Client login Discuss your program

Post-Market Clinical Evidence, EU MDR

PMCF Study and Post-Market Evidence CRO

Design and execute prospective PMCF studies, retrospective evidence programs and medical-device registries that address defined clinical evidence gaps and feed directly into ongoing clinical evaluation.

Discuss Your PMCF Study Request a PMCF Study Design Workshop
Reviewers who previously assessed clinical evidence inside a notified body Submissions prepared and filed in house Class IIa to Class III

How the program is built

  1. The CER or PMCF plan identifies an unresolved clinical question
  2. Existing PMS, literature and complaint data cannot answer it
  3. The evidence model is selected against that question
  4. The program is designed around the question, not around a template
  5. European sites, data and biometrics deliver the evidence
  6. Output feeds the PMCF Evaluation Report and the clinical evaluation

The decision

What Clinical Question Must Your PMCF Program Answer?

Before study type, sample size or budget, one question decides everything downstream: what remains genuinely unresolved about this device in clinical use.

Safety

Do residual or emerging risks require structured clinical follow-up?

Performance

Does routine clinical use continue to confirm the intended clinical performance?

Long-term evidence

Are durability, survivorship, revision, recurrence or long-term performance questions unresolved?

Clinical use

Are there unanswered questions around patient selection, subgroups, clinical pathways or routine use?

The evidence model should follow the unresolved clinical question, not the other way around.

Threshold

When General PMCF Activities Are No Longer Enough

Surveys, complaint analysis, literature monitoring and registry scanning are legitimate PMCF activities. They stop being sufficient at the point where a device-specific clinical question needs device-specific clinical data.

  • The clinical evaluation identifies a device-specific clinical evidence gap
  • Long-term performance needs confirmation beyond the available follow-up
  • Residual risk requires structured, prospective follow-up
  • Class III or implantable device evidence remains insufficient for the claimed indications
  • Specific populations or subgroups require additional data
  • Objective endpoints or imaging are required and cannot be collected passively
  • Literature, PMS, complaints and surveys cannot answer the question for this device
  • A notified body question requires new device-specific clinical evidence

Do Not Commission a Registry Simply Because "PMCF Requires a Registry"

A registry is one evidence model among four. It is the right instrument when the question is longitudinal and depends on large routine-use populations observed over years. It is the wrong instrument when the question needs predefined endpoints, standardized visits or imaging that routine care does not produce. Committing to a registry before the clinical question is defined usually produces a long, expensive dataset that still does not close the gap. The clinical question decides whether the correct model is prospective PMCF, a registry, a retrospective program or a hybrid.

Model selection

Choose the PMCF Model That Matches the Evidence Gap

Four models, four different questions. Read the left column first: if none of those gap signatures matches your device, the model is not the right one, whatever the budget says.

Gap signature

The endpoint does not exist in routine records and has to be created: defined visits, defined measurements, defined timing.

Prospective PMCF study

For predefined endpoints, standardized visits, prospective safety and performance data, imaging, PROMs and structured follow-up. The protocol decides what is measured, so the data can be interpreted against a stated hypothesis rather than reconstructed after the fact.

Proven in practice: Case 01, RegenLab

Choose when

The evidence must be comparable across countries and sites, and the gap is precise enough to state as an endpoint.

Gap signature

The question is about what happens over years, in the whole treated population, not in a selected sample.

Medical device registry

For longitudinal real-world evidence, large routine-use populations, survivorship, revision, recurrence and long-term implant follow-up. A registry earns its cost through denominator, continuity and traceability, not through volume alone.

Design and delivery model, section below

Choose when

Follow-up must outlast any single study, and the endpoints are events that routine care already records.

Gap signature

The device has been in use long enough that the answer may already sit in existing clinical records.

Retrospective evidence program

When existing records may contain sufficient device-specific clinical data, subject to feasibility, missingness and selection-bias assessment. The feasibility work is the deliverable that protects the budget, not an administrative step before it.

Assessed before it is proposed, never assumed

Choose when

Device identification, consecutive cases and endpoint availability survive a pilot chart review.

Gap signature

Part of the answer exists historically, part of it has to be collected going forward.

Hybrid model

When the most efficient answer combines prospective follow-up, retrospective records, registry data or external linkage. Hybrid designs are demanding to defend, so the boundary between the historical and prospective components has to be defined before enrollment, not negotiated afterwards.

Requires an explicit data-provenance model

Choose when

Historical data cover the baseline but not the endpoint, or cover the endpoint but not long enough.

Where observational real-world programs sit. They draw on routine clinical use as the evidence source without building registry infrastructure, which makes them the efficient answer when the question is device-specific rather than population-wide. Case 02 below is a portfolio-scale example across three device families.

Prospective PMCF

Prospective PMCF Studies Designed Around the Evidence Gap

Every design decision below is tested against one question: can this program close the defined clinical evaluation and PMCF evidence gap? A study that cannot answer that question is expensive whatever it costs.

Define

From evidence gap to protocol

  • Evidence objective stated as an answerable question
  • Protocol built to that objective
  • Endpoint framework, primary, secondary and safety
  • Sample-size assumptions traced to the endpoint

Enable

From protocol to open sites

  • Country and site feasibility against the real population
  • Site selection and investigator engagement
  • Regulatory and ethics submissions, prepared and filed in house
  • Contracting and site activation

Deliver

From first patient to reportable evidence

  • On-site and central monitoring proportionate to risk
  • EDC and eCOA configured to the endpoint framework
  • Imaging workflows where the endpoint requires them
  • Safety and device deficiency handling
  • Data management, biostatistics and reporting

The test applied at every design gate is the same one the notified body will apply to the output: does this dataset answer the question the clinical evaluation could not? Scope that does not serve that answer is removed rather than priced.

Registries

Continuous Data Collection Built Around the Evidence Question

"Registry" describes four quite different instruments. Choosing the wrong one is the most common and most expensive mistake in long-term device evidence.

Model 01

Sponsor-owned registry

The manufacturer defines the dataset, the follow-up and the governance. Maximum control over endpoints and traceability, and full responsibility for recruitment, retention and data quality.

Model 02

Existing national or specialty registry

Where a national or specialty registry already captures the population, the first task is to evaluate whether its dataset, device granularity and follow-up can answer the question. Often faster and more defensible than building a parallel structure.

Model 03

Registry-linked study

A defined study sits alongside or inside a registry, adding the variables the registry does not hold while inheriting its denominator and its continuity.

Model 04

Long-term implant registry

Longitudinal follow-up of implanted devices, where survivorship, revision and time-to-event analysis are the point and the horizon is measured in years.

What Actually Determines Registry Quality

A registry is only as defensible as its denominator and its follow-up. These are the parameters that decide whether the output can be used in a clinical evaluation, and they are set at design time, not recovered later.

  • Eligibility definition and a denominator you can defend
  • Consecutive enrollment rather than convenience sampling
  • Device and UDI traceability down to the implanted unit
  • Longitudinal follow-up model and realistic retention assumptions
  • Endpoint definitions stable across sites and across years
  • Site contribution variability, and what it does to the analysis
  • Loss to follow-up, measured and reported rather than absorbed
  • Data completeness rules and missing-data handling
  • Governance, oversight and publication rules agreed up front
  • Comparator suitability where a comparison is claimed
  • PROMs and imaging where the endpoint depends on them

On proof. Eclevar designs registries and runs the clinical, data and biometric operations that registries depend on. No delivered Eclevar registry case is presented anywhere on this page, because none is documented for publication. The case studies below are what they say they are: prospective and observational PMCF programs.

Retrospective

Retrospective PMCF When the Evidence Already Exists in Clinical Records

Retrospective collection is chosen because the records can answer the question, not because it looks faster or less expensive. The difference between those two reasons is usually discovered late and paid for twice.

Assessed before any scale-up

  • Device identification, can the exact device be isolated in the record
  • Consecutive case identification, or a defensible reason it is not possible
  • Endpoint availability in the source records
  • Follow-up completeness over the required horizon
  • Missing data, quantified by category rather than described
  • Safety-event documentation and how it was captured
  • Site record quality and consistency across sites
  • Selection bias, and whether it can be mitigated or only disclosed

The decision gate

Pilot chart review and data-availability assessment

Proceed retrospective

Use a hybrid or prospective collection

One outcome or the other, decided on evidence

The pilot answers a measurable question: of the records that exist, how many are analyzable against this endpoint. That number, reported per site, is what a retrospective budget should be built on.

Delivery

PMCF Site Operations Across Europe

One accountable CRO for protocol, sites, data and evidence integration. The handoffs between those four are where post-market programs usually lose time, and they do not exist here.

Before activation

  • Country feasibility against the actual treated population
  • Investigator identification and engagement
  • Site qualification
  • Contracting
  • Regulatory and ethics submissions

In conduct

  • Site activation
  • CRA monitoring
  • Central monitoring and risk signals
  • Recruitment tracking against forecast
  • Retention, which decides a longitudinal program

Through to close

  • Data quality at source
  • Site performance management
  • Close-out

Submissions across these programs have been prepared and filed by Eclevar rather than subcontracted, which is why country strategy and site strategy are decided together.

Data architecture

PMCF Data Designed for Longitudinal Evidence

The value is not that a sponsor is given an EDC. The value is that the dataset is designed so it can actually answer the PMCF question and be carried into the clinical evaluation without reconstruction.

Capture

  • EDC configured to the endpoint framework, not to a template
  • eCOA and PROMs where the endpoint is patient-reported
  • Device and UDI traceability carried in the record
  • Imaging acquisition and transfer where required

Hold over time

  • Longitudinal follow-up structure across the full horizon
  • Visit windows defined so that late visits stay analyzable
  • Coding applied consistently across years and sites
  • Missing-data control, measured by category and reported

Make usable

  • Central review of the variables the endpoint depends on
  • Registry reconciliation where a registry is part of the design
  • Analysis-ready datasets with documented provenance

Eclevar runs data management and biostatistics on Milo Studio, its own EDC platform, so endpoint changes reach the database without a vendor negotiation.

Closing the loop

Close the Evidence Loop From PMCF Question to Clinical Evaluation

The sponsor is not buying a standalone study report. The program should produce evidence that can be integrated into the post-market clinical evidence system, which is where the value is realized or lost.

  1. 01CER or PMCF evidence gap
  2. 02PMCF question, stated so it can be answered
  3. 03Study, registry or retrospective program
  4. 04Clinical data
  5. 05Analysis
  6. 06PMCF Evaluation Report

Feeds directly into

Clinical Evaluation Report

PSUR

SSCP where applicable

Risk management file

Medical writing sits inside the program rather than after it. The people who will write the PMCF Evaluation Report are involved when the endpoints are defined, which is the only reliable way to avoid a dataset that is complete, expensive and still not usable in the clinical evaluation. The loop closes, or the evidence was not worth generating.

Evidence delivered

PMCF Evidence Programs Delivered in Practice

Different evidence gaps require different study models. The examples below show prospective multicountry PMCF delivery and observational real-world evidence programs across medical-device categories.

Case 01 Prospective PMCF, advanced wound care

RegenLab

Autologous platelet-rich plasma, chronic wound care

Client

Multicountry Prospective PMCF Across 14 Sites in 5 European Countries

The evidence need

Post-market clinical follow-up had to generate prospective data on device use in routine European practice, in two chronic wound populations that are not interchangeable. The program was therefore built as two parallel cohorts, diabetic foot ulcer and venous leg ulcer, rather than one pooled observation.

Study architecture

  • 160Participants
  • 14Clinical sites
  • 5European countries

Cohort ADiabetic foot ulcer

Cohort BVenous leg ulcer

SharedOne standardized European delivery model, identical evidence lines in both cohorts

OutputProspective PMCF dataset, reported by indication

Four-step preparation pathway: venous blood collection in a closed single-use tube, standardized centrifugation on a single fixed protocol, separation into autologous platelet-rich plasma above a red cell fraction, and application to the wound bed, assessed at each visit.
The standardized preparation and application pathway. Holding this sequence constant is what allows records collected in five countries to be compared. Schematic illustration, not study data.
Schematic of the plantar surface of a foot, marking the plantar forefoot as the assessed site and the heel as a secondary location.
Cohort A, diabetic foot ulcer. Anatomical schematic, no patient images.
Schematic of a lower leg below the knee, marking the gaiter area as the assessed site and the medial malleolus as a landmark.
Cohort B, venous leg ulcer. Anatomical schematic, no patient images.

What Eclevar did

  • PMCF strategy, protocol and endpoint architecture
  • Regulatory and ethics submissions prepared and filed in house, not subcontracted
  • Country and site delivery across five European countries
  • Site activation and monitoring
  • Data management on Milo, Eclevar's own EDC platform
  • Statistical analysis and final study reporting

Why it was operationally difficult

  • Five national regulatory and ethics routes, each with its own submission requirements
  • Fourteen sites operating inside different national wound-care pathways
  • Two indication-specific populations that had to stay separable in the data
  • One data model, so that records from five countries remained comparable
  • Clinical operations and data management run by one accountable team

Intended evidence use

The program feeds prospective post-market clinical data into the sponsor's ongoing clinical evaluation and PMCF reporting cycle, reported per indication rather than pooled. Eclevar delivered the study and the reporting. Conformity decisions remain with the manufacturer and its notified body.

Still frame from a recorded discussion: Antoine Turzi of RegenLab seated in conversation. Recorded discussion

Client perspective

RegenLab on running a multicountry PMCF program

A recorded discussion with the client about the European PMCF program covering diabetic foot ulcers and venous leg ulcers.

Eclevar, with its tailor-made approach and advanced Milo Studio platform, represents a major strategic asset.

Antoine Turzi, Chief Executive Officer, RegenLab

Case 02 Observational PMCF, Class III implantables

Perouse Medical, Vygon Group

Class III implantable vascular devices

Client

Six Registered Observational PMCF Studies Across a Class III Vascular Portfolio

The evidence need

Three families of Class III implantable vascular devices each required device-specific post-market clinical evidence. Run as unrelated studies, they would have produced three incompatible datasets, three sets of definitions and three governance chains for one manufacturer.

  • Vascular patches
  • Vascular prostheses
  • Implantable ports

The evidence model

Six registered observational studies, built on one harmonized evidence model applied across the three device families. This is an observational program, not a registry. The distinction matters on this page: the model was selected because routine clinical use could answer the question, not because longitudinal registry scale was required.

What Eclevar contributed

  • Portfolio-level evidence architecture across three device families
  • Harmonized observational study design, one model applied six times
  • Study registration and supporting documentation
  • Clinical operations and structured data collection
  • Analysis and PMCF reporting feeding clinical evaluation

Why this case matters

A structured observational program can be the more appropriate route when the evidence question and the available clinical use support it. The reusable asset here is the architecture: consistent definitions, traceable records and one governance model across a portfolio, instead of six unrelated studies that happen to share a manufacturer.

Intended evidence use

Eclevar structured the program architecture and supports delivery. The intended use of the output is device-family-specific post-market clinical evidence for ongoing clinical evaluation. No conformity assessment outcome, notified body position or market-access result is claimed or implied.

Capability Registry delivery model

When the Evidence Question Requires a Registry

The two programs above are studies. Some evidence questions cannot be answered by a study at all, because they depend on how a device behaves in large routine-use populations over years rather than months. That is where a registry is the right instrument, and where Eclevar designs one.

Signals that point to a registry

  • Longitudinal follow-up beyond a conventional study horizon
  • Large routine-use populations rather than a selected sample
  • Implant survivorship and revision
  • Recurrence and long-term safety signals
  • Device traceability down to the implanted unit
  • Patient-reported outcomes collected repeatedly over time
  • Imaging held at the site and needed for the endpoint
  • Linkage to an existing specialty or national registry

What Eclevar designs and delivers

  • Registry strategy and scientific design
  • Registry protocol and the applicable qualification route
  • Dataset, variable definitions and traceability rules
  • Site network design and investigator engagement model
  • EDC and data architecture on Milo
  • Follow-up and retention model across the registry horizon
  • Statistical analysis plan, interim and final reporting
  • Evaluation of whether an existing registry already answers the question

Registry design starts with whether a registry is warranted at all. Where an existing specialty or national registry already captures the population and the variables, evaluating that registry is usually faster and more defensible than building a parallel one.

Supporting proof, PMCF into clinical evaluation

PMCF Design Integrated With Clinical Evaluation Requirements

Client: Mölnlycke

Clinical evidence strategy and PMCF programs across an advanced wound care portfolio under EU MDR. Eclevar ran the gap analysis against MDCG guidance, designed the PMCF studies and authored the clinical evaluation reports, so that the evidence generated was the evidence the clinical evaluation actually needed. Two distinct multicenter PMCF programs were designed in France, one in diabetic foot ulcer and one in venous leg ulcer. The output supports successive clinical evaluation cycles.

01Evidence gap identified against MDCG guidance

02PMCF study designed to close that specific gap

03Clinical evidence generated

04Integrated into the clinical evaluation report

Eclevar does not force every PMCF evidence gap into the same study model. We define the clinical question, select the evidence model and deliver the data through to PMCF reporting and clinical evaluation integration.

Discuss Your PMCF Study

Accountability

The People Behind the PMCF Evidence Program

Five functions decide whether a post-market evidence program is defensible. These are the people who hold them, and the decision each one influences.

Mark Da Costa

Chief Operating Officer and Head of Cardiovascular

Twenty-five years in cardiac surgery, and a former notified body Team Leader and Senior Clinical Reviewer. Decides: whether the proposed evidence model will read as sufficient to the reviewer who eventually assesses it.

Nikhil Khadabadi

Chief Medical Officer, Orthopaedics and Spine

Former notified body reviewer, working on implant and long-term follow-up evidence. Decides: whether the endpoints and follow-up horizon are clinically meaningful for the therapeutic area, not just statistically convenient.

Susanne Höfer

Head of Clinical Operations, DACH Region

Clinical operations across German-speaking Europe. Decides: whether the country and site strategy can realistically deliver the population, the recruitment rate and the retention the design assumes.

Sébastien Meier Piantanida

Chief Data Officer

Data management and biostatistics across the programs. Decides: whether the dataset as specified can support the analysis, and what missing data will do to the conclusion before enrollment starts.

Pierre-Marie Boutanquoi

Head of Medical Writing

PMCF Evaluation Reports and clinical evaluation integration. Decides: whether the output can be written into the clinical evaluation as it stands, which is tested at design time rather than at reporting.

Dawn Heimer, PhD

Strategic Clinical Advisor, United States

Provides strategic input on U.S. medical-device clinical operations, regulatory considerations and real-world evidence programs. Advises on: U.S. clinical strategy, operations and real-world evidence.

Where it starts

Start With the CER Gap, Not With a Data Collection Budget

A PMCF study design workshop turns an unresolved clinical evaluation finding into a defined evidence question and a recommended model. It is the cheapest point in the program at which to change your mind.

What you bring

  • Device and classification
  • CE-marked intended purpose
  • Clinical Evaluation Report
  • PMCF plan
  • PMS and PSUR
  • Residual risks
  • Existing clinical evidence
  • Target population
  • Current evidence gaps
  • Expected endpoints
  • Current sites and users
  • Required follow-up

What you leave with

  • A defined PMCF evidence question
  • A recommended evidence model
  • Prospective, retrospective, registry or hybrid decision
  • Endpoint architecture
  • Sample-size assumptions
  • Follow-up model
  • Country and site strategy
  • Data architecture
  • Major delivery risks
  • Budget drivers
  • Reporting and CER integration plan
Request a PMCF Study Design Workshop

Questions

PMCF Studies and Evidence Programs, Answered

When does a medical device need a prospective PMCF study?

When the clinical evaluation identifies a device-specific question that existing data cannot answer, and the answer requires measurements that routine care does not produce. Typical triggers are residual risks needing structured follow-up, insufficient evidence for a claimed indication or population, endpoints that need standardized timing or imaging, and notified body questions asking for new device-specific clinical data.

When is a registry appropriate for PMCF?

When the question is longitudinal and depends on the whole treated population rather than a selected sample. Survivorship, revision, recurrence, long-term implant performance and rare events observed over years are registry questions. If the endpoint requires predefined visits, standardized measurement or imaging that routine care does not generate, a registry is the wrong instrument however large it becomes.

Can retrospective clinical data be used for PMCF?

Yes, when the records can carry the question. That is established by assessment rather than assumption: device identification, consecutive case identification, endpoint availability, follow-up completeness, missing data, safety-event documentation, site record quality and selection bias. A pilot chart review and data-availability assessment answers how many existing records are actually analyzable against the endpoint before any large extraction is committed.

How many patients should a PMCF study include?

There is no default number, and any figure offered before the endpoint is defined is a guess. Sample size follows from the primary endpoint, the expected event rate or effect size, the variability in the population, the required follow-up duration and the acceptable precision. Those assumptions are derived in the design workshop and stated explicitly, so the number can be defended rather than justified after the fact.

Can an existing national or specialty registry be used for PMCF?

Sometimes, and it is worth checking first. The questions are whether the registry identifies the device at the level of granularity needed, whether it holds the endpoint, whether follow-up runs long enough, and whether data access and publication terms allow the intended regulatory use. Where those hold, using or linking to an existing registry is usually faster and more defensible than building a parallel one.

Can Eclevar run a multicountry PMCF study in Europe?

Yes. Eclevar has delivered multicountry prospective PMCF across European sites, and prepares and files the regulatory and ethics submissions in house rather than subcontracting them. Country strategy, site strategy, monitoring, data management and biostatistics are held by one accountable team.

What information is needed to budget a PMCF study?

Device and classification, the CE-marked intended purpose, the clinical evaluation and PMCF plan, the residual risks, the existing clinical evidence, the target population, the expected endpoints, the required follow-up duration and the current sites or users. From those, the cost drivers become visible: number of countries, number of sites, visit schedule, follow-up horizon, imaging, and the monitoring intensity the risk profile justifies.

How does a PMCF study feed into the CER, PSUR and SSCP?

The analysis output becomes the PMCF Evaluation Report, which is the document the clinical evaluation consumes. From there the findings update the Clinical Evaluation Report, inform the PSUR, feed the SSCP where one applies, and return to the risk management file. Designing the endpoints with the writers of those documents is what keeps the loop closed.

Publications and external evidence

Published Work and Industry Collaboration

Published material and externally judged results, rather than claims about ourselves.

Published whitepaper, BSI and Eclevar

EU MDR post-market clinical follow-up: gaining real-world evidence

A practical reading of the clinical evidence expectations under EU MDR, written with the notified body BSI and relevant to any high-risk device program.

Read it in the whitepaper library

Whitepaper in preparation

Clinical evidence under EU MDR, read from the reviewer's side

What notified bodies expect from a clinical evaluation and a PMCF strategy for Class III and implantable devices, co-written with a former TUV SUD senior reviewer.

Coming soon

Platinum CRO Award 2026 badge issued to Eclevar MedTech, an initiative co-funded by the European Commission.

Industry collaboration, xShare and EUCROF

EHDS and Clinical Research Open Call

The Platinum Award, top tier of the xShare and European CRO Federation Open Call on EHDS and clinical research, was presented to Eclevar MedTech and its Milo Health platform at the EUCROF 2026 conference in Amsterdam on 2 February 2026. The Open Call is co-funded by the European Union under Horizon Europe. It is recognition of the submitted use case, judged externally, and not an assessment of the services on this page.

Read the independent report from partner CVBF

Next step

Define the Evidence Model Before Committing the PMCF Budget

Send the device, the clinical evaluation gap, the existing evidence, the target population, the current PMCF plan and the follow-up you expect to need. We will come back with the evidence question as we read it and the model we would recommend, before anything is priced.

Discuss Your PMCF Study Request a PMCF Study Design Workshop
  • Device and classification
  • The clinical evaluation gap
  • Existing clinical evidence
  • Target population
  • Current PMCF plan
  • Intended follow-up

Related capabilities

Before the market

Clinical investigations

Pre-market clinical investigation design and delivery, where the evidence question comes before CE marking rather than after it.

The parent capability

Post-market clinical follow-up

The full PMCF capability, of which studies and registries are the execution end. Start there if the PMCF plan itself is the open question.

Where the data lives

Clinical data management

EDC, eCOA and biostatistics on Milo Studio, including the longitudinal structures a registry depends on. See statistical analysis plans and endpoint analysis.

Evidence in practice

Client success stories

The wider set of client programs, including the case studies summarized on this page and several outside post-market evidence.

Talk it through

Contact

Bring the clinical evaluation gap and the current PMCF plan. The first conversation is about the question, not the quote.

Therapeutic areas

Cardiovascular and structural heart Orthopaedics, spine and surgical robotics Advanced wound care Neurology and neuromodulation

Reforming Clinical Evaluation of Medical Devices in Europe