ISO 14155:2026: What Changes for Your Clinical Investigations

A manufacturer starting a clinical investigation in October 2026 writes its clinical investigation plan (CIP) against a standard that changed in March. ISO published the fourth edition of ISO 14155 on March 23, 2026. It replaces the 2020 edition with no transition period. Yet the standard harmonized under the MDR is still the 2020 edition. This article sets out what the 2026 edition changes in the CIP, in your committees and in data capture, and what to settle before you submit your next investigation.

What changed between ISO 14155:2020 and ISO 14155:2026?

ISO 14155 is the good clinical practice standard for clinical investigations of medical devices in human subjects. The 2026 edition separates risks from device use and risks from study procedures, gives clinical events committees their own section, strengthens the role of the data monitoring committee and requires more precise statistical definitions.

• Risks: risks from device use go through the ISO 14971 assessment, while procedures the CIP imposes outside routine practice get a descriptive, proportionate assessment. Residual risk assessment becomes explicit.
• Clinical events committee (CEC): a dedicated section for the first time, with documented governance: independence, conflicts of interest, risk of bias.
• Data monitoring committee (DMC): the charter sets the conditions for suspending or stopping the study. Without a DMC, the CIP justifies why.
• Statistics: the ICH E9(R1) estimand framework enters the standard. The CIP defines intercurrent events, missing data handling and sensitivity analyses.
• Adverse events: a more proportionate approach lets you skip routine collection of some events when the CIP states the reason, and a new category covers events related to CIP procedures.

These points come from the analyses of QbD Group and MH Analytics, which agree. The text of the standard is paid: check the clause numbers in the edition you buy.

Does the 2026 edition already apply to your investigation?

The 2026 edition replaces the 2020 edition with no transition period. Under the MDR, however, presumption of conformity rests on a harmonized standard, and that is still EN ISO 14155:2020/A11:2024. Write your new CIPs to the 2026 edition, and keep the 2020 edition as your conformity reference until the Commission harmonizes the new one.

The Commission harmonized EN ISO 14155:2020/A11:2024 through Implementing Decision (EU) 2026/193 of January 28, 2026, two months before the 2026 edition came out. As of September 30, 2026, we found no harmonization decision for the 2026 edition. The text you apply as the state of the art and the one that gives you presumption of conformity are two different documents.

For a study already authorized, run a gap analysis between your CIP and the 2026 edition, then decide on amendments point by point rather than a full rewrite.

How do you document risks in the CIP under the 2026 edition?

Separate two families of risks in the CIP. Risks from device use fall under ISO 14971 and are weighed against the study parameters. Risks from procedures the CIP imposes outside routine practice are described proportionately. State the residual risk for each family.

For a multi-country study, QbD Group notes that the 2026 edition also asks you to document what counts as routine clinical practice in each region. For example, in a post-market study of an implant run in France and Germany, an X-ray at six weeks may be routine practice in one country and not in the other. The CIP then documents the risk of that exam for the sites in the second country.

Which committees and statistical rules should you set before the first patient?

Before the first patient is enrolled, set the CEC membership, the DMC charter or the justification for not having one, and the statistical definitions of the CIP. The table shows what each topic requires in the CIP and what it changes in your electronic case report form.

The Datacapt eCRF module lets you configure these forms in the same database as eConsent and ePRO.

What should you do before your next submission?

Four steps cover the essentials: cite the 2026 edition in every new CIP, name the CEC, write the DMC charter or justify its absence, and define each intercurrent event in the statistical analysis plan.

• Cite "ISO 14155:2026" in the reference list of every new CIP, with the harmonized 2020 edition alongside for presumption of conformity.
• Name the CEC members and have them sign their conflict of interest declarations before the first patient.
• Write the DMC charter with the conditions for suspension and stopping, or state in the CIP why the study has none.
• List each expected intercurrent event in the statistical analysis plan (treatment discontinuation, device change, death) with the handling strategy chosen.

[[cta]]

Where does Datacapt fit in a medical device clinical investigation?

Datacapt brings eCRF, ePRO and eConsent together in one platform, for pre-market clinical investigations as well as post-market clinical follow-up. You configure the capture of events, consents and patient-reported data in a single database. The medical devices page details the modules, and the article on post-market clinical follow-up covers PMCF. For long-term follow-up studies, the registries use case explains the setup.

Until the Commission harmonizes the 2026 edition, your file cites two editions: 2020 for presumption of conformity, 2026 for the state of the art. The day the decision comes out, review the reference list of your CIPs and of your technical documentation.

Frequently asked questions about ISO 14155:2026

[[faq]]

Sources

• ISO, ISO 14155:2026 Clinical investigation of medical devices for human subjects — Good clinical practice, edition 4 (iso.org), 2026
• European Commission, Commission Implementing Decision (EU) 2026/193 of 28 January 2026 amending Implementing Decision (EU) 2021/1182 as regards harmonised standards for […] clinical investigation of medical devices for human subjects […] (EUR-Lex), 2026