Clinical investigation is a systematic, planned study of a medical device in human subjects to assess its safety and clinical performance for regulatory purposes. Governed by ISO 14155, EU MDR 2017/745, and FDA 21 CFR Part 812, it generates the clinical evidence used to support market approval and ongoing device claims.
What is a clinical investigation?
A clinical investigation is a structured study that tests a medical device on human participants to confirm it works as intended and is safe under real use. In the device world, the terms “clinical investigation,” “clinical study,” and “clinical trial” are used interchangeably. It sits in the clinical evaluation stage of the device lifecycle, generating primary clinical data when existing literature or an equivalence argument is not enough to prove conformity.
The reference standard is ISO 14155, updated to its fourth edition (ISO 14155:2026) in March 2026, which sets good clinical practice (GCP) for designing, running, recording, and reporting these studies. It applies to most devices and software as a medical device (SaMD), but not to in vitro diagnostic (IVD) products, which follow ISO 20916.
Why clinical investigation matters in medical device development
A clinical investigation produces the evidence regulators use to decide whether a device reaches the market. Get the design or conduct wrong, and the consequences are concrete: rejected submissions, repeat studies, delayed launches, and exposure during audits.
For higher-risk devices, the study is often mandatory. Under EU MDR 2017/745, Class III and implantable devices generally require clinical investigation data unless equivalence to an existing device is firmly established (Article 61). In the United States, significant-risk devices need an approved Investigational Device Exemption before a study can begin. Weak clinical evidence is one of the most common reasons high-risk submissions stall, so the quality of this work shapes both timeline and budget.
The study also protects participants. Ethical oversight, informed consent, and risk management are built into the process, and they are the parts regulators and ethics committees scrutinize first.
How a clinical investigation works
A clinical investigation follows a regulated sequence. The exact steps vary by region, but the core flow is consistent:
- Plan the study. The sponsor writes a clinical investigation plan (CIP) that states the research question, endpoints, sample size, and statistical method. ISO 14155:2026 introduces an estimand framework, which forces teams to define precisely what effect they are measuring, in whom, and under what conditions.
- Manage risk. Device risks are assessed against ISO 14971, and the residual risk must be justified against expected clinical benefit before any participant is enrolled.
- Get approval. In the EU, the sponsor applies to the competent authority of each member state and secures a positive ethics committee opinion, with documentation per Annex XV. In the US, significant-risk devices need an FDA-approved IDE under 21 CFR Part 812, plus Institutional Review Board sign-off.
- Obtain informed consent. Every participant must understand the study and agree voluntarily, consistent with the Declaration of Helsinki.
- Conduct and monitor. Sites collect data under GCP, with monitoring, source-data verification, and adverse event reporting throughout.
- Report. Results go into a clinical investigation report that feeds the clinical evaluation report (CER) and the technical documentation.
Standards and regulations interlock here. ISO 14155 sets the practice, EU MDR Articles 62 to 82 set the EU legal path, FDA 21 CFR Part 812 sets the US path, and ISO 13485 governs the quality system around it.
Common challenges and best practices
The frequent failure is a vague research question. If endpoints and the statistical plan are loose, the data rarely supports the claims a team wants to make. The estimand approach in ISO 14155:2026 exists to close that gap, so adopt it early rather than retrofitting it.
Underpowered studies are another trap. Sample sizes set for convenience instead of statistical reality produce inconclusive results and force a second study. Build the risk management file and the CIP together, so that identified risks map directly to what the investigation measures.
Documentation gaps surface during audits. Traceability from risk to endpoint to result is what reviewers look for, and reconstructing it after the fact is slow. Strong teams also engage notified bodies or the FDA early through pre-submission meetings, which surface objections while they are still cheap to fix.
How SJML helps with clinical investigation
SJML supports the regulatory and clinical-evidence side of device development as part of its Compliance-as-a-Service offering. The QARA team helps determine whether a clinical investigation is needed, plans clinical evaluation activities (CEP and CER), and manages clinical data and literature review. SJML also plans post-market clinical follow-up (PMCF), aligns studies with ISO 14155 and EU MDR expectations, and maintains the ISO 14971 risk files that underpin study justification. This connects clinical strategy to design, manufacturing, and submission work under one roof.
Frequently Asked Questions
Yes. In the medical device field, clinical investigation, clinical study, and clinical trial mean the same thing: a planned study of a device in human participants to assess safety and performance. The device sector tends to use “clinical investigation,” while “clinical trial” is more common with drugs. ISO 14155 is the governing standard for device studies.
It is usually required for higher-risk devices. Under EU MDR, Class III and implantable devices generally need one unless equivalence is firmly demonstrated. In the US, significant-risk devices require an approved IDE before the study starts. Lower-risk devices can often rely on existing clinical data, literature, or equivalence instead.
ISO 14155 is the primary international standard, now in its fourth edition, ISO 14155:2026. It defines good clinical practice for designing, conducting, recording, and reporting device studies in human subjects. It works alongside EU MDR 2017/745 and FDA 21 CFR Part 812, and it does not cover in vitro diagnostic devices, which follow ISO 20916.
A clinical investigation is one method of generating data. Clinical evaluation is the broader, ongoing process of gathering and appraising all clinical evidence for a device, documented in a clinical evaluation report. An investigation feeds the evaluation when existing data is insufficient. Every device needs clinical evaluation; not every device needs a new investigation.
Related Terms
- Clinical Evaluation Report (CER)
- Post-Market Clinical Follow-up (PMCF)
- ISO 14155
- Investigational Device Exemption (IDE)
- ISO 14971
