Contents
InsightsJournal and publisher policy
Policy briefGlobal
A Trial Registered After the First Patient Will Not Be Considered
The ICMJE requires a clinical trial to be registered “at or before the time of first patient enrollment” as a condition of consideration for publication, and states plainly that approval from an ethics committee does not satisfy that requirement. The companion rule is the data sharing statement: it must say whether individual deidentified participant data will be shared, and “‘undecided’ is not an acceptable answer” (ICMJE, 2026).
- Register before patient one
- Ethics approval is not registration
- Data sharing statement required
- “Undecided” is not an answer
- Condition of consideration
- Instrument
- ICMJE Recommendations, Sections III.L.1 and III.L.2
- Issued by
- International Committee of Medical Journal Editors
- Version
- Updated January 2026
- Applies to
- Clinical trials submitted to any journal that follows the ICMJE Recommendations
- What you must do
- Register the trial in a public registry at or before the enrolment of the first participant, and submit a data sharing statement that says whether individual deidentified participant data will be shared, what, when and on what access criteria.
- What does not count
- Ethics committee approval. Retrospective registration. A statement that the decision on data sharing is undecided.
- Primary source
- ICMJE Recommendations, updated January 2026
International Committee of Medical Journal Editors · read 8 October 2026
In force Global
What this policy is, in plain words.
Two requirements that between them decide whether a trial can be published at all, and both are settled long before anything is written.
Registration. The trial must be registered “at or before the time of first patient enrollment”. The ICMJE says this is a condition of consideration, which means a journal is not obliged to look at the paper if it was not met. And it closes the most common misunderstanding directly: approval from an ethics committee does not satisfy the requirement. They are different acts with different purposes — one protects participants, the other makes the existence of the trial, and its intended outcomes, public before the data exist.
Data sharing. A clinical trial submission needs a data sharing statement saying whether individual deidentified participant data will be shared. The statement must also cover what data, when, and under what access criteria. The ICMJE forecloses the obvious evasion: “‘undecided’ is not an acceptable answer”. “No” is an acceptable answer. Not having decided is not.
Why it exists.
Prospective registration exists because of what happens without it. A trial that is registered after the results are known can have its primary outcome chosen to match them, and no reader can detect that from the paper. Registering before the first participant fixes the question in public while the answer is still unknown.
It also makes trials that were never published visible. A registry entry with no corresponding publication is evidence that something was run and not reported, which is otherwise invisible to anyone outside the sponsor.
The data sharing rule works on the same principle applied after the fact. Requiring a decision, rather than a reservation, means the reader of the published paper knows whether the underlying data can ever be examined.
What you have to do.
Five steps, four of them before anybody writes anything
Retrospective registration cannot be repaired. It is the one failure on this page with no remedy at writing-up: the date is in a public record, and an editor can see it. Some journals will consider a retrospectively registered trial with a declaration; many will not.
Who and what it reaches.
What it reaches — by subject, by the people it falls on, and by territory.
Subjects
Clinical trials in every specialty
Any prospective study assigning participants to an intervention to study the causal relationship with a health outcome.
Population
Trialists, physician-investigators, sponsors
The obligation falls on whoever is responsible for the trial at enrolment — which is usually not the person who will later write the paper.
Countries
No territorial limit, any WHO-standard registry
The ICMJE accepts registries meeting the WHO International Clinical Trials Registry Platform requirements, which includes national registries in many countries alongside ClinicalTrials.gov.
Which journals apply it, and how.
How a journal applies it. Journals following the ICMJE ask for the registration number at submission and check the registration date against the enrolment date. The BMJ, JAMA, The Lancet, NEJM and Annals of Internal Medicine are among the journals whose editors sit on the ICMJE itself.
CONSORT 2025 reinforces the same ground from the reporting side: registration, protocol and analysis plan access and data sharing are now grouped as open science items on the checklist, so a trial report is asked the same questions twice, by the journal and by the checklist.
The questions people actually ask.
Be straight about it. Give the enrolment date and the registration date, explain what happened, and let the editor decide. Some journals will consider a retrospectively registered trial with a clear declaration, some will not consider it at all, and a few will publish it with an editorial note. What ends a submission is not the late registration itself but the discovery that it was presented as though it had been prospective.
The ICMJE requirement is framed around clinical trials. Some journals encourage or require registration of observational studies too, and some funders do. Check the specific journal: this is an area where journal policy commonly goes beyond the ICMJE.
Yes. “No” is an acceptable answer, and for some studies it is the only honest one — a small rare-disease cohort may not be deidentifiable in any meaningful sense. What is not acceptable is declining to decide.
From the journal’s point of view, the authors are, because they are the ones submitting. This is why the registration date is worth confirming at the start of writing rather than at submission, particularly on a trial you joined after it began.
1BEFORE SUBMISSIONwhile the choices are still open
Journal selectionJournals whose own policies your manuscript can actually satisfy, decided from the manuscript.Medical writingThe manuscript written to the reporting guideline and the journal’s own conditions together.Sources
Every claim above rests on one of these. Each was read on the date shown; a policy can move after that.
- 1International Committee of Medical Journal Editors. (2026). Recommendations for the conduct, reporting, editing, and publication of scholarly work in medical journals. Updated January 2026. Sections III.L.1 (clinical trial registration) and III.L.2 (data sharing and authors’ access to data).
https://www.icmje.org/icmje-recommendations.pdf Read 8 October 2026. - 2International Committee of Medical Journal Editors. (2026, January). Up-dated ICMJE recommendations. Naming clinical trial registration among the sections revised in this version.
https://www.icmje.org/news-and-editorials/updated_recommendations_jan2026.html Read 8 October 2026.
clinical trialstrial registrationICMJEdata sharingprospective registrationresearch transparency
Your next step
Does this rule reach your work?
- 1Check whether it binds the work you have in hand
- 2Put the obligation where the deadline will be met
- 3Keep the evidence that you did
Or read how a compliance statement is checked, on the compliance statement checking page.
More in Journal and publisher policy
How to cite this article
Pick your referencing style and copy it. Your university or target journal may vary the punctuation — check its guide if it has one.
Pubrica Insights. (2026, October 8). A Trial Registered After the First Patient Will Not Be Considered. Pubrica. https://pubrica.com/insights/icmje-trial-registration-data-sharing/
Pubrica Insights. “A Trial Registered After the First Patient Will Not Be Considered.” Pubrica, 8 Oct. 2026, pubrica.com/insights/icmje-trial-registration-data-sharing/.
Pubrica Insights (2026) A Trial Registered After the First Patient Will Not Be Considered. Pubrica. Available at: https://pubrica.com/insights/icmje-trial-registration-data-sharing/ (Accessed: 15 October 2026).
Pubrica Insights. “A Trial Registered After the First Patient Will Not Be Considered.” Pubrica. October 8, 2026. https://pubrica.com/insights/icmje-trial-registration-data-sharing/
Pubrica Insights, ‘A Trial Registered After the First Patient Will Not Be Considered’ (Pubrica, 8 October 2026) <https://pubrica.com/insights/icmje-trial-registration-data-sharing/> accessed 15 October 2026.
Pubrica Insights. A Trial Registered After the First Patient Will Not Be Considered. [Internet]. Pubrica; 2026 Oct 8 [cited 2026 Oct 15]. Available from: https://pubrica.com/insights/icmje-trial-registration-data-sharing/
Pubrica Insights, “A Trial Registered After the First Patient Will Not Be Considered,” Pubrica, Oct. 8, 2026. [Online]. Available: https://pubrica.com/insights/icmje-trial-registration-data-sharing/
@misc{icmje_trial_registration_data_sharing_2026,
author = {{Pubrica Insights}},
title = {A Trial Registered After the First Patient Will Not Be Considered},
howpublished = {Pubrica},
year = {2026},
month = {oct},
url = {https://pubrica.com/insights/icmje-trial-registration-data-sharing/},
note = {Accessed: <span data-cite-acc="iso">2026-10-15</span>}
}This article is written to be learned from. Cite it if you are referring to what it says, and cite the original sources for any evidence you take from it.