Home / Clinical Data Management

Clinical data management for medical devices

Clinical Data Management Services for Medical Device Clinical Trials

Eclevar provides specialized clinical data management for medical device investigations, from protocol review, eCRF design and database build through data cleaning, reconciliation and database lock.

A traceable clinical database, designed to support inspection readiness and downstream statistical analysis, without a fragmented chain between clinical operations, data management and biometry.

Clinical data management run inside a medical device CRO, not bolted onto a pharmaceutical one

  • Medical device specialization
  • Dedicated data management and biometry function
  • EU MDR clinical investigations and FDA IDE studies
  • ISO 14155:2026, ICH GCP and 21 CFR Part 11 frameworks applied to data processes
  • Clinical operations, data management and quality in one organization
  • Your EDC environment or ours
30+ yearsClinical biometry leadership
10+ yearsSenior clinical data management expertise
160 · 14 · 5Participants, clinical sites and countries in one medical device data program
EU + USMedical device clinical programs

European delivery, with clinical operations covering the DACH region.

Where we usually come in

When sponsors bring Eclevar into a study

Most conversations start from one of six situations, each with a different scope and a different entry point.

Your protocol is ready, but the database is not

We translate protocol-defined endpoints, visit requirements and operational data needs into eCRFs and database structures designed for reliable downstream analysis.

Your current CRO has a data cleaning backlog

We can add independent data management capacity without replacing your clinical operations team. Site relationships stay where they are; the queries, coding and reconciliation move.

Database lock is slipping

We review open and aging queries, coding status, external data reconciliation, critical data completeness and lock documentation, then work the remaining list in priority order.

Your eCRFs are generating too many queries

High query volume can reflect avoidable design issues, not only site performance. We review data capture design, edit check logic and form structure to remove unnecessary site burden.

You need an independent data management partner

Some sponsors prefer data management held separately from the party running the sites. Eclevar can act as a full service partner, an embedded resource or a specialized remediation team.

You are preparing a regulatory submission

Traceability from the source record to the analysis dataset is built during the study rather than reconstructed afterward. We work backward from the evidence the submission has to carry.

The lifecycle

Nine stages, one accountable chain

Eclevar can own the full sequence or take a defined part of it. What does not work is a chain with a gap in the middle, where nobody owns the point at which one stage hands over to the next.

The Eclevar clinical data management lifecycle, in four phases and nine stages, from protocol review to an analysis-ready dataset.
Phase 1 · Design
  • Protocol reviewEndpoints, visit structure, critical data and external sources identified before any build.
  • Data management strategyScope, systems, roles, review model and the route to lock agreed in writing.
  • eCRF designDevice, procedure and endpoint data captured in a form sites can complete accurately.
Phase 2 · Build
  • EDC database buildVisit schedule, forms, edit checks, permissions, workflows and audit trail configured.
  • Database validation and UATSpecifications, test scripts, user acceptance testing, issue resolution and controlled release.
Phase 3 · Conduct
  • Data cleaning and queriesContinuous review of critical data, with query volume and aging managed as they arise.
  • Coding and reconciliationMedical coding, safety reconciliation and alignment with imaging, laboratory and device datasets.
Phase 4 · Close
  • Database lockA documented readiness review against defined criteria, then a controlled lock.
  • Analysis-ready datasetStructured, documented data handed to the statistician in the agreed format.

Scope is agreed stage by stage. A sponsor may hand over the full sequence, or only phases 3 and 4 on a study that is already running.

Study setup

From protocol to a validated database

Eclevar can take responsibility for the complete clinical data management lifecycle, or integrate with your existing CRO, EDC vendor, statistics group or internal clinical organization. Data issues are easier and less disruptive to address during protocol and eCRF design than immediately before database lock.

Protocol review and data management strategy

Data management should begin before the database build. We read the protocol as a data specification and agree what is collected, at which visit, and what happens to it afterward.

  • Endpoint and data requirement review against the visit structure
  • Critical data and critical study processes identified
  • External data sources, transfer routes and reconciliation needs
  • Coding requirements and expected statistical outputs
  • A database lock strategy agreed at the start

Data management plan

The data management plan is the operational control document for the study, where decisions that would otherwise be made informally are written down and agreed with the sponsor.

  • Roles, responsibilities and escalation routes
  • Data review model, frequency and depth
  • Query management conventions and closure rules
  • Medical coding and reconciliation procedures
  • Quality control checks and the database lock procedure

Medical device eCRF design

Device studies collect information a pharmaceutical eCRF is not built to hold. Depending on the device and study design, the form may need to carry the procedure, the device and the technical performance of both.

  • Implantation and procedural information, including concomitant procedures
  • Device characteristics, identifiers, and software or firmware versions
  • Device deficiencies, malfunctions, revisions and explants
  • Technical performance endpoints, imaging assessments and adjudicated events
  • Device accountability where the design requires it

eCRF design influences site burden, query volume, missing data, and whether an endpoint stays interpretable once the data are in.

EDC database build

We build the study in the environment that suits the program: a sponsor system, a third party platform, or the Milo environment used on Eclevar studies. What we sell is data management expertise, not a software license.

  • Visit schedule, forms and dynamic logic
  • Edit checks written from the data review rules, not added afterward
  • Role-based access, user permissions and automated workflows
  • Audit trail configuration and system integrations where applicable

Database validation and user acceptance testing

A database goes live once it has been shown to behave as specified, and once that demonstration is documented. Retrofitting the documentation later is the most common gap we find on rescue engagements.

  • Database specifications maintained as a controlled document
  • Test scripts covering forms, logic, derivations and edit checks
  • User acceptance testing with the sponsor, and documented issue resolution
  • Controlled release to sites, with change control on later versions

Data flow and external source mapping

Before the first participant is entered, we map the datasets that will exist outside the primary EDC and agree who owns each transfer, in which format, at which frequency.

  • Data transfer specifications agreed with each external provider
  • Identifier conventions that allow datasets to be matched reliably
  • Reconciliation rules defined before data start arriving
  • Responsibility for discrepancy resolution written into the plan
Study conduct

Keeping the database clean while the study runs

The alternative is a backlog discovered at the end, when the sites have moved on and every question costs more to answer than it would have at the time.

Clinical data cleaning

Data review is continuous and prioritized by what the study has to demonstrate. Critical and endpoint data are reviewed on a defined cycle; lower-risk fields are handled through automated checks.

  • Missing data, incomplete forms and overdue visits
  • Internal inconsistencies, impossible values and outliers
  • Temporal inconsistencies across visits and procedures
  • Device and procedure related inconsistencies
  • Protocol deviations, in coordination with clinical operations

Query management

Queries are a cost borne by the sites, so the objective is the smallest number that resolves the real issues. We track aging as closely as volume.

  • Automated checks tuned to reduce avoidable queries
  • Manual review where logic cannot replace clinical judgment
  • Aging monitored per site, with defined escalation
  • Traceability from issue raised to query closed

Medical coding

Coding is performed against the dictionary versions agreed in the data management plan, with a documented review route for terms needing clinical input.

  • Adverse events and medical history coded in MedDRA
  • Concomitant medications coded in WHO Drug where the study requires it
  • Version control and documented upversioning decisions
  • Coding review with the medical monitor where the protocol provides for one

External data reconciliation

Medical device studies may generate substantial evidence outside the primary EDC, including imaging, core-lab, device-generated, safety and eCOA data. The clinical database should remain consistent with the external evidence used to evaluate device safety and performance.

  • Safety data and, where applicable, adjudicated event data
  • Imaging, central reading and core laboratory datasets
  • External laboratory data and randomization system records
  • Device-generated data, ePRO and eCOA datasets
  • Wearable and connected device data where the design includes them

Risk-based data management

Not every field deserves the same attention. Review effort is concentrated where an error would change the conclusion of the study, and the reasoning is documented so it can be explained later.

  • Critical to quality factors identified with the sponsor
  • Primary endpoint and safety variables reviewed at the highest intensity
  • Device deficiency reporting treated as critical data
  • Data review aligned with the monitoring plan rather than duplicating it

Data review with clinical operations

Our data managers work alongside the people who run the sites, so a pattern in the data can be assessed as a training issue, a form design issue or a clinical signal without waiting for the next steering meeting.

  • Joint data review with monitoring and project management
  • Site-level trends fed back into training and monitoring focus
  • Recurrent form issues resolved at source through change control
Close-out

Database lock and analysis-ready data

Lock is the point at which the study stops being an operation and becomes evidence. It should be a controlled milestone, not an emergency clean-up exercise.

Before we recommend a lock, we work through a readiness review against criteria agreed in the data management plan. As applicable to the study, that covers:

  • Critical queries resolved and remaining queries documented
  • Required forms complete for all participants in scope
  • Medical coding finalized and reviewed
  • External datasets reconciled against the clinical database
  • Planned data review completed and recorded
  • Protocol deviations addressed with clinical operations
  • Quality control checks executed and documented
  • Lock documentation and approvals complete

A lock that is a surprise to anyone in the study team was not a controlled lock.

The handover is a structured, documented dataset in the format agreed with the statistician, with the definitions and derivations that go with it. Controlled and traceable documentation is built during the study and designed to support inspection readiness, rather than reconstructed at the end. If an unlock becomes necessary, it is handled under change control, with the reason, the change and the re-lock documented.

What you receive

Clinical data management deliverables

Scope varies by study. This is the standard deliverable set named in an Eclevar data management scope and budget.

Study stageTypical Eclevar deliverable
PlanningProtocol and data strategy review
SetupData management plan
Data captureeCRF specifications and site completion guidelines
DatabaseConfigured EDC database
ValidationUAT and validation documentation
ConductData review and query management
CodingMedical coding
IntegrationExternal data reconciliation
QualityData quality reviews
Close-outDatabase lock documentation
HandoverClean analysis-ready clinical data
Device-specific data

Clinical data management built for medical devices

A device study is not a drug study with a different intervention. The unit of analysis often includes an implant, an operator and a procedure, and a substantial share of the evidence may arrive from outside the EDC.

The medical device clinical data ecosystem: four categories of source data converge into one clinical database, which produces the analysis-ready dataset.
Site and participant dataDemographics, visits, assessments, outcomes and participant-reported measures.
Device and procedure dataIdentifiers, characteristics, implantation detail, deficiencies, revisions and explants.
Imaging and core laboratoryCentral reading output, adjudication decisions and laboratory datasets.
Safety and connected sourcesSafety records, randomization systems, ePRO, eCOA and device-generated data.

One clinical database

Each source that applies is mapped, transferred against a specification and reconciled on a defined cycle, so the clinical database stays consistent with the external evidence used to evaluate safety and performance.

Analysis-ready dataset

Structured for the statistical analysis the protocol requires, with definitions, derivations and a traceable route back to the source record.

Which sources apply depends on the device and the study design. A single-center early feasibility study and a multicenter pivotal investigation carry very different reconciliation loads.

Device and procedural data

Implant and procedure characteristics, operator and technique variables, and the device identifiers that let a later question about a specific unit be answered.

Device deficiencies and malfunctions

Structured collection, classification and reconciliation of deficiencies, malfunctions and the events they lead to, aligned with the reporting obligations that apply to the study.

Imaging and core laboratory data

Integration and reconciliation of independent assessments, including adjudicated endpoints, where the clinical conclusion depends on a reading performed away from the site.

Longitudinal post-market follow-up

PMCF and long-term observational designs can put years between the procedure and the last data point, which changes how identifiers, visit windows and missing data are handled.

Digital and connected devices

Higher-frequency, machine-generated data needs a defined transfer, storage and reduction rule before it reaches the clinical database, or it becomes unusable volume.

Early feasibility through pivotal

Early feasibility databases have to absorb evolving procedural experience. Pivotal databases are built tightly around prespecified endpoints. The design decisions are not interchangeable.

Therapeutic and program experience

  • Cardiovascular and structural heart
  • Vascular access and grafts
  • Orthopedics and spine
  • Advanced wound care
  • Regenerative medicine
  • Neuromodulation
  • Neurological diagnostics
  • Continence and urology
  • Clinical nutrition
  • Dental
  • Early feasibility and pivotal investigations
  • FDA IDE studies
  • PMCF and long-term observational follow-up
Remediation

Clinical data management rescue

Already working with another CRO or EDC provider? You may not need to replace the entire study team.

Many data management problems are contained. They sit in the database, in the query backlog or in the documentation, and a separate team can work on them without disturbing site relationships that took a year to build. Eclevar can be engaged for a defined remediation scope with a defined end point.

  • eCRF remediation and database design review
  • Data cleaning backlog and aging queries
  • Medical coding backlog
  • External data reconciliation gaps
  • Database validation and documentation gaps
  • Pre-lock data review and database lock preparation
  • Vendor transition and migration between EDC environments
  • Legacy data review before analysis
  • Independent review of a database before a sponsor decision
  • Additional capacity for a defined period
Outsourcing models

Three ways to buy clinical data management

The right model depends on what you already have in-house, not on what is easiest to sell.

Full-service clinical data management

We want one partner to own CDM

Eclevar manages the function from protocol review through database lock and dataset handover, with a named data manager, a data management plan and one accountable route for every data question.

Embedded and FSP clinical data management

We need additional capacity inside our existing structure

Eclevar data managers work inside your organization, your systems and your procedures. You keep the platform, the standards and the oversight; we supply the people and the device experience.

Rescue and specialized support

We need a specific data problem fixed without replacing the entire CRO

A scoped engagement against a defined problem: database remediation, a cleaning or coding backlog, reconciliation, validation documentation or lock preparation, with a clear definition of done.

Selected program

Data management on a multicountry PMCF program

RegenLab Regenerative medicine · Chronic wound care · Europe Ongoing program

160 participants, 14 sites, 5 European countries, one clinical database

Challenge

A randomized post-market clinical follow-up program in diabetic foot ulcers and venous leg ulcers, across France, the United Kingdom, Germany, Italy and Spain. Two wound populations with different assessment schedules had to produce data reviewable on a common structure.

Eclevar role

Eclevar designed and is managing the program, including the clinical data management function across all participating countries.

Delivery

eCRF design across both cohorts, database build in the Milo environment, data review, query management and reconciliation across five countries, on one data structure and one set of review rules.

Program evidence

Evidence intended to support EU MDR clinical evaluation and post-market surveillance documentation for the sponsor's chronic wound care portfolio.

Program described with the sponsor's permission. The program is ongoing. Figures describe the program design as planned, not completed follow-up. No clinical result, performance conclusion, regulatory outcome or authority acceptance is claimed or implied.

The people who do the work

Meet the clinical data and quality team

Data management is a named-person service. These are the people a sponsor deals with, not a resourcing pool.

Portrait of Sebastien Meier Piantanida, Chief Data Officer and Biometry Lead at Eclevar MedTech

Sebastien Meier Piantanida

Chief Data Officer · Biometry Lead

30+ years in clinical biometry. Sets data strategy and clinical data architecture across Eclevar programs, and owns the interface between data management and biostatistics, so databases are designed for the analysis they will have to support and not only for the data they will collect. Oversight of complex multicountry medical device data environments, across EDC and eCOA.

Portrait of Mathilde Renier, Senior Clinical Data Manager at Eclevar MedTech

Mathilde Renier

Senior Clinical Data Manager

10+ years in clinical data management. The hands-on lead on sponsor studies: eCRF design, EDC database build and configuration, validation, data cleaning, query management, consistency checks and database lock readiness through to the final analysis-ready dataset. If you engage Eclevar for data management, this is the level at which your study is run day to day.

Portrait of Jimmy Andrew Hayek, Head of Quality and Compliance at Eclevar MedTech

Jimmy Andrew Hayek

Head of Quality & Compliance

The quality and governance interface supporting data management: QA oversight, controlled processes, documentation and data governance, procedural compliance and the quality system interface, with inspection readiness designed into the workflow. He is not the data manager on your study; his role is why data management sits inside a documented quality framework rather than individual habit.

One accountable chain from clinical data collection through quality oversight to analysis-ready datasets.

See the full Eclevar leadership team.

The interface

From clean data to statistical analysis

Every link in this chain is a decision, and a decision taken at one end constrains the other.

  • Protocol endpoint
  • eCRF variable
  • Database structure
  • Data review rule
  • Analysis dataset
  • Statistical output

Eclevar's clinical data management and biostatistics teams work within the same broader data and biometry capability, helping reduce handoff gaps between database design, database lock and final analysis.

Standards

The frameworks these processes are written against

We are precise about this, because imprecision here is expensive later. Data management processes are designed against the requirements below. Meeting the requirements of one framework does not establish anything about another.

  • ISO 14155:2026 for clinical investigations of medical devices in human subjects
  • EU MDR 2017/745 clinical investigation and post-market clinical follow-up requirements
  • Data management processes designed to support studies conducted under applicable FDA requirements
  • 21 CFR Part 11 requirements for electronic records and electronic signatures, where applicable
  • ICH GCP principles where the study design makes them applicable
  • ALCOA+ data integrity principles applied to data review and documentation
  • Risk-based quality management applied to data review intensity
  • Applicable data protection requirements, including the GDPR in the European Union

Eclevar does not represent that any study, dataset or system is certified or accepted by a Notified Body or a competent authority, and does not claim a guaranteed inspection outcome.

The case for Eclevar

Why device sponsors choose Eclevar for clinical data management

Medical device specialization

Data management designed around device, procedural and operator data, not adapted from a pharmaceutical template each time a device study arrives.

Senior data leadership

Data strategy set by the head of the biometry function, not delegated to whoever has capacity in the resourcing model that quarter.

Clinical operations in the same organization

Our data managers understand how the data were generated because they work alongside the monitors and project managers who generated them.

Quality oversight built in

Data management operates under documented procedures with independent quality review, rather than the personal method of the data manager assigned.

Flexible outsourcing

Full service, embedded resource or a scoped rescue. Sponsors move between models as a program matures, and the contract should allow it.

An analysis-ready mindset

The question we ask at eCRF design is what the statistician will need at the end. It is the cheapest question in the study, and the one most often skipped.

Questions

Clinical data management, answered

What does a clinical data management CRO do?

It is responsible for the clinical data of a study from the moment the protocol defines what will be collected to the moment a locked dataset reaches the statistician: data management strategy, eCRF design, the EDC database and its validation, data review and cleaning, query management, medical coding, reconciliation with external datasets, and database lock. In a device study it also covers the device, procedural and technical performance data a general clinical database is not designed to hold.

When should clinical data management start in a medical device study?

Before the protocol is final, where possible. A data manager reading a protocol draft will ask which variable each endpoint resolves to, which data arrive from outside the EDC, and how missing data will be handled. Those questions change the protocol, and they are easier to answer then than immediately before lock.

Can Eclevar manage data management without running the entire clinical trial?

Yes. Data management is regularly contracted as a standalone function while another party runs the sites, or while the sponsor runs them internally. What we agree at the start is the interface: who owns the site relationship, who raises and chases queries, who owns the protocol deviation record, and who signs the lock.

Can Eclevar take over data management from another CRO?

Yes, and it is a common engagement. A transition starts with an independent review of the database, the validation documentation, the open query and coding position, and the state of external data reconciliation, so both sides know what is being inherited. We then agree a scope with a defined end point rather than an open-ended remediation.

What is included in clinical data management outsourcing?

Typically the data management plan, eCRF specifications and completion guidelines, the configured EDC database, validation and user acceptance testing documentation, data transfer specifications, ongoing data review and query management, medical coding, reconciliation of external datasets, quality review records, the lock readiness review and the final analysis-ready dataset. The exact list is written into the scope.

How is clinical data management different for medical device trials?

The data model has to carry the device and the procedure, not only the participant: identifiers, characteristics, implantation detail, deficiencies, malfunctions, revisions and explants, and often technical performance measured by the device itself. A larger share of the evidence may sit outside the EDC, in imaging, core laboratory and adjudication datasets, so reconciliation carries more weight. Post-market follow-up periods are also longer, which changes how identifiers, visit windows and missing data are handled.

Can Eclevar support both EU and US medical device clinical studies?

Yes. We manage data for clinical investigations conducted under EU MDR and for medical device studies conducted under applicable FDA requirements, including IDE studies, as well as for programs running in both regions. The data management plan records which requirements apply to which part of the program, because the answer is often not the same across a single study.

Next step

Need a clinical data management partner for your medical device study?

Whether you need full-service clinical data management, additional CDM capacity or targeted support to bring a delayed database to lock, Eclevar can provide a specialized medical device data team aligned with your clinical and regulatory program.

Reforming Clinical Evaluation of Medical Devices in Europe