Article type
Year
Abstract
Background: The World Health Organization (WHO) developed the International Clinical Trials Registry Platform (ICTRP) to collect data from trial registries to ensure transparent clinical trial processes from protocol to publication. In 2004, the International Committee of Medical Journal Editors (ICMJE) published a statement supporting prospective registration and in 2007, made prospective registration in a WHO-approved registry a publication pre-requisite. Increasingly registries are accepting retrospectively registered trials. Thus, reporting registration ID numbers does not fulfil the same checks that it had when registries only accepted prospectively registered trials. Now the onus to check registration timing, to align with ICMJE prospective registration mandates, falls on journal editors.
Objectives: To determine whether retrospectively registered trials are reported transparently in top medical journals.
Methods: We identified three ICMJE-member journals according to impact factor (Lancet —33.63, Annals of Internal Medicine—16.72, PlosMed—15.67). We hand-searched the journals between October and December 2011 for all randomised controlled trials (RCTs). Data was extracted, including: title, principle investigator, start date, registry ID number, intervention, primary outcome(s), location, funding source and close-date. Findings were cross-referenced against the registration entry. Descriptive analysis was conducted.
Results: We identified 30 RCTs: The Lancet (19), Annals of Internal Medicine (6) and PlosMed (5). Fourteen of 30 RCTs registered prospectively, and 14 registered retrospectively. One trial registered in a non-ICMJE/WHO-recognised registry, and one was not registered. Of 14 prospectively registered RCTs, all were registered after 2006. Of 14 retrospectively registered RCTs, trial start-dates ranged from 1992 to 2008. Four (29%) prospectively and two (14%) retrospectively registered RCTs reported different outcomes from those in the registry. Funder data was reported inconsistently between registry and publication.
Conclusions: Half of the sampled RCTs are retrospectively registered, showing that registration uptake is still poor. The results indicate that journals are accepting publications that have not complied with ICMJE requirements.
Objectives: To determine whether retrospectively registered trials are reported transparently in top medical journals.
Methods: We identified three ICMJE-member journals according to impact factor (Lancet —33.63, Annals of Internal Medicine—16.72, PlosMed—15.67). We hand-searched the journals between October and December 2011 for all randomised controlled trials (RCTs). Data was extracted, including: title, principle investigator, start date, registry ID number, intervention, primary outcome(s), location, funding source and close-date. Findings were cross-referenced against the registration entry. Descriptive analysis was conducted.
Results: We identified 30 RCTs: The Lancet (19), Annals of Internal Medicine (6) and PlosMed (5). Fourteen of 30 RCTs registered prospectively, and 14 registered retrospectively. One trial registered in a non-ICMJE/WHO-recognised registry, and one was not registered. Of 14 prospectively registered RCTs, all were registered after 2006. Of 14 retrospectively registered RCTs, trial start-dates ranged from 1992 to 2008. Four (29%) prospectively and two (14%) retrospectively registered RCTs reported different outcomes from those in the registry. Funder data was reported inconsistently between registry and publication.
Conclusions: Half of the sampled RCTs are retrospectively registered, showing that registration uptake is still poor. The results indicate that journals are accepting publications that have not complied with ICMJE requirements.