PMCF: Understanding Post-Market Clinical Follow-Up for Medical Devices

Obtaining CE marking is not the end of the clinical story for a medical device. Once the device is on the market, the manufacturer must continue to verify that safety, performance, and clinical benefit remain controlled under real-world conditions of use.

This is the role of PMCF, for Post-Market Clinical Follow-up.

PMCF is not only an MDR deliverable. It is a continuous process of collecting and analyzing clinical data on a device that is already CE marked and used according to its intended purpose. It feeds the clinical evaluation, post-market surveillance, and monitoring of the benefit-risk ratio throughout the device lifecycle.

One important nuance should be made from the start: for in vitro diagnostic devices covered by the IVDR, the term is not PMCF but PMPF, for Post-Market Performance Follow-up. The logic is similar, but the terminology and expectations focus on performance.

Short answer: choose the PMCF method from the clinical question that stays open after your clinical evaluation, not from what is easiest to run. Literature screening and user feedback cover general questions. A registry covers long-term performance and rare events. A dedicated PMCF study answers a precise question that no other source can. MDR Annex XIV Part B requires you to justify that choice in the PMCF plan.

What Is PMCF?

In practical terms, PMCF consists of proactively collecting and evaluating clinical data from the use of the device after it has been placed on the market.

These data are used to verify that the observed performance and safety remain consistent with the initial clinical evaluation. They also make it possible to identify emerging risks, rare adverse events, residual clinical questions, or uses that may not have been sufficiently documented before commercialization.

PMCF is therefore not limited to monitoring incidents. It aims to generate or complete clinical evidence through a planned approach that is proportionate to the device.

PMCF, PMS, and Vigilance: What Are the Differences?

PMCF, PMS, and vigilance are connected, but they do not serve the same purpose.

PMCF therefore does not replace PMS or vigilance. It provides a specific clinical component: confirming safety, performance, and the benefit-risk ratio over time.

[[cta]]

Which Regulatory Framework Applies to PMCF?

PMCF is part of the requirements of MDR 2017/745 on clinical evaluation and post-market surveillance. It is described in particular in Annex XIV, Part B, which presents PMCF as a continuous process for updating the clinical evaluation.

Two MDCG guidance documents are particularly used by manufacturers and notified bodies:

• MDCG 2020-7, for the PMCF plan
• MDCG 2020-8, for the PMCF evaluation report

These documents do not replace the MDR, but they provide an expected structure for documenting the PMCF strategy, collection methods, results obtained, and their impact on the clinical evaluation.

PMCF must also remain connected to PMS documentation. Depending on the device class, post-market data feed into the PMS Report or the PSUR, as well as updates to the CER, the Clinical Evaluation Report.

What Is PMCF Used For in Practice?

PMCF is first used to confirm that safety, performance, and clinical benefit remain valid over time. The data collected after commercialization may relate to broader populations, more varied practices, or contexts of use that differ from those documented before CE marking.

It also helps detect clinical signals that were not visible or sufficiently documented at the start: emerging risks, rare effects, late effects, misuse, or off-label uses.

Finally, it feeds regulatory documentation: CER, PMS documentation, and, where applicable, PSUR. PMCF therefore works as a continuous loop: plan, collect, analyze, document, update, and then reassess.

Which Deliverables Are Expected?

PMCF is materialized through several key documents. They must be consistent with each other and integrated into the device technical documentation. For the data side, you write a data management plan before collection starts.

In practice, the PMS Report applies to Class I devices, while the PSUR applies to Class IIa, IIb, and III devices.

The challenge is not only to produce these documents. It is to show that the data collected are relevant, analyzed, and used to update the device evaluation.

Which Methods Can Be Used to Collect PMCF Data?

Collection methods depend on the device, its risk class, its clinical history, and the questions that remain open.

The most structured approaches include dedicated PMCF studies, post-market observational studies, registries, cohort follow-up, or multicenter studies. These formats are often necessary when the available data are not sufficient to meet clinical or regulatory expectations.

Other methods can complement data collection: patient questionnaires, healthcare professional questionnaires, quality-of-life measures, real-world use data, performance data in routine clinical practice, or longitudinal follow-up.

ePRO tools can be useful when PMCF relies on patient-reported data: symptoms, quality of life, satisfaction, tolerance, ease of use, or long-term perception. The value lies in structuring collection, limiting manual entry, and facilitating follow-up over time.

Depending on the PMCF protocol, an eCRF can also be used to collect data entered by investigators or healthcare professionals. In some cases, hybrid or remote approaches can also help follow participants without multiplying travel constraints.

How Do You Choose a PMCF Method?

MDR Annex XIV Part B asks the PMCF plan to state two kinds of methods: general methods (gathering clinical experience and user feedback, screening scientific literature and other sources of clinical data) and specific methods (evaluating suitable registers, running PMCF studies). The plan must also give the rationale for why these methods suit the device. MDCG 2020-7 provides the expected structure for that plan.

Start from the open questions in your Clinical Evaluation Report, then match each question to the lightest method that can answer it.

Four checks keep the choice defensible with a notified body:

• Every method answers a named question from the CER. A method with no question behind it draws findings.
• The intensity is proportionate to the device class, its clinical history, and its residual risks. A PMCF you consider unnecessary needs a documented justification in the clinical evaluation.
• The data you collect can feed the PMCF evaluation report (MDCG 2020-8) and the next CER update without re-work.
• The data management plan is written before the first patient is followed.

For long-term follow-up across many sites, an electronic data capture system and ePRO collect investigator and patient data in the same study. Datacapt also supports registries, and the medical device page describes how teams set up post-market studies.

What Does the Proposed MDR Revision Change for PMCF?

On 16 December 2025, the European Commission published a proposal for a targeted revision of the MDR and IVDR, covering simplification, notified body capacity and device shortages. It remains a proposal. The European Parliament and the Council are still examining it, and it creates no obligation today.

Until a final text is adopted, MDR Annex XIV Part B, MDCG 2020-7 and MDCG 2020-8 remain the reference for your PMCF plan and evaluation report. Build the plan on the current text, and revisit it if the adopted text changes the requirements.

Can PMCF Be Adapted?

PMCF is expected under the logic of the MDR, but its level of intensity must be proportionate. Not all devices require the same volume of data or the same collection methods.

The intensity of PMCF depends in particular on:

• the device risk class
• its clinical history
• the data already available
• the level of innovation
• residual risks
• remaining clinical uncertainties
• notified body expectations

When the manufacturer considers that PMCF is not necessary, this position must be justified and documented in the clinical evaluation. In practice, this justification is harder to defend for high-risk devices, implantable devices, or devices for which clinical data remain limited.

What Are the Risks of Insufficient PMCF?

Poorly designed or poorly executed PMCF can weaken the entire clinical dossier.

The consequences can be concrete: insufficient clinical data, difficulty confirming the benefit-risk ratio, notified body findings, requests for additional data, delays in certification processes, or difficulties maintaining an up-to-date clinical evaluation.

The risk is not only documentary. If post-market data are not collected or analyzed correctly, the manufacturer may miss useful signals on safety, performance, real-world use, or the limits of the device.

Why Digitize PMCF Data Collection?

PMCF can quickly become complex when data remain scattered across Excel files, emails, paper questionnaires, separate databases, and manual exports. This fragmentation complicates analysis, reduces traceability, and increases the workload needed to prepare regulatory deliverables.

A digital approach can help structure collection from the start: standardized forms, longitudinal follow-up, eCRF and ePRO data, reminders, dashboards, usable exports, and traceability of actions. The benefits of an eCRF solution apply to PMCF as to any other study.

The objective is not to digitize for the sake of digitizing. The challenge is to secure the data chain, from collection through to analysis, in order to facilitate updates to the CER, PMCF report, PMS Report, or PSUR.

PMCF is an essential component of post-market clinical follow-up for medical devices under the MDR. It helps maintain the clinical evaluation up to date, document the benefit-risk ratio, and respond more robustly to notified body expectations.

Its value depends above all on the quality of the strategy implemented: relevant clinical questions, appropriate methods, structured collection, regular analysis, and integration of results into regulatory documentation.

When well designed, PMCF helps the manufacturer monitor the device over time, with clinical data that are genuinely usable.