The journal editors' trial registration requirement
Journals that follow the International Committee of Medical Journal Editors consider a clinical trial for publication only if it was registered in a public registry before its first participant was enrolled. The requirement fixes which registries count, which studies count as clinical trials and from which date each kind of trial is covered, and journals differ in how they treat a trial registered late.
How the requirement arose and spread
In September 2004 the committee's 12 member journals announced that, for a clinical trial that began recruiting on or after 1 July 2005, they would consider the results for publication only if investigators had reported certain details of the trial's design in a public registry before patient enrolment. By 2007 dozens of other journals had adopted that policy. By 2014 trial registration was required as a condition of publication in many journals. Prospective registration, before the first patient is enrolled, is a condition for publishing a trial's results in all journals that have adopted the committee's principles, as their instructions for authors show.
As it had promised, the committee re-evaluated the policy two years after it took effect. In June 2007 the member journals published a joint editorial announcing three sets of changes: acceptance of additional registries, a change in the definition of a clinical trial, and guidance for avoiding charges of prior publication.
Since the requirement was introduced, trial registrations have skyrocketed and opportunities for later selective or biased reporting of trials have plummeted. By 2019, registering a new trial in a registry the committee accepts was an established procedure for most trial sponsors.
Registries the committee accepts
Registration needs to be done in a public database such as ClinicalTrials.gov. Before June 2007 the committee recognised five registries: www.actr.org.au, www.ClinicalTrials.gov, www.ISRCTN.org, www.umin.ac.jp/ctr and www.TrialRegister.nl. From June 2007 it also accepted registration in any of the primary registries participating in the World Health Organization's International Clinical Trials Registry Platform. In 2007 the platform's primary registries were to provide data directly to the WHO, and its associate registries were to feed data to primary registries. The committee did not accept associate registries, because they may be managed by for-profit entities. By 2019 the committee accepted registration in ClinicalTrials.gov and in any of the primary registries that participate in the platform.
Which trials must be registered
In accord with the WHO, the committee defines a clinical trial as "any research study that prospectively assigns human participants or groups of humans to one or more health-related interventions to evaluate the effects on health outcomes". Health-related interventions include any intervention used to modify a biomedical or health-related outcome, such as drugs, surgical procedures, devices, behavioural treatments, dietary interventions and process-of-care changes. Health outcomes include any biomedical or health-related measures obtained in patients or participants, including pharmacokinetic measures and adverse events. This definition applies to trials that begin enrolment on or after 1 July 2008. In 2019 the committee endorsed the dissemination of information based on this WHO definition.
In May 2005 the committee had changed its original definition of a clinical trial to exclude phase I trials, the preliminary trials designed to study pharmacokinetics or major unknown toxicity. The committee explained its reversal of that exclusion by stating that having information about phase I trials in the public domain can be beneficial, because these studies can guide future research or signal safety concerns. All phase 2 and 3 trials that started enrolling patients on or after 1 July 2005, and all phase 1 studies started on or after 1 July 2008, should be prospectively registered before publication.
One 2014 account gives the committee's definition of a clinical trial as "any research project that prospectively assigns human subjects to intervention and comparison groups to study the cause-and-effect relationship between a medical intervention and a health outcome", and states that the World Health Organization supports it as the world standard. In 2020 one practitioner described the committee as making it mandatory to register any clinical trial and to disclose any result, negative ones included, for any trial that prospectively assigns a treatment to people or to groups of people, that is, for interventional trials. The same practitioner said that many initiatives were under way to widen the set of clinical trials that carry these obligations.
Checking registration dates and accepting late registration
Timely registration can be easily established, because all trial registries record the date the trial was first registered and the date the first participant was enrolled or randomised. A journal's editorial staff routinely cross-check these dates when a manuscript is submitted.
In 2014 most medical journals followed the committee's guidelines and were therefore very strict about accepting for publication only trials that had been prospectively registered, but many still accepted retrospective registration in certain cases. Some journals may reject a manuscript that does not fulfil the committee's registration criteria, while others may be more lenient. Even so, in the experience of two practitioners, all journals insist on registration in a registry the committee accepts as a condition of reviewing the manuscript, even when the trial was registered retrospectively. If a trial is not registered when the manuscript is prepared, it is worth asking the target journal whether it will accept retrospective registration and whether a rationale for the delay is required.