCF methods — what actually works

Annex XIV Part B names several categories of PMCF methods. In practice, startups pick from a menu that includes the following, and usually combine more than one.

Post-market clinical studies. A structured clinical study on the device after it is CE marked, conducted under EN ISO 14155:2020 + A11:2024 for the aspects that apply. Post-market studies are the strongest form of PMCF evidence and also the most expensive. They are often unavoidable for implantable and Class III devices, and sometimes for novel Class IIb devices with specific questions that cannot be answered by other methods.

Patient and device registries. Registries that systematically collect data on the device or the condition it treats, across many centres, over long time periods. Registries are powerful for long-term follow-up and for devices where the lifetime of the device exceeds the practical length of a dedicated study. Participation in an existing registry is often much cheaper than running a study, and for some device categories it is the pragmatic default.

User and clinician surveys. Structured surveys of healthcare professionals and, where applicable, patients, with pre-specified questions about clinical performance, adverse events, usability, and off-label use. Surveys are feasible for small companies and useful for devices where qualitative clinical signals matter. The weakness is that response rates are usually low and the data is self-reported.

Real-world data and post-market evidence collection. Structured collection of data from routine clinical use, including electronic health records where legally accessible, device telemetry for connected devices, and clinician-reported outcomes. Software devices and connected devices often have a natural advantage here because the data is already being generated.

Focused literature surveillance. Structured ongoing review of scientific literature relevant to the device, the technology, similar devices, and the clinical condition. Literature surveillance is the cheapest PMCF method and is required in almost every PMCF plan as at least one component, because Annex XIV Part B explicitly names screening of scientific literature as a general method.

Follow-up of equivalent or similar devices. Structured monitoring of the clinical experience with devices of the same generic group or based on equivalent technology. This is explicitly required by Annex XIV Part B — "an evaluation of the clinical data relating to equivalent or similar devices" — and it is often the element startups forget entirely.

The right combination depends on the device, the clinical claims, the risk class, the novelty, and the resources of the manufacturer. For a novel Class III implantable, a post-market study plus a registry plus literature surveillance might be the baseline. For a low-risk Class IIa software device, literature surveillance plus a structured user survey plus real-world data from the software itself might be sufficient. The PMCF plan must justify the choice, not simply name the methods.

For the methods-by-class breakdown, see PMCF methods for startups.

Writing the PMCF evaluation report

Annex XIV Part B requires the manufacturer to analyse the findings of the PMCF and document the results in a PMCF evaluation report that is part of the clinical evaluation report and the technical documentation.

The PMCF evaluation report is the document that closes the loop. It is not a re-statement of the PMCF plan. It is a report of what the plan actually produced. A well-written PMCF evaluation report has a recognisable structure, even though the Annex does not prescribe a rigid template.

It restates the PMCF plan objectives, so the reader knows what the activities were supposed to achieve. It summarises the PMCF activities actually performed during the reporting period, against the plan, with any deviations explained. It presents the data collected, organised by source and method. It appraises the data against the pre-specified criteria from the plan and the CER. It analyses the findings, including whether the clinical performance and safety remain consistent with the clinical evaluation, whether any previously unknown side-effects have emerged, whether identified risks are occurring at frequencies and severities consistent with the risk file, whether the benefit-risk ratio remains acceptable, and whether any systematic misuse or off-label use has been observed. It states the conclusions of the analysis and names the specific actions that follow — updates to the CER, updates to the risk file under EN ISO 14971:2019 + A11:2021, updates to the IFU or labelling, corrective actions under the CAPA system, or confirmation that no updates are needed with the reasoning written out.

The PMCF evaluation report is what Article 61(11) names as the input to the next clinical evaluation update. The report is also the document the notified body will read most carefully during surveillance audits of higher-class devices, because it is where the clinical feedback loop either demonstrably runs or visibly does not.

For a step-by-step guide to writing the report, see the PMCF evaluation report under MDR.

When PMCF is not required — the justification

Annex XIV Part B allows the manufacturer to state that PMCF is not applicable to a specific device, provided the PMS plan contains a justification for that decision. This is not a blanket exemption. It is a specific written statement that must address the Annex XIV Part B criteria and survive notified body review.

A PMCF non-applicability justification has to explain, in terms of the device and its clinical profile, why the PMCF objectives set out in Annex XIV Part B do not require active post-market clinical data collection for this specific device. Acceptable reasoning is narrow. It typically applies to devices where the clinical performance is fully established, the risks are fully characterised, the technology is mature, the intended purpose is narrow, and the residual clinical uncertainty is effectively zero. Even then, the justification must be written, referenced to the clinical evaluation conclusions, and signed off as part of the PMS plan.

What is not acceptable as justification: "startup resource constraints," "the device is Class I so PMCF is optional" (it is not — class alone does not exempt), "we will do it later," or "our notified body did not ask about it" (they will, eventually, and probably at the worst moment). Notified bodies scrutinise non-applicability claims carefully because a wrong non-applicability claim is the cheapest way for a clinical safety signal to go undetected.

For most startup devices with any meaningful clinical claim, PMCF is applicable in some form, and the honest move is to design the lightest defensible PMCF programme rather than to argue it away. The Subtract to Ship logic, applied here, is: do the minimum PMCF that actually satisfies Annex XIV Part B, and cut everything that does not trace to one of the Annex XIV Part B objectives.

For the non-applicability decision framework, see when is PMCF not required under MDR.

The lifecycle loop — PMCF and clinical evaluation

Article 61(11) ties the loop together. The clinical evaluation and its documentation shall be updated throughout the life cycle of the device concerned with clinical data obtained from the implementation of the manufacturer's PMCF plan referred to in Annex XIV Part B and the post-market surveillance plan referred to in Article 84. For Class III devices and implantable devices, the clinical evaluation report shall be updated at least annually with the data from the PMCF.

The loop runs like this. The clinical evaluation is performed under Article 61 and Annex XIV Part A, and the CER documents it at CE marking. The PMCF plan is written as a section of the PMS plan and as a component of the clinical evaluation strategy. The device is placed on the market. The PMCF plan executes. Data flows in from the chosen methods. The PMCF evaluation report is produced on the cadence defined in the plan. The PMCF evaluation report feeds into the next CER update. The updated CER reflects the new clinical picture. The updated CER informs the next PSUR under Article 86 (for Class IIa, IIb, and III) or the next PMS Report under Article 85 (for Class I). For Class III and implantable devices, the updated clinical conclusions also feed the next update of the Summary of Safety and Clinical Performance under Article 32. And the cycle repeats.

That loop is the backbone of the post-market clinical compliance architecture under MDR. A startup that understands the loop, even if the execution is lean, will build a file that survives audit. A startup that treats PMCF as a document exercise disconnected from the CER, the risk file, the PSUR, and the SSCP will not.

For the pillar on the clinical evaluation side of the loop, see what is clinical evaluation under MDR. For the PMS side, see what is post-market surveillance under MDR.

PMCF for software devices

Software as a medical device deserves a short note because the PMCF economics are different. A connected software device generates clinical usage data as a by-product of being used. The question is not "how do we collect data" but "which data do we already have, which pieces of it are clinically relevant, and how do we use them to answer the PMCF objectives."

Real-world evidence from the software itself — usage patterns, clinical outputs, user interactions, flagged events — can be a very strong PMCF source if the data pipeline is designed with PMCF in mind before CE marking. The Annex XIV Part B objectives still apply: confirm safety and performance, detect emerging risks, monitor identified risks, identify off-label use, keep the benefit-risk ratio current. The methods look different but the test is the same.

Software startups should design the PMCF data pipeline at the same time as the clinical evaluation plan, not after CE marking. Retrofitting a PMCF data flow onto a software architecture that was not built to support it is expensive and sometimes impossible within GDPR and data minimisation constraints.

For the software-specific walkthrough, see PMCF for software as a medical device.

Common mistakes

A short list of PMCF mistakes that recur across startup projects.

Treating PMCF as post-launch work. The PMCF plan must be written and approved before CE marking, as part of the technical documentation. Treating it as post-launch is a documentation non-conformity from day one.

Writing a PMCF plan with one generic method and no rationale. Annex XIV Part B requires a rationale for the appropriateness of the chosen methods. "User surveys quarterly" is not a rationale.

Forgetting the equivalent-or-similar-device evaluation. Annex XIV Part B explicitly requires it, and it is the element most often missing from startup PMCF plans.

Claiming non-applicability without a real justification. A non-applicability statement that does not address the Annex XIV Part B objectives is worse than not having one, because it signals to the auditor that the manufacturer did not read the Annex.

Disconnecting the PMCF evaluation report from the CER update. The whole point of PMCF is to keep the CER current. A PMCF evaluation report that does not produce a documented CER update, or a documented reasoned decision that no update is needed, has broken the loop.

Promising PMCF activities the company cannot fund. A PMCF plan that commits to a post-market study the startup has no budget for is worse than a PMCF plan that commits to a smaller set of methods that will actually run. Auditors read the plan against the evidence that it ran.

The Subtract to Ship angle — lean PMCF that closes the loop

The Subtract to Ship framework for MDR applied to PMCF produces a short rule. For each activity in the PMCF plan, trace it to a specific objective in Annex XIV Part B and to a specific part of the clinical evaluation it is meant to update. If an activity cannot trace to both, cut it. If an Annex XIV Part B objective has no activity in the plan, add the lightest activity that can genuinely address it.

A lean PMCF programme for a Class IIa startup device typically includes a structured literature surveillance process on a defined cadence with documented search terms and databases, a structured user or clinician feedback channel with pre-specified questions that map to the clinical claims, an evaluation of clinical data on equivalent or similar devices using the same literature surveillance infrastructure, integration with the PMS complaint intake so clinical signals are routed into the PMCF evaluation and not only into the CAPA system, and a defined cadence for producing the PMCF evaluation report and feeding it into the CER update cycle.

That programme traces to Annex XIV Part B line by line. It is runnable by a small team. It survives notified body surveillance because every element has an audit trail. It is not a token programme. It is the minimum real programme, and minimum real beats elaborate fictional every time.

For the lean PMCF implementation walkthrough, see minimum viable PMCF for Class IIa startups.

Reality Check — where do you stand?

  1. Do you have a written PMCF plan that is part of the technical documentation submitted for CE marking, or is PMCF a section you intend to write later?
  2. Does your PMCF plan address every element listed in Annex XIV Part B — general methods, specific methods, rationale, link to CER, link to risk file, objectives, equivalent-or-similar-device evaluation, relevant standards and guidance, time schedule?
  3. If your plan claims PMCF is not applicable, does the justification address the Annex XIV Part B objectives one by one, in writing, reviewed by someone who understands the Regulation?
  4. For each PMCF activity in the plan, can you name the specific Annex XIV Part B objective it addresses?
  5. Has your PMCF plan been reviewed against the clinical evaluation plan and the risk management file, so the three documents speak the same language about claims, risks, and acceptance criteria?
  6. Do you have a defined cadence for producing the PMCF evaluation report, and does that cadence match the update cadence of the CER (annually for Class III and implantables, proportionate for lower classes)?
  7. If you are a software as a medical device company, was the PMCF data pipeline designed before CE marking, or are you planning to retrofit it?
  8. Have you identified the equivalent or similar devices you will monitor, and do you have a method to actually monitor them?
  9. When a complaint arrives tomorrow, is there a documented rule for routing clinically relevant signals into the PMCF evaluation and not only into the CAPA system?
  10. Have you read MDCG 2025-10 for the operational context on how PMCF sits inside the broader PMS system?

Frequently Asked Questions

Is PMCF mandatory for every medical device under MDR?

PMCF is mandatory unless the manufacturer documents in the PMS plan a specific justification for why PMCF is not applicable to the device, based on the Annex XIV Part B objectives. The default is that PMCF applies. Non-applicability is the exception, and the burden of proof is on the manufacturer.

Is PMCF the same as post-market surveillance?

No. Post-market surveillance is the overall system required by Articles 83 to 86 for collecting and acting on real-world data about a device. PMCF is the clinical arm of that system, required by Annex XIV Part B, focused specifically on clinical data and clinical evaluation updates. PMCF sits inside the PMS system and feeds the clinical evaluation update cycle under Article 61(11).

Is a PMCF study the same as a clinical investigation?

A PMCF study is a specific type of clinical study conducted on a CE marked device after it is placed on the market, following EN ISO 14155:2020 + A11:2024 for the aspects that apply. A pre-market clinical investigation under Articles 62 to 82 of the MDR has different regulatory obligations and submission routes. Both generate clinical data, but they sit at different points in the device lifecycle.

How often does the PMCF evaluation report need to be produced?

The cadence is defined in the PMCF plan itself, proportionate to the risk class, the novelty of the device, and the clinical evaluation update cycle. For Class III and implantable devices, Article 61(11) requires the clinical evaluation to be updated at least annually with PMCF data, which in practice means the PMCF evaluation report runs on at least an annual cadence for those classes. For lower-risk devices, the cadence is justified in the plan.

Can a startup use literature surveillance as the main PMCF method?

For some devices, yes. Annex XIV Part B explicitly names screening of scientific literature as a general PMCF method, and for lower-risk devices with well-characterised clinical profiles, structured literature surveillance combined with user feedback and similar-device monitoring can be sufficient. For novel Class III and implantable devices, literature surveillance alone is almost never sufficient.

What is the link between PMCF and the PSUR?

Article 86 requires the Periodic Safety Update Report to state the main findings of the post-market clinical follow-up. This is a direct link: the PSUR cannot be complete without the PMCF findings being summarised inside it. For Class IIa, IIb, and III devices, the PMCF evaluation report is one of the inputs that populates the PSUR.

What is the link between PMCF and the SSCP?

Article 32 requires manufacturers of Class III devices and implantable devices (other than specific exemptions) to produce a publicly available Summary of Safety and Clinical Performance. The SSCP reflects the current clinical picture of the device, including PMCF findings, and is updated as the clinical evaluation is updated. PMCF data that changes the clinical evaluation also changes what the SSCP says to the public.

Sources

  1. Regulation (EU) 2017/745 of the European Parliament and of the Council of 5 April 2017 on medical devices, Article 32 (Summary of Safety and Clinical Performance), Article 61 (clinical evaluation), Article 61(11) (PMCF update of clinical evaluation), Articles 83 to 86 (post-market surveillance system, plan, PMS Report, and PSUR), and Annex XIV Part B (post-market clinical follow-up). Official Journal L 117, 5.5.2017.
  2. MDCG 2025-10 — Guidance on post-market surveillance of medical devices and in vitro diagnostic medical devices. Medical Device Coordination Group, December 2025.
  3. EN ISO 14155:2020 + A11:2024 — Clinical investigation of medical devices for human subjects — Good clinical practice.

This post is part of the Post-Market Surveillance & Vigilance series in the Subtract to Ship: MDR blog. Authored by Felix Lenhard and Tibor Zechmeister. PMCF is where the clinical story of the device continues after CE marking, and the gap between a plan that looks good on paper and a plan that actually runs is where most of the real risk lives.