top of page

ISO 14155:2026 Is Here — What Changed and What It Means for Your Medical Device Trial

  • May 2
  • 6 min read

Updated: May 2

ClinIQ's Compliance Corner | May 2026


If you work in medical device clinical research, whether as a sponsor, CRO, monitor, investigator, or site coordinator, ISO 14155 is the international standard that shapes how your clinical investigation should be designed, conducted, and reported. The 2026 edition is now published, and it brings meaningful updates that touch everything from how you define your research question to how you classify adverse events and oversee data quality.


This post walks through what changed, why it matters, and what you need to think about as you implement the updated standard.


What Is ISO 14155, and Who Has to Follow It?

ISO 14155 is the international standard for good clinical practice in the clinical investigation of medical devices for human participants. It applies to sponsors, principal investigators, monitors, CROs, IRBs and ethics committees, and anyone involved in the design, conduct, recording, or reporting of a medical device clinical study.


While the FDA has not formally adopted ISO 14155 the way it has adopted ICH E6 for drug and biologic trials, the FDA does recognize the standard and accepts clinical data from studies conducted under it. Adherence to ISO 14155 supports compliance with FDA IDE requirements (21 CFR Part 812), making it best practice to follow both for U.S. medical device trials and essential for any study with an international footprint.


The 2026 update replaces ISO 14155:2020 and is applicable immediately.


A Quick Look at the Structure

ISO 14155:2026 is organized into 10 clauses and 11 annexes. The clauses move from scope and definitions through good clinical practices, ethical considerations, clinical investigation planning and conduct, suspension and close-out, and the responsibilities of both sponsors (Clause 9) and principal investigators (Clause 10).


Annexes are either normative, meaning they carry regulatory weight and are required, or informative, meaning they represent best practice guidance. Annexes A, B, and D (Clinical Investigation Plan, Investigator’s Brochure, and Clinical Investigation Report) are normative. The remaining annexes, including the brand new Annex K on study design, are informative but represent important best practice guidance.


What Changed in ISO 14155:2026?

The 2026 update reflects the growing complexity of modern medical device trials, stronger alignment with risk management principles, and a push toward more rigorous and transparent study design. Changes fall into six broad areas, with several additional updates worth knowing about.


Six major areas of change in ISO 14155:2026 including risk management, governance bodies, study design, participant protections, equipment and documents, and adverse events

1. Risk Management

Device-related risks must now explicitly follow ISO 14971, the international standard for medical device risk management. The connection is made more explicit in 2026, and residual risks must now be formally evaluated and documented.


Procedure-related risks received their own distinct treatment for the first time. Non-routine clinical procedures required by the study protocol, meaning procedures outside standard clinical care, now require a descriptive risk assessment. This is an important distinction for sites and sponsors designing studies that involve procedures beyond routine practice.


Annex H has been expanded to provide more practical guidance on how general device risk management principles translate into study-specific considerations.


2. Governance Bodies — CECs and DMCs

Clinical Events Committees (CECs) — New

CECs are entirely new to ISO 14155:2026. Section 6.12 formally introduces the CEC as an independent committee of clinical experts that sponsors should consider establishing before a clinical investigation begins. The purpose of a CEC is to ensure consistent, unbiased classification of safety and effectiveness events across all participating sites in a multi-center trial and mitigate inadequate reporting risks.


The CEC charter must document the methods for central review and classification, the criteria for determining whether events meet CIP endpoint definitions, and the methods for standardizing endpoint outcomes for statistical analysis.


For sites: If your study has a CEC, you are responsible for reporting events to the sponsor promptly and completely.

For sponsors and CROs: The charter must be in place before the study starts.


Data Monitoring Committees (DMCs) — Updated

Sponsors must now either establish a DMC or formally justify in the CIP why one is not needed. If established, the DMC must be involved in confirming the conditions that would trigger a study suspension or termination.


3. Study Design and Statistics — Estimands and Annex K

This is arguably the most conceptually significant change in the 2026 edition. An estimand is a precise, pre-specified description of the treatment effect a clinical investigation is designed to measure. Think of it as the exact clinical question your study is trying to answer, stated with enough precision that everyone agrees on what is being measured, in whom, under what conditions, and how complications will be handled.


ISO 14155:2026 introduces the estimand framework through Section 6.4 and the new Annex K. An estimand has five defined attributes:

  • Treatment: What intervention is being studied?

  • Population: Which participants does the question apply to?

  • Variable (endpoint): What outcome is being measured and when?

  • Population-level summary: How will the treatment effect be expressed statistically?

  • Intercurrent events: What happens when something disrupts the planned course of treatment?


Intercurrent events are events occurring after a participant starts the intervention that can affect the interpretation of results, such as withdrawal, treatment switch, or death. These must be defined and handled prospectively in the statistical analysis plan, along with non-inferiority margins and missing data strategies.


Annex K contains an example of an estimand and its attributes.

able K.1 from ISO 14155:2026 showing an example estimand and its five attributes for a chronic pain medical device study

For sites: Follow the CIP’s instructions on what data to collect when intercurrent events occur.

For sponsors and CROs: The estimand must drive the protocol and cannot be defined after the fact.


4. Participant Protections

  • Deviating from inclusion or exclusion criteria now requires a formal protocol amendment and cannot be handled as a site-level deviation or exception.

  • Informed consent must be obtained from the participant’s legally designated representative where applicable, and individuals must be given the opportunity to discuss participation with family members before consenting.

  • Informed consent forms must now describe both the future use of biological samples and the future use of health-related data.

 

5. Equipment and Documents

  • Equipment calibration is now explicitly mandatory and must be addressed in the CIP.

  • Implant cards must be addressed in the CIP and provided to participants who receive an implanted device.

  • The CIP must now address the methods and timing for data analysis.

  • The CIP must explicitly address requirements for participant follow-up and continued care in the event of study suspension or termination, including care that differs from routine clinical practice.

  • The Investigator’s Brochure must now address in-silico data from preclinical testing, device training requirements, and device-specific precautions.

 

6. Adverse Events

The 2026 edition restructures adverse event categories into three distinct types aligned with the updated risk management framework:

  • Non-device-use related

  • Device-use related

  • Non-routine, protocol-related


A new category for adverse events associated with a device deficiency has also been added (Figures F.1 and F.2). Reduced adverse event recording and reporting is now explicitly permitted, but only when justified in the CIP and pre-approved by the IRB or ethics committee.


Additional Changes Worth Knowing

Beyond the six major categories, the 2026 foreword identifies several other updates:

  • Revised Definition of Clinical Performance Shifted from device behavior to an outcomes-oriented definition centered on achieving intended purpose and clinical benefit for participants.

  • Procedure for Suspension and Termination: Updated Section 8.2.2 & 8.2.3 Overhauled to separate suspension and premature termination into distinct procedures, each with specific requirements for proportionate response, participant follow-up, and communication obligations.

  • Clinical Quality Management: Updated Section 9.1 Formalizes the expectation that sponsors apply quality management principles across the full clinical investigation lifecycle.

  • Development Stage Applicability: Updated Annex I Clarifies how ISO 14155 requirements apply across different clinical development stages, from first-in-human through confirmatory investigations.

 

Putting It All Together

ISO 14155:2026 reflects a maturing field. The addition of estimands, CECs, strengthened risk management integration, and formalized quality management signals that medical device clinical research is being held to an increasingly rigorous standard, one that demands earlier planning, more precise thinking, and stronger cross-functional collaboration between clinical, statistical, regulatory, and site teams.


For sites The most immediate implications are in adverse event categorization, participant protections, equipment calibration, and understanding how CEC reporting works if your study has one.

For sponsors and CROs The estimand framework, CEC charter development, risk-based monitoring plans, and the new clinical quality management requirements deserve early and serious attention in study planning.


The standard is available for purchase directly from ISO at iso.org. As always, ClinIQ’s Compliance Corner will keep pointing you straight to the source.




 
 
 

Comments


bottom of page