Clinical Testing

« Back to Glossary Index

What it is

Clinical Testing (devices/IVDs) is a planned investigation in human subjects to assess a device’s safety, performance, and—where applicable—clinical benefit. In the EU this is a clinical investigation (MDR 2017/745, Art. 2(45), Annex XV) or an IVD clinical performance study (IVDR 2017/746, Art. 2(44), Annex XIII). In the U.S., it is a clinical investigation that may require an IDE (21 CFR 812) with human subject protections (21 CFR 50, 56). These studies inform conformity/clearance decisions and post-market risk control.

Human subjectsSafety & performancePre- & post-market

Regulatory framework

  • US (FDA): IDE requirements and exemptions (21 CFR 812); informed consent (21 CFR 50); IRB review (21 CFR 56); records/reports (21 CFR 812 Subpart G); significant vs non-significant risk device determinations.
  • EU (MDR/IVDR): Clinical investigations (MDR Arts. 62–82, Annex XV); IVD clinical performance studies (IVDR Arts. 57–77, Annex XIII); ethics approval and competent authority notifications/authorizations; vigilance during studies (MDR Art. 80; IVDR Art. 76).
  • Foundational standards: ISO 14155:2020 (clinical investigation of medical devices); ISO 20916:2019 (IVD clinical performance studies); Declaration of Helsinki.

Key elements

  • Clear objectives and endpoints tied to intended purpose/claims.
  • Risk management linkage and monitoring plan (ISO 14971, ISO 14155 §4).
  • Defined population, inclusion/exclusion, and study sites.
  • Protocol, investigator brochure/IFU, and statistical plan.
  • Safety reporting and device accountability.

Process — how it works

  • Plan: Define endpoints and sample size; align with risk and claims; then finalize protocol and CIP.
  • Approve: Obtain ethics approval and, as required, CA/FDA authorization (IDE or EU submission route).
  • Start: Train sites; enroll subjects with consent; implement monitoring and data capture.
  • Monitor: Track AEs/SAEs, device deficiencies, and deviations; report per timelines (e.g., MDR Art. 80; 21 CFR 812.150).
  • Analyze: Lock database; perform statistics; compare to success criteria.
  • Report: Issue clinical investigation report and integrate evidence into CER/CEP or performance evaluation.

Common pitfalls

  • Endpoints not linked to intended purpose, which weakens evidence.
  • Wrong risk classification (NSR vs SR) and missing IDE/authorization.
  • Poor safety reporting or late submissions to IRB/CA.
  • Underpowered sample size or unclear statistical success criteria.
  • Logs and accountability gaps (device version, UDI, lot controls).

Quick checks

  • Do endpoints, statistics, and risk controls support the claims?
  • Are IDE/CA and ethics approvals in place before first subject?
  • Are AE/SAE and device deficiency flows defined with timelines?
  • Will results map cleanly into CER/PEP and labeling?

FAQ

Is clinical testing always required?

Not always. For lower-risk devices with adequate clinical evidence from literature or equivalence, regulators may accept non-interventional evidence; however, new or higher-risk claims generally need a study (MDR Arts. 61–62; 21 CFR 812).

What is the difference between SR and NSR in the U.S.?

Significant Risk (SR) studies usually need an IDE; Non-Significant Risk (NSR) may proceed under abbreviated IDE with IRB approval (21 CFR 812.2(b)).

How does IVD clinical testing differ?

IVDs run clinical performance studies to show clinical performance measures (e.g., sensitivity/specificity) and may need specimen handling plans and comparator methods (IVDR Annex XIII; ISO 20916).

Do we need to report SAEs during the study?

Yes. Sponsors must assess, document, and submit required reports to regulators/ethics bodies within set timelines (MDR Art. 80; 21 CFR 812.150).

Can we use post-market data instead of a new study?

Sometimes. If data quality matches the needed endpoints and risk context, literature and real-world data may support claims; nonetheless, gaps may still require a new study.