To write a PMCF plan under MDR Annex XIV Part B, you document the general and specific methods you will use to proactively collect clinical data on your CE marked device, the rationale for choosing those methods, the explicit link to your clinical evaluation report and risk management file, the specific PMCF objectives from Annex XIV Part B that each activity addresses, an evaluation of clinical data on equivalent or similar devices, references to applicable standards and guidance, and a justified time schedule. The plan is written before CE marking, sits inside the post-market surveillance plan required by Articles 83 and 84, and feeds the clinical evaluation update cycle under Article 61(11).

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

TL;DR

  • The PMCF plan is a mandatory section of the technical documentation under Annex XIV Part B of Regulation (EU) 2017/745, embedded inside the PMS plan required by Article 84.
  • Annex XIV Part B fixes the required contents: general methods, specific methods, rationale, link to the CER and risk file, specific objectives, equivalent or similar device evaluation, applicable standards and guidance, and a justified time schedule.
  • Method selection is a function of risk class, novelty of the device, residual clinical uncertainty, and the resources available to actually run the chosen methods.
  • Proportionality is the test: the plan must be intensive enough to answer the open clinical questions and lean enough that every activity will actually run.
  • A PMCF plan that does not trace to the clinical evaluation report and the risk file has broken the Article 61(11) loop before it starts.
  • MDCG 2025-10 (December 2025) is the current operational guidance on how PMCF fits inside the PMS system, and it is the document to read alongside the Annex.

Why the PMCF plan deserves its own post

Post 181 is the pillar on post-market clinical follow-up under MDR. It walks through PMCF as a concept, the regulatory architecture, the feedback loop to clinical evaluation, and the common pitfalls. This post narrows in on one specific deliverable: the PMCF plan itself. The reason it needs its own walkthrough is that in 15 years of reviewing technical files, the PMCF plan is the document most often written badly. Not because founders are lazy. Because the Annex XIV Part B text is compact, the requirements are easy to miss on a first read, and the default startup instinct is to name a generic method and move on. A notified body auditor reading such a plan will press on every missing element, and each missing element is a non-conformity waiting to be written.

This post is the how-to. Read the pillar for the why.

The purpose of the PMCF plan

The PMCF plan exists for one reason: to specify, in advance, how the manufacturer will keep the clinical picture of a CE marked device current throughout its lifetime. Annex XIV Part B of Regulation (EU) 2017/745 defines PMCF as a continuous process that updates the clinical evaluation referred to in Article 61 and Annex XIV Part A, and that is specifically addressed in the manufacturer's post-market surveillance plan. The plan is the proactive commitment. It says what the manufacturer will do, with which methods, on what cadence, to answer which clinical questions, so that when PMCF findings arrive, they can be compared against a pre-specified set of objectives and criteria and turned into documented updates of the clinical evaluation report.

Without the plan, PMCF becomes reactive. Data arrives, gets filed somewhere, and nothing happens until a complaint escalates or an audit triggers a retrospective effort to reconstruct what the manufacturer "meant" to do. That is the scenario the Annex is written to prevent. Article 61(11) reinforces the point: 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. The loop only runs if the plan exists and is implemented.

The required contents of a PMCF plan

Annex XIV Part B specifies the contents of the plan. A compliant plan addresses, at a minimum, every element below. The order in the document is a matter of style. The presence of every element is not.

The general methods and procedures of the PMCF to be applied. This is the baseline layer of activities that apply to almost every plan: the gathering of clinical experience gained with the device, feedback from users, screening of scientific literature, and screening of other sources of clinical data. Name the specific channels you will use, not the category.

The specific methods and procedures of the PMCF to be applied. This is the device-specific layer: evaluation of suitable registers, PMCF studies, or other structured activities designed to answer the specific open clinical questions for this device. If the device has a specific residual uncertainty, this is where the activity that addresses it lives.

A rationale for the appropriateness of the methods and procedures applied. This is the element most often missing. The plan has to explain why the chosen methods are appropriate for this device, these risks, these claims, and this class. A list of methods with no rationale is not a plan. It is a menu.

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 plan must point explicitly to the CER sections and the risk file sections it is designed to keep current. If a risk in the risk file has no PMCF activity backing it, the plan has a gap. If the plan names an activity with no link to the CER or the risk file, the activity has no justification.

The specific objectives to be addressed by the PMCF. Annex XIV Part B lists the PMCF objectives the plan must aim at: confirming the safety and performance of the device throughout its expected lifetime, identifying previously unknown side-effects and monitoring 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. Each activity in the plan must trace to at least one of these objectives.

An evaluation of the clinical data relating to equivalent or similar devices. Explicitly required, explicitly often forgotten. The plan must describe how the manufacturer will monitor clinical experience with devices of the same generic group or based on equivalent technology, and what will be done with the findings.

Reference to any relevant common specifications, harmonised standards when used by the manufacturer, and relevant guidance on PMCF. Where PMCF activities qualify as clinical investigations, EN ISO 14155:2020 + A11:2024 applies for the aspects that apply. MDCG 2025-10 is the operational guidance on PMS and PMCF. The plan names the standards and guidance it follows.

A detailed and adequately justified time schedule for PMCF activities. The time schedule specifies when each activity runs, when interim analyses happen, and when the PMCF evaluation report is produced. "Ongoing" is not a time schedule. "Annual literature search with a PMCF evaluation report produced within three months of each search" is.

How to select PMCF methods

Method selection is where proportionality becomes concrete. The Annex names a menu of methods; the plan chooses from it. The selection is driven by four factors.

Risk class. Class III and implantable devices carry the heaviest PMCF expectations because Article 61(11) requires the clinical evaluation for these classes to be updated at least annually with PMCF data, and Article 86 names PMCF findings as a mandatory element of the PSUR. For these classes, a structured post-market study, registry participation, or both are often the baseline. For Class IIb novel devices, similar expectations may apply depending on the clinical questions. For Class IIa and well-characterised Class IIb devices, lighter combinations can be proportionate. For Class I devices with clinical claims, structured literature surveillance, similar-device monitoring, and user feedback are usually the starting point.

Novelty. A novel device with limited pre-market clinical evidence carries more residual clinical uncertainty than a well-characterised device with decades of equivalent-device experience. The more novel the device, the more intensive the PMCF activities need to be, regardless of class.

Open clinical questions. The clinical evaluation report states the conclusions of the pre-market clinical evaluation and, implicitly or explicitly, the open questions that pre-market evidence could not close. Each open question is a target for a PMCF activity. Matching activities to open questions one-for-one is the cleanest way to justify the plan.

Resources that actually exist. An elaborate PMCF plan the manufacturer cannot fund or staff is worse than a lean plan that runs. Auditors assess the plan against the evidence that it ran. A post-market study named in the plan and never started is a documented failure. A literature search named in the plan and performed on schedule is a documented success. Method selection must respect the reality of what the company can execute.

The typical menu — post-market clinical studies, registries, user and clinician surveys, real-world data from connected devices, focused literature surveillance, and follow-up of equivalent or similar devices — is covered in post 181. For a class-by-class breakdown, see PMCF methods for startups. Where PMCF activities qualify as clinical investigations, EN ISO 14155:2020 + A11:2024 applies for the aspects that apply; see PMCF studies and EN ISO 14155 under MDR.

Proportionality — the test the plan must pass

Proportionality is the principle that the intensity of PMCF activities must match the clinical risk and the residual uncertainty of the device. An over-intensive plan is not safer; it is wasteful and, because it will not actually run, is arguably less safe than a lean plan that does. An under-intensive plan leaves clinical questions unanswered and is a non-conformity against Annex XIV Part B.

The proportionality test has three questions. Does each Annex XIV Part B objective have at least one activity that can genuinely address it for this device? Is each activity staffed and funded at a level where it will actually run on the stated cadence? Can the combination of activities produce a PMCF evaluation report substantive enough to feed a meaningful CER update? If the answer to any of the three is no, the plan is not proportionate — either too thin or too fictional.

For Class III devices, proportionate usually means more, not less. For Class IIa software devices with real-world data pipelines built in, proportionate can mean lean but structured. The label is not the test. The test is whether the open clinical questions get answered.

For the proportionality decision framework, see PMCF proportionality for startups.

The PMCF plan is not a standalone document. It is one vertex of a triangle that also includes the clinical evaluation report and the risk management file under EN ISO 14971:2019 + A11:2021. Annex XIV Part B requires the plan to reference the relevant parts of the CER and the risk file explicitly, and the reason is structural: if the three documents speak different languages about claims, risks, and acceptance criteria, the post-market clinical loop cannot close.

Concretely, the plan should identify for each clinical claim in the CER which PMCF activity will monitor whether the claim continues to hold in real-world use. It should identify for each major residual risk in the risk file which PMCF activity will monitor the frequency and severity of the risk and confirm that the pre-market assumptions remain valid. It should identify the acceptance criteria — the thresholds at which a finding triggers a CER update, a risk file update, a labelling change, or a corrective action. Those criteria belong in the plan, not invented later when findings arrive.

For the integration walkthrough, see integrating PMCF with risk management and CER.

Common PMCF plan mistakes

Six mistakes recur across startup PMCF plans. Avoid them in the draft and the review effort shrinks dramatically.

Naming one generic method with no rationale. "Quarterly user surveys" is not a plan. Annex XIV Part B requires a rationale for the appropriateness of the chosen methods. Write the rationale.

Omitting the equivalent or similar device evaluation. Explicitly required by Annex XIV Part B, explicitly the element most often missing. Include it, name the devices or device group, and describe the method.

No link to the CER or the risk file. If the plan does not point to specific CER sections and risk file entries, it cannot close the loop Article 61(11) requires. Add the references.

No acceptance criteria for PMCF findings. Without pre-specified thresholds, a finding cannot be compared to anything, and the PMCF evaluation report becomes narrative instead of analysis. Pre-specify the criteria.

A time schedule that says "ongoing." "Ongoing" is not a schedule. Specify cadences — monthly, quarterly, annual — and the trigger dates for interim analyses and the PMCF evaluation report.

Committing to activities the company cannot execute. A plan that names a multi-centre registry the manufacturer has no contract for is a plan that will fail at audit. Commit to what will actually run.

The Subtract to Ship angle — lean plans that close the loop

The Subtract to Ship framework for MDR applied to the PMCF plan produces a short rule. For each activity in the plan, trace it to a specific Annex XIV Part B objective and to a specific CER section or risk file entry. If the trace does not exist, cut the activity. For each Annex XIV Part B objective and for each major clinical claim or residual risk, name the lightest activity that can genuinely address it. If no activity exists, add one. The plan that results is the minimum real plan — lean, but every element traces and every element runs.

Minimum real beats elaborate fictional every time. The notified body auditor who reads a lean plan with clean tracing and running evidence is happier than the auditor who reads a bloated plan with no evidence behind it. The file that survives is the file that runs.

Reality Check — where do you stand?

  1. Does your PMCF plan address every element of Annex XIV Part B — general methods, specific methods, rationale, link to CER, link to risk file, objectives, equivalent or similar device evaluation, standards and guidance, time schedule — or can you name an element that is missing?
  2. For each activity in the plan, can you name the specific Annex XIV Part B objective it addresses and the specific CER section or risk file entry it is meant to keep current?
  3. Are the acceptance criteria for PMCF findings pre-specified in the plan, or will they be decided retrospectively when findings arrive?
  4. Does the time schedule specify cadences and trigger dates, or does it rely on words like "ongoing" and "continuous"?
  5. Have you evaluated clinical data on equivalent or similar devices, and is the method for doing so described in the plan?
  6. If a PMCF study is named in the plan, do you have the budget, the centres, and the protocol to actually start it?
  7. Does the plan reference EN ISO 14155:2020 + A11:2024 for the aspects that apply to any PMCF activity that qualifies as a clinical investigation?
  8. Does the plan reference MDCG 2025-10 as the operational guidance on how PMCF fits inside the PMS system?
  9. Has the plan been reviewed side by side with the CER and the risk file, so the three documents speak the same language about claims, risks, and acceptance criteria?
  10. If the plan had to be executed by the current team starting tomorrow, would every activity actually happen on schedule?

Frequently Asked Questions

When must the PMCF plan be written?

Before CE marking. The PMCF plan is part of the technical documentation and a component of the PMS plan required by Articles 83 and 84. Writing it after CE marking is a documentation non-conformity that a notified body auditor will find immediately.

Is the PMCF plan a separate document from the PMS plan?

The PMCF plan is structurally a section of the PMS plan, but it is often maintained as a separate document that the PMS plan references. Either form is acceptable as long as the PMCF plan contents required by Annex XIV Part B are all present and the link between the PMS plan and the PMCF plan is clear.

How often must the PMCF plan be updated?

The plan is updated whenever the PMCF activities change — new methods, new cadences, new objectives — or whenever PMCF findings reveal that the existing plan does not adequately cover the residual clinical questions. The Regulation does not set a fixed calendar cadence for the plan itself; it is updated when the underlying reality changes.

Does the PMCF plan need to reference EN ISO 14155?

Where the PMCF plan includes activities that qualify as clinical investigations — typically post-market clinical studies with human subjects — EN ISO 14155:2020 + A11:2024 applies for the aspects that apply, and the plan references it. For PMCF activities that are not clinical investigations (literature surveillance, registry participation, user surveys), EN ISO 14155 may not be directly applicable, but the plan still references the standards and guidance it does follow.

Can a PMCF plan be lean and still pass audit?

Yes, if every element required by Annex XIV Part B is present, every activity traces to an objective and to the CER and risk file, and every activity actually runs. Lean plans that meet these tests pass audit. Bloated plans that do not run fail audit. The discriminator is not volume but tracing and execution.

What is the difference between the PMCF plan and the PMCF evaluation report?

The PMCF plan is written before CE marking and specifies what will be done. The PMCF evaluation report is produced as the plan runs and documents what actually happened, what the findings were, and what they mean for the CER, the risk file, the labelling, and the benefit-risk determination. The plan is the commitment; the report is the delivery.

Sources

  1. Regulation (EU) 2017/745 of the European Parliament and of the Council of 5 April 2017 on medical devices, Article 61(11) (PMCF update of clinical evaluation), Articles 83 to 86 (post-market surveillance system, plan, PMS Report, and PSUR), Annex I Sections 1 and 9 (benefit-risk determination within the general safety and performance requirements), Annex XIV Part A (clinical evaluation), 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.
  4. EN ISO 14971:2019 + A11:2021 — Medical devices — Application of risk management to medical devices.

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. A PMCF plan is the document where the post-market clinical story of the device is committed to, and the gap between a plan that names methods and a plan that actually answers the open clinical questions is where most of the real audit risk lives.