“Clinical trial registration” and “results reporting” are often treated as one requirement, but they are legally and administratively distinct obligations that happen to converge on the same public record. Registration is largely a publication-eligibility rule enforced by journal editors; results reporting is a separate federal disclosure obligation enforced by regulators and funders, with its own deadlines, its own definition of which trials are covered, and its own penalties. A trial can be fully compliant on one and non-compliant on the other. This guide covers why registration exists, what actually counts as a “clinical trial” under each rule (the definitions genuinely differ), how ClinicalTrials.gov relates to the World Health Organization’s global registry network, and what results-reporting compliance requires once a trial is registered.
For adjacent CASRAI Dictionary entries, see ICH GCP (Good Clinical Practice), IRB (Institutional Review Board), informed consent, and pre-registration.
Why registration exists: ICMJE’s publication-eligibility requirement
The registration norm originated with journal editors, not regulators. In September 2004, the International Committee of Medical Journal Editors (ICMJE) announced that its member journals would require registration in a public trials registry as a condition of consideration for publication. The policy took effect for trials that began enrollment on or after 1 July 2005; trials already enrolling before that date had to be registered by 13 September 2005 to remain publishable in an ICMJE journal.
ICMJE’s current recommendations set specific criteria for what counts as an acceptable registry and when registration must happen:
- Prospective registration — at or before the time of first patient consent for enrollment. Registering retrospectively, at the time of manuscript submission, does not satisfy the requirement. ICMJE treats the date the registration was submitted to the registry as the operative date, not the date it was posted publicly.
- An acceptable registry must be free and publicly accessible, open to all prospective registrants at no charge, managed by a not-for-profit organization, electronically searchable, and must capture a minimum data set — currently the WHO Trial Registration Data Set (24 items) — at the time of registration. ClinicalTrials.gov and any WHO ICTRP primary registry (below) qualify.
- Since 1 January 2019, trials beginning enrollment on or after that date must also include a data sharing plan as part of the registration itself — a separate but related requirement (see the data-sharing section below).
What actually counts as a “clinical trial” — two different definitions
This is the part most explainers skip, and it’s where registration and results-reporting compliance genuinely diverge: ICMJE and the US federal regulation that governs ClinicalTrials.gov do not use the same definition of “clinical trial.”
ICMJE’s definition is broad and phase-agnostic: “any research project that prospectively assigns people or a group of people to an intervention, with or without concurrent comparison or control groups, to study the relationship between a health-related intervention and a health outcome.” That covers drugs, devices, surgical procedures, behavioral treatments, educational programs, and dietary interventions — and it includes Phase 1 trials.
The US regulatory definition, under Section 801 of the Food and Drug Administration Amendments Act of 2007 (FDAAA), is narrower and is called an “applicable clinical trial” (ACT): controlled trials of drugs and biologics other than Phase 1 investigations, plus certain applicable device trials (including required pediatric postmarket surveillance). Phase 1 drug trials are explicitly excluded from the FDAAA registration/results mandate even though ICMJE would still require them to be registered for publication purposes.
A separate, broader NIH policy (below) closes part of that gap for NIH-funded research specifically, by requiring registration and results reporting for all NIH-funded clinical trials, not only FDAAA-defined ACTs. The practical takeaway for a research office: “is this trial covered?” has to be answered three times, against three different definitions (ICMJE, FDAAA/ACT, NIH), not once.
ClinicalTrials.gov: the primary US registry
ClinicalTrials.gov is operated by the National Library of Medicine (NLM), part of the National Institutes of Health, in collaboration with the FDA. It functions as both the venue most US-based trials use to satisfy ICMJE’s publication-eligibility rule and the system of record for the separate US regulatory reporting obligation under FDAAA 801, described below. It is also one of the “primary registries” recognized by the WHO International Clinical Trials Registry Platform (next section), so a trial registered there is automatically discoverable through WHO’s global search.
Under FDAAA 801 and its implementing regulation, an applicable clinical trial initiated after 27 September 2007 must be registered on ClinicalTrials.gov within 21 days of enrolling its first participant — a firm regulatory deadline, distinct from ICMJE’s “before first patient consent” rule, which is a journal-editorial standard rather than a federal one.
WHO’s International Clinical Trials Registry Platform, and other primary registries
The WHO International Clinical Trials Registry Platform (ICTRP) is not itself a registry that trials register with directly. It is a search platform that aggregates records from a network of national and regional “primary registries” that each independently meet WHO’s registry criteria — ClinicalTrials.gov (US), the EU’s Clinical Trials Information System (CTIS, which became the sole mandatory EU registration route as of 31 January 2025, replacing the older EudraCT/EU Clinical Trials Register), ISRCTN, ANZCTR (Australia/New Zealand), CTRI (India), ChiCTR (China), DRKS (Germany), JPRN (Japan), and others. A trial only has to register with one primary registry to be indexed by ICTRP; it does not need to register with WHO separately.
Two mechanisms keep a trial identifiable across this distributed network:
- The WHO Trial Registration Data Set — the 24-item minimum data set every primary registry must capture, which is what makes cross-registry search and comparison possible at all.
- The Universal Trial Number (UTN) — a voluntary identifier a trial can obtain from WHO, independent of any specific registry’s own ID (like an NCT number from ClinicalTrials.gov). Because a multi-country trial is sometimes registered in more than one national registry, the UTN exists to unambiguously tie those separate registration records back to the same underlying trial — a role broadly analogous to how a DOI or RAiD disambiguates a single research output or project across systems, though the UTN is specific to trial registrations and is not itself a DOI-style resolvable identifier.
Results-reporting compliance: what’s required beyond registering
Registering a trial is necessary but not sufficient. Two separate rules govern what has to happen after the trial is registered:
FDAAA 801 Final Rule (42 CFR Part 11)
The HHS Final Rule implementing FDAAA 801’s results-reporting mandate took effect 18 January 2017. For an applicable clinical trial, summary results — including participant flow, baseline characteristics, outcome measures, and adverse event data — must be submitted to ClinicalTrials.gov no later than 12 months after the trial’s primary completion date, with a possible extension of up to two years in specific circumstances (e.g., the product is not yet approved and is still under active development for the studied use). The “responsible party” — generally the trial sponsor, or in some cases the principal investigator — is accountable for both registration and results submission.
Noncompliance carries real consequences, not just a reputational risk: FDA can issue a public Notice of Noncompliance, which starts a 30-day remediation clock, and can pursue civil monetary penalties of up to $10,000 per day of continued violation after that window (a statutory base figure, subject to periodic inflation adjustment), in addition to potential grant-funding actions by the awarding federal agency.
NIH’s own policy — broader than FDAAA alone
A separate NIH policy, the NIH Policy on the Dissemination of NIH-Funded Clinical Trial Information, also took effect 18 January 2017. It applies the same registration and results-reporting expectation to every NIH-funded clinical trial, using NIH’s own, broader definition of “clinical trial” — not only the narrower FDAAA “applicable clinical trial” category. That means an NIH-funded behavioral or Phase 1 trial that falls outside FDAAA’s ACT definition can still be subject to NIH’s own registration and reporting expectation, enforced through the grant’s terms and conditions of award rather than FDA’s civil-penalty authority. (This detail is well corroborated across university research-office guidance interpreting the policy; the primary grants.nih.gov policy page returned an access error during this review and was not independently re-fetched — reviewers relying on the exact current text should check the live NIH page directly.)
The data-sharing plan folded into registration
Since ICMJE’s June 2017 policy announcement, journals following ICMJE recommendations have required a data sharing statement with manuscripts reporting clinical trial results submitted on or after 1 July 2018, and — as noted above — a data sharing plan as part of the trial’s registration itself for trials beginning enrollment on or after 1 January 2019. The requirement is one of transparency, not a mandate to share data: it obliges investigators to state clearly whether individual participant data will be shared, and under what conditions, not to commit to sharing it. See CASRAI’s guide on how to write a data availability statement for reproducibility for how this plays out in the eventual manuscript.
Why this matters for research-information systems, not just compliance offices
A clinical trial’s registry number (an NCT number from ClinicalTrials.gov, or an equivalent from another WHO primary registry) functions as a de facto persistent identifier for the trial itself — the anchor that a CRIS/RIM system, a funder’s reporting system, and the eventual publication all need to reference consistently to establish that they’re describing the same underlying study. In practice, that linkage is still mostly manual: registries, publications, and grant records aren’t guaranteed to cross-reference each other automatically the way a DOI resolves to a canonical record or an ORCID iD anchors a researcher’s identity across systems. Institutions building out CRIS interoperability increasingly want a trial’s registry number treated the same way — a stable key that ties protocol, funding record, ethics approval, and eventual publication together — rather than a field that only matters at the point of journal submission.
Frequently asked questions
Is clinical trial registration mandatory?
It depends which rule you mean. ICMJE registration is a condition of publication in ICMJE-following journals, not a law. FDAAA 801 registration (for FDAAA-defined “applicable clinical trials”) is a US federal legal requirement, with a 21-day deadline and civil penalties for noncompliance. NIH adds its own, broader funding-conditioned requirement for NIH-funded trials. A given trial can be subject to one, two, or all three simultaneously.
What happens if a clinical trial isn’t registered or results aren’t reported?
For ICMJE purposes, an unregistered trial’s results generally can’t be published in a journal that follows ICMJE recommendations. For FDAAA-covered applicable clinical trials, FDA can issue a public Notice of Noncompliance, followed by civil monetary penalties (up to $10,000 per day of continued violation after a 30-day remediation window) if the deficiency isn’t corrected. Funders, including NIH, can also take funding actions independent of FDA’s enforcement.
What’s the difference between ClinicalTrials.gov and the WHO ICTRP?
ClinicalTrials.gov is a primary registry — trials register there directly, and it’s the US system of record for FDAAA 801 compliance. WHO ICTRP is not a registry itself; it’s a search platform that aggregates records from ClinicalTrials.gov and other national/regional primary registries worldwide, so a trial only needs to register with one primary registry to be discoverable through WHO’s global search.
How long do I have to report clinical trial results after a trial ends?
For an FDAAA-applicable clinical trial, summary results are due on ClinicalTrials.gov no later than 12 months after the trial’s primary completion date, with a possible extension (up to two years in some cases) if the studied product isn’t yet approved for the relevant use.
Does every clinical trial need to be registered on ClinicalTrials.gov specifically?
No. ICMJE accepts registration with any WHO ICTRP primary registry, not only ClinicalTrials.gov — the EU’s CTIS, ISRCTN, ANZCTR, and others also qualify, provided they meet WHO’s registry criteria. ClinicalTrials.gov is specifically required, rather than merely accepted, only for trials that fall under the US FDAAA 801 / NIH reporting mandates.
What is the WHO Trial Registration Data Set?
It’s the minimum set of data items (currently 24) that a WHO-recognized primary registry must capture about a trial at the time of registration — things like the scientific title, health condition studied, intervention, key inclusion/exclusion criteria, and primary outcome. It’s what makes the registries in WHO’s network structurally comparable and searchable as one system, despite being independently operated.
Related CASRAI resources
See also the Integrity & Compliance pillar page for the broader cluster this guide belongs to; the CASRAI Dictionary’s ICH GCP (Good Clinical Practice), IRB (Institutional Review Board), informed consent, pre-registration, and data availability statement entries; the data availability statement guide; and the Compliance and regulatory and Research integrity and misconduct dictionary domains.
