Information on adverse events in randomised clinical trials assessing drug interventions published in four medical journals with high impact factors.

Abstract

The impact of the extension of the Consolidated Standards of Reporting Trials (CONSORT) statement, which was published in 2004 and aimed to improve the quality of the safety information presented in clinical trials, remains uncertain. To assess the incorporation of the CONSORT statement extension's recommendations in randomised clinical trials (RCTs) evaluating drug therapies published in high-impact medical journals. Using Medline, 122 RCTs published in 2009 were selected from BMJ, JAMA, Lancet, and NEJM. A structured form was used to identify the harms information reported in the RCTs, following the recommendations of the CONSORT statement extension. The most frequently met CONSORT recommendation was the mention of harms in the title or abstract of the paper (72.1% of the papers analysed); the least-met recommendation was the reporting of how the harms information was collected (10.7%). The studies that focused on harms presented better information on safety, but only 10.8% met all recommendations in the CONSORT statement. The adverse event information was insufficient for the RCTs published in four high-impact medical journals five years after the publication of the extension of the CONSORT statement.