A research registry is a public database where investigators record a study’s design, outcomes, and identifying details before (or, less desirably, after) the study is conducted. Registries exist for several distinct purposes — clinical trials, systematic reviews, and general study protocols each have their own registries, run by different organizations for different reasons. This guide orients research administrators and investigators to the registry landscape: why registration matters, which registry fits which kind of study, and when registration is a legal or editorial requirement versus a strongly encouraged best practice.
Why Prospective Registration Matters
Registration is only meaningful if it happens prospectively — before a study’s results are known, ideally before enrollment or data collection begins. The core problem prospective registration solves is publication bias and selective outcome reporting: without a public, time-stamped record of what a study set out to measure, researchers (or sponsors) can quietly drop disappointing outcomes, add favorable post-hoc ones, or simply decline to publish a “negative” study at all. A registry entry creates a fixed reference point that reviewers, editors, and later meta-analysts can check a published paper against.
The International Committee of Medical Journal Editors (ICMJE) treats the date registration materials were submitted to a qualifying registry — not the date they were posted — as the operative timestamp for judging whether registration was genuinely prospective. Registering only at the time of manuscript submission, after results are already known, does not satisfy ICMJE’s policy, even though the registry record technically exists.
The Registry Landscape, by Study Type
“Research registry” is not one system — it is a family of registries, each scoped to a different kind of research output.
Clinical Trial Registries
ClinicalTrials.gov is operated by the U.S. National Library of Medicine (NLM), part of NIH, in collaboration with the FDA. It is the largest single clinical trial registry and one of the World Health Organization’s recognized “primary registries.” U.S. law (FDAAA 801, expanded by the Final Rule) requires registration and results reporting on ClinicalTrials.gov specifically for a narrower legal category called an “applicable clinical trial” (ACT) — most Phase 2-4 drug and device trials, with Phase 1 trials explicitly out of scope for that particular mandate.
ISRCTN was the first primary registry recognized by WHO’s International Clinical Trials Registry Platform (ICTRP), initiated in 2000. It is owned by a non-profit and administered by BMC (part of Springer Nature). Unlike ClinicalTrials.gov’s narrower ACT-focused legal mandate, ISRCTN accepts all clinical research studies, not just drug and device trials, and records the WHO Trial Registration Data Set before feeding it into ICTRP.
Beyond these two, WHO ICTRP recognizes a network of national and regional primary registries — including the EU’s Clinical Trials Information System (CTIS, the mandatory registration/authorisation route for EU trials since the 2021 Clinical Trials Regulation’s transition completed on 31 January 2025), ANZCTR (Australia/New Zealand), CTRI (India), ChiCTR (China), DRKS (Germany), and JPRN (Japan). A trial registers with one primary registry in the network; it does not register with WHO directly. ICTRP is a search platform across all of them, not a registry itself.
Systematic Review Registries
PROSPERO, maintained by the Centre for Reviews and Dissemination (CRD) at the University of York and funded by the UK’s National Institute for Health and Care Research (NIHR), is the primary international registry for systematic review protocols. Registering a review before it progresses beyond data extraction assigns it a permanent, publicly citable CRD42-prefixed identifier, and locks in the review question, eligibility criteria, search strategy, and planned analysis in advance — preventing the systematic-review equivalent of selective outcome reporting, where a review’s stated conclusions shift to match whatever the included studies happened to show. PROSPERO’s scope covers any systematic review with a health-related outcome, which since a 2021 scope expansion includes reviews from fields such as education or social policy provided the outcome measured is health-related.
Other Study Types
Not every study fits a trial or systematic-review registry. Observational studies, secondary-data analyses, and non-clinical research more broadly are increasingly registered through general-purpose preregistration platforms such as OSF Registries, which lock in a study’s hypotheses and analysis plan without requiring the study to meet any single registry’s eligibility criteria. See CASRAI’s Pre-registration entry for the broader reproducibility concept this serves, and the Preregistration of a Study Protocol guide for how to do it in practice. A Registered Report goes a step further, adding independent peer review of the protocol itself before results exist.
Required vs. Encouraged: Who Actually Has to Register
| Situation | Registration status |
|---|---|
| Trial meeting FDAAA 801’s “applicable clinical trial” definition, conducted in the US | Legally required — registration and results reporting on ClinicalTrials.gov, with civil and funding consequences for noncompliance |
| Trial submitted to an ICMJE-following journal for publication | Editorially required as a condition of consideration for publication, in any WHO ICTRP-qualifying primary registry (not limited to ClinicalTrials.gov) |
| EU-conducted clinical trial since 31 January 2025 | Legally required via the EU Clinical Trials Information System (CTIS) |
| Systematic review with a health-related outcome | Strongly encouraged / often expected by funders, Cochrane, and journals following PRISMA reporting guidance, though PROSPERO itself does not carry a legal mandate |
| Observational study, exploratory analysis, or other non-trial research | Encouraged as a transparency and reproducibility practice, via general preregistration platforms; requirements vary by funder and journal rather than being universal |
The common thread: registration requirements are set by whoever has leverage over the study’s continuation — a funder, a regulator, or a journal — not by the registries themselves. A registry is infrastructure; the mandate to use it comes from elsewhere.
When to Register
“Prospective” has a specific, checkable meaning across every registry type discussed here: the registration record must exist, and be locked, before the outcome is knowable — before first patient enrollment for a trial, before screening begins for a systematic review. Registering retrospectively, after data collection is complete, defeats the purpose even if the registry accepts the submission; ICMJE explicitly does not treat retrospective registration at manuscript-submission time as satisfying its policy. Research administrators supporting investigators should build registry submission into the study start-up checklist, alongside IRB/ethics approval, not as a pre-publication afterthought.
Frequently Asked Questions
Is registering a clinical trial always legally required?
Only for studies that meet the specific legal definition of an “applicable clinical trial” under FDAAA 801 (US) or that fall under the EU Clinical Trials Regulation. Many other trials are registered because a journal requires it as a publication condition (the ICMJE policy) or because a funder mandates it, not because a standalone law requires that specific study to register.
Do I have to use ClinicalTrials.gov specifically?
Only if your trial is legally an “applicable clinical trial” under US law. For ICMJE’s publication-eligibility purposes, registering with any WHO ICTRP-recognized primary registry — ClinicalTrials.gov, ISRCTN, or another national registry in the network — satisfies the policy.
Does PROSPERO accept every systematic review?
No. PROSPERO’s eligibility criteria require a health-related outcome and registration before the review progresses beyond data extraction; a narrative review with no pre-specified protocol, or a review that has already completed data extraction, is not eligible. See CASRAI’s PROSPERO Registration entry for worked examples.
What happens if a trial is never registered?
Consequences depend on which requirement applies: an unregistered “applicable clinical trial” in the US can face FDA civil monetary penalties and funding-agency action; an unregistered trial submitted to an ICMJE-following journal is typically simply not considered for publication.
How does the WHO International Clinical Trials Registry Platform (ICTRP) fit in?
ICTRP is not itself a registry a trial registers with directly — it is a global search platform that aggregates data from WHO’s network of recognized national and regional primary registries (including ClinicalTrials.gov and ISRCTN) via the shared WHO Trial Registration Data Set, so a single search can find a trial regardless of which primary registry it was submitted to.
For the compliance mechanics specific to clinical trials — ICMJE versus FDAAA 801, results-reporting deadlines, and data-sharing plans — see CASRAI’s Clinical Trial Registration and Reporting Compliance guide. For the underlying concept of a single registered trial identifier, see Clinical Trial Registration and Prospective Clinical Trial Registration.
