Ask everyONE: Policy on clinical trials
The following post is by the Senior Editor of PLoS Medicine, Emma Veitch, and is a re-post of the “What is PLoS Medicine’s policy on trials registration?” article written for the Speaking of Medicine blog. Since PLoS ONE also publishes clinical trials, we thought it would be useful to share with everyONE. Emma is also a consulting editor for PLoS ONE.
It’s now well over five years since the International Committee of Medical Journal Editors announced a policy requiring that clinical trials considered for publication in its journals must have been previously registered in a public database. This policy was originally intended to promote greater transparency in clinical trials research, prevent the suppression of trials with “negative” results, and prevent duplication of research. However, PLoS journals still receive many submissions a month reporting results of unregistered trials. So, here’s a quick update of our policy in this area:
Q. What do you define as a “clinical trial” for the purposes of trial registration?
A. We follow the World Health Organisation (WHO) definition. This states that “A clinical trial is 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. Interventions include but are not restricted to drugs, cells and other biological products, surgical procedures, radiologic procedures, devices, behavioural treatments, process-of-care changes, preventive care, etc”. This definition is broader than some other requirements for trial registration (eg those of the Food and Drug Administration in the US). PLoS journals do require that single arm trials, “pilot studies” and trials of non drug interventions must still be registered, whilst some countries’ legislation about trial registration does not cover such studies.
Q. When should trials be registered?
A. All trials should be registered before enrollment of the first participant. However, we do apply exceptions for very old trials that were started before 1 July 2005; these must be registered before submission to our journals.
Q. Which registries are appropriate?
A. Triallists must use a registry that is free, publicly available to all, and that fulfills certain criteria regarding its minimum dataset and operation. PLoS journals require that authors use a registry which is part of the WHO’s network, ensuring that the registry meets these criteria.
Q. How do editors and reviewers use the information in trial registries?
A. Information in a trial registry record reveals key features of a trial, corresponding to the design stage before participants have been enrolled or the findings are known. Therefore editors and reviewers can use the information in trial registry records to check that data corresponding to all prospectively planned trial endpoints have been reported in the paper, using the registry as a tool to address bias from selective outcome reporting.
Q. In Europe many drug trials must by law be registered with EudraCT. Is EudraCT an accepted registry for the purposes of publication in PLoS journals?
A. No. At present access to EudraCT is limited to trial sponsors and regulatory authorities only; it does not fulfill WHO criteria of enabling public access to records in this database. However, the European Commission has stated that ultimately some information from trial records in EudraCT will be made publicly available.