When Should You Search Clinical Trial Registries for Completed but Unpublished Studies?

Could completed trials be missing from the published literature?

You search bibliographic databases carefully and retrieve dozens of journal articles. It is tempting to assume that you have found the relevant evidence. For research involving clinical trials, however, some studies may have been initiated and completed without ever appearing as full journal publications.

A clinical trial registry can provide a record of a study independently of whether its findings eventually reach a journal. That makes registries particularly useful when you need to identify unpublished or ongoing studies, investigate possible missing evidence, or reconstruct a more complete set of trials conducted on an intervention.

The practical question is not simply whether registries exist. It is when searching them adds enough value to justify the additional work, and what you can legitimately conclude when a registry record has no corresponding publication.

Search registries when missing trials could materially affect your evidence base

What trial registries can reveal that journal databases cannot

A registry records the existence of a trial, not merely its publication

Bibliographic databases primarily help you find research reports that have entered the scholarly literature. Trial registries serve a different function. They record information about clinical trials, often including the study design, intervention, eligibility criteria, outcomes, sponsor, recruitment status, dates, locations, and identifying information.

The World Health Organization defines a clinical trial, for registration purposes, as research that prospectively assigns human participants or groups to health-related interventions to evaluate effects on health outcomes. The WHO International Clinical Trials Registry Platform, or ICTRP, provides a search portal drawing records from contributing registries. ClinicalTrials.gov is another major registry and results database.

This distinction matters because the published literature may not represent every study that was conducted. Cochrane's guidance on missing evidence specifically recognizes trial registers as sources for identifying initiated, ongoing, and completed studies that may not have been published.

When is a registry search particularly important?

The strongest case arises when you are conducting a systematic review or another evidence synthesis intended to identify eligible clinical trials comprehensively. If your eligibility criteria include randomized or other registered intervention trials, searching only journal databases may leave a systematic blind spot.

A registry search becomes particularly informative when an intervention has been evaluated through regulated clinical development, when you suspect that not all known trials have publications, when published evidence appears unusually sparse relative to the amount of research activity, or when you are specifically assessing whether studies may have disappeared from the visible literature.

It can also help when a published article mentions a trial registration number. Searching that identifier may reveal the original registered outcomes, study dates, recruitment information, posted results, or other records that can be compared with the publication.

Which registries should you search?

Two major starting points are ClinicalTrials.gov and the WHO ICTRP search portal. The latter aggregates trial registration data supplied by contributing registries, which can broaden international coverage.

For comprehensive evidence synthesis, however, one portal should not automatically be treated as a perfect substitute for every relevant registry. Registry coverage, search functionality, indexing, duplicate records, and retrievability can differ. The appropriate sources should therefore be planned according to the review question, intervention, geography, and applicable methodological standards.

What does a completed status actually tell you?

A study marked as completed has, according to the registry record, finished the relevant study activity. That status does not by itself establish what happened to its findings.

Several possibilities remain. Results may have been published under a title that differs substantially from the registry title. A publication may omit or incorrectly report the registration identifier. Results may be posted directly in a registry. A manuscript may exist but be difficult to retrieve. Publication may still be pending. Alternatively, results may genuinely remain unavailable.

Can the registry itself contain results?

Yes. A trial register and a trial results database are conceptually distinct, although a platform may provide both functions. WHO describes a trial results database as a source of results from completed studies accompanied by methodological information needed to interpret them. Those results may or may not have undergone peer review.

ClinicalTrials.gov has accepted structured summary results for registered studies since 2008. Consequently, a trial with no identifiable journal article may still have usable results available through its registry record.

This can matter empirically. The current Cochrane Handbook discusses research in which adding results identified through trial registries changed effect estimates in some existing meta-analyses. Registry searching therefore is not merely administrative detective work. In some reviews, evidence found there can influence the quantitative synthesis itself.

A registry record without results still contains information

Suppose a completed trial has no posted results and no identifiable publication. You cannot infer what its findings were. In particular, you should not assume that the result was negative, null, harmful, or statistically non-significant.

You have nevertheless learned something important: a potentially eligible study appears to have existed, yet its results are unavailable to your review. That information may be relevant when evaluating whether the accessible evidence could give an incomplete picture.

The registry may also tell you the planned sample size, outcomes, intervention groups, completion date, and sponsor. Those details can help determine whether the missing study is potentially consequential and whether further retrieval efforts are justified.

Registry records can help detect missing outcomes as well as missing studies

The problem is not limited to entire unpublished trials. A trial may be published while some prespecified outcomes remain unreported. When earlier registry information is available, researchers can compare registered outcomes with results reported in publications and other sources.

Cochrane's guidance on bias due to missing evidence recommends assembling available sources for a study, which may include registry records, protocols, statistical analysis plans, journal reports, clinical study reports, and information obtained from investigators or sponsors. Discrepancies require careful interpretation because legitimate protocol changes can occur and registry records themselves can be amended.

Registry coverage is not complete across all research areas or historical periods

Trial registration has become substantially more prominent over time, but registry searching does not guarantee that every eligible trial will be found. Older studies may never have been registered. Registration practices also vary across intervention types, jurisdictions, sponsors, and periods.

This is why registries supplement rather than replace conventional literature searching. They are one component of a broader strategy for finding missing evidence, alongside bibliographic databases and, where appropriate, contacting investigators about unpublished work, searching regulatory sources, and examining other forms of grey literature.

How a completed registry record can change your search

Common mistakes when using trial registries

Decide whether registry searching is proportionate to your research question

Not every literature review requires an exhaustive search for unpublished clinical trials. A narrative overview designed to introduce a broad topic has different retrieval requirements from a systematic review estimating an intervention effect. The more strongly your conclusion depends on identifying the complete set of eligible trials, the stronger the rationale for searching registries.

There is also a question of proportionality. Searching for unpublished evidence can consume considerable time, especially when records must be matched manually across multiple sources. The appropriate stopping point depends on the purpose of the review, the likely consequences of missing studies, and the resources available. Those trade-offs are better handled as part of deciding how much effort to spend pursuing unpublished evidence.

Before concluding that your trial evidence is complete

Questions about finding unpublished trials in registries

Use trial registries to look beyond what reached a journal

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.