Under the Medical Device Regulation, a clinical investigation is how you generate clinical data on a device when literature and equivalence are not enough. This guide covers the route from the decision to investigate through application, conduct, safety reporting and closure, and what your data system has to support. Not legal advice.
Free sandbox · No credit card · 21 CFR Part 11 aligned
Decide an investigation is needed
Clinical evaluation shows a gap
Prepare Annex XV documentation
Application form, IB, plan, supporting documents
Apply to the member state(s)
Via the electronic system, with ethics review per national law
Conduct the investigation
Consent, monitoring, safety reporting
Report changes and events
Substantial modifications and serious events
Close out and report
End notification and report with summary
The short answer
Scope
The MDR requires a manufacturer to plan and carry out clinical evaluation for every device. That evaluation uses clinical data, which can come from the literature, from equivalent devices where the stringent conditions are met, from post-market surveillance, and from clinical investigations of the device itself. For implantable devices and class III devices, the regulation expects a clinical investigation to be carried out unless a listed exemption applies, for example for a modification of a device already on the market under conditions the text sets out. Your notified body and regulatory team decide which applies. A thin literature review is not a substitute, and a gap identified in clinical evaluation is the usual trigger for an investigation.
Software that is a medical device is in scope too. If your product is software as a medical device with a clinical claim, the same evaluation logic applies, and so does the question of whether you need prospective data. For AI-based devices the validation design has its own questions, covered in our guide to EDC for AI medical device clinical validation.
Which route
A simplified map. Confirm the route for your case with your regulatory adviser and the competent authority.
| Type of study | Typical purpose | MDR treatment |
|---|---|---|
| Investigation to demonstrate conformity (pre-market) | Generate clinical data for conformity assessment of a device without a CE mark, or outside its intended purpose | Full application under Article 70, assessment by the member state(s), ethics review, conduct under Article 72 and Annex XV |
| Investigation of a CE-marked device within its intended purpose with additional invasive or burdensome procedures | Post-market clinical follow-up (PMCF) with extra procedures | Notification route under Article 74, with Annex XV documentation, before the study starts; check the notice period in the text |
| Other investigations (not for conformity purposes) | Academic questions, comparisons, outcome research | Article 82 leaves these to national law, with some MDR principles still applying; check the national authority |
PMCF studies without additional procedures are usually run under the post-market framework rather than as a clinical investigation.
Application
Article 70 governs the application for investigations to demonstrate conformity. The sponsor submits to the member state or states where the investigation will take place, using the electronic system the MDR provides for (EUDAMED). The EUDAMED modules have been brought into use in phases, so check the Commission notices for the current position on which modules are mandatory and what applies in your member state in the meantime. The member state validates the application, assesses it, and may ask for more information. Multi-state investigations can use a coordinated assessment procedure under the regulation, covered in Article 78. The exact time limits are in the text and in national guidance, and are deliberately not copied here.
The documentation is set out in Annex XV Chapter II. In general terms it includes the application form, the investigator brochure, the clinical investigation plan, and other documents, among them evidence of insurance, informed consent materials, information on the investigators and sites, and the proof that the device meets the general safety and performance requirements apart from the aspects under investigation. The regulation also requires that the investigation be ethically reviewed, with the process set by national law. Most countries have a separate ethics committee route that runs in parallel with the authority assessment.
Annex XV expects the plan to define the objectives, design, endpoints, the statistical approach, the data management and quality assurance arrangements, and how deviations are handled. That is exactly what you need to build the case report forms and visit schedule from. Build from the approved plan, and keep the study configuration in step with every amendment. See protocol amendments without change orders for the practical method.
The MDR sets the legal requirements, while ISO 14155 on good clinical practice for device investigations is the standard most sponsors, notified bodies and ethics committees expect you to follow. Its harmonisation status and the current edition change, so check both before you cite it. Our page on ISO 14155 clinical investigation software maps the standard to software controls.
Conduct
The MDR has specific provisions on informed consent, and on investigations involving incapacitated subjects, minors, pregnant or breastfeeding women and emergency situations. Consent is given freely, after an interview with a member of the investigating team and written information that is comprehensive, concise, clear and understandable to the participant, and the participant can withdraw at any time without detriment. Insurance or an equivalent arrangement for damage compensation is required under the regulation, with the specifics set by national law. Read the relevant articles in full for any vulnerable group before you design the consent process, and see eConsent software for clinical trials for what to check in the system.
Safety reporting is its own workstream. The sponsor records and reports serious adverse events, device deficiencies that could have led to a serious event, and new findings in relation to them, to the member states where the investigation takes place, within the time limits the text sets. MDCG guidance on safety reporting in clinical investigations explains the definitions and the report format. Substantial modifications to the investigation, for example changes to the plan, the device or the population that affect safety, rights or data reliability, need to be notified and, where the regulation requires, assessed before implementation. If the investigation is halted temporarily or ended early, notification duties follow. Our adverse event reporting software guide covers the data capture side.
Upload the plan, let the AI study builder draft the visits and forms, review every item, and test edit checks, skip logic and the audit trail in the free sandbox.
End to end
A working sequence for a pre-market conformity investigation. Timelines depend on the member state and the regulation.
Document the gap in clinical evaluation, agree the route with your notified body and regulatory adviser, and decide which member states and sites are involved.
Draft the clinical investigation plan, the investigator brochure and the risk management links, with a clear statistical and data management section.
Application form, plan, brochure, consent and participant information, insurance evidence, site and investigator suitability, and the other listed documents.
Submit through the electronic system, track the validation and assessment, and answer requests for information within the stated time.
Competent authority decision and a favourable ethics opinion per national law, site agreements, trained staff and a tested, released study build.
Consent, data capture, monitoring, safety reports and notifications of substantial modifications. Keep the audit trail and the document set current.
Notify the end of the investigation, lock the data, and prepare the report with a summary that the intended user can understand. The report and summary timing is set in Article 77 and Annex XV, so check the text for the dates that apply.
Data system
| Obligation (in general terms) | What to be able to show | In Capture |
|---|---|---|
| Data reflect the plan | Forms and visit schedule match the approved plan version | Draft to approved form lifecycle; only approved forms are in the casebook |
| Traceable changes to data | Who changed what, when, why | Field-level audit trail with timestamp, user, old value, new value and reason, exportable |
| Controlled access | Roles, least privilege, protection of identity | Role-based access, PII segregation, row-level security |
| Consent documented | Version, date and signatory for each participant | eConsent with email OTP, countersignature and PDF export with timestamps |
| Serious events recorded and reportable | A complete list with dates for notification | Adverse event forms and safety workflows in one system |
| Data available for the report | Locked, reproducible dataset with dictionary | CSV and Excel exports with data dictionary; SDTM as SAS XPT with Define-XML |
Capture supports your process. The sponsor remains responsible for the application, the plan and the regulatory submissions.
In Capture
A visit-by-assessment schedule built from the plan shows where device-specific measurements, follow-ups and safety assessments sit. Because the form lifecycle is draft to approved, the version that goes live is the one you approved, and the blank eCRF PDF export gives you a printable copy for ethics submissions. Capture is not a regulatory submission tool and does not connect to EUDAMED.
| Assessment | Screening | Implant | Day 7 | Month 1 | Month 6 | Month 12 |
|---|---|---|---|---|---|---|
| Informed consent | ||||||
| Device procedure record | ||||||
| Device deficiency check | ||||||
| Adverse event review | ||||||
| Primary performance measure | ||||||
| Patient-reported outcome |
Demo schedule only; real visits come from your plan.
A systematic investigation involving one or more human subjects, undertaken to assess the safety or performance of a device. The MDR regulates those carried out for conformity assessment purposes in Articles 62 to 82 and Annex XV, and leaves other investigations largely to national law.
Chapter VI, commonly cited as Articles 62 to 82, with Annex XV setting the general requirements, the documentation and the sponsor obligations. Check the consolidated text, since amendments have changed some provisions.
No. Clinical evaluation can use literature, equivalence and existing clinical data. Implantable and class III devices are expected to have an investigation unless a listed exemption applies, and your notified body will tell you what your device needs.
Devices and medicines have separate regulations, separate applications and separate portals. The MDR route uses its own application under Article 70 and the electronic system provided for in the MDR, not CTIS. Drug-device combinations need advice on which framework applies.
In general terms Article 70 is the application for investigations to demonstrate conformity, while Article 74 covers investigations of CE-marked devices within the intended purpose that add invasive or burdensome procedures, via a notification with Annex XV documentation. Confirm the route and the time limits in the text.
The MDR itself is the legal requirement. ISO 14155 is the recognised good clinical practice standard for device investigations and is widely expected by ethics committees and notified bodies. Check the current edition and its harmonisation status.
Capture is a data capture platform with EDC, eCRF, ePRO, eConsent and safety reporting workflows, a field-level audit trail and 21 CFR Part 11-aligned controls. It supports the data side of an investigation. It does not submit applications or reports to authorities.
No. It is orientation for sponsors and regulatory teams. Check the current Regulation (EU) 2017/745, MDCG guidance and your national competent authority before you act.
Keep exploring
How to run a medical device clinical study
End-to-end process, EU and US.
ISO 14155 software
The standard mapped to controls.
EDC for medical device trials
Data capture for device studies.
Early feasibility and pivotal device studies
Staging device evidence.
Software as a medical device
SaMD evidence needs.
Adverse event reporting
Safety data capture.
EDC
The Capture EDC.
Free sandbox with every feature. No credit card, and you pay only when you go live with real participants.