Year: 2013
Type: article
Abstract: In the early 1990s, around the same time as The Cochrane Collaboration was being established 1, groups of clinical trialists, journal editors, systematic reviewers and others came together to try to improve the quality of reporting of randomised trials 2, 3. They had all recognised that shortcomings in trial reporting made it difficult for users of these reports to appraise their quality and to understand key elements in the conduct of the trial. In particular, it was difficult to assess the potential for bias in the trial's methodology and, therefore, whether or not it provided a reliable and robust guide to the relative effects of the interventions compared in the study. These gatherings nearly 20 years ago, led to the publication of the first CONSORT statement in 1996 4. CONSORT stands for CONsolidated Standards Of Reporting Trials. It provided a framework for the reporting of randomised trials, with a list of 21 items that should be included in a trial report to help the authors and journal editors to provide key information to their readers on the design, conduct, analysis and interpretation of the trial. This became known as the CONSORT checklist and although it is not intended as a critical appraisal or quality assessment tool, it is widely used to appraise trials. Along with the checklist, CONSORT recommended the inclusion of a flow diagram showing the passage of potential participants through a trial. This includes information on the number of people assessed for eligibility; and the numbers randomised, treated, followed-up and analysed. It also encourages the authors to describe clearly the loss of any randomised participants. CONSORT has been updated twice since the first version, with publications in 2001 5, 6 and 2010 7, 8. The latest version has an updated 25-item checklist and flow diagram, and has been endorsed by numerous journals around the world. A key aim of CONSORT is to improve the use of studies that have evaluated healthcare interventions and its effect has, itself, been evaluated many times over the last decade. In 2012, an international team of researchers, some of whom are involved in the development of CONSORT, prepared a Cochrane Methodology Review bringing this evidence together 9. They included 50 evaluations, which had assessed a collective total of more than 16,000 randomised trials, and had compared the quality of reports in, for example, journals that had endorsed CONSORT and those that had not done so. The Cochrane researchers examined the effects of CONSORT on 27 outcomes which they chose as being related to the completeness of reporting. Almost all of these appeared to favour CONSORT-endorsing journals over non-endorsers. This impact was statistically significant for five items: reporting of the methods for sequence generation and allocation concealment, scientific rationale and background in the report's Introduction, sample size and a composite outcome that counted the number of reported CONSORT items. As well as the revisions to the CONSORT Statement itself, the last decade has also seen the preparation of extensions for particular types of randomised trial and reporting guidelines for several other types of research. There are too many to list here, but to see them and to keep up to date as new guidelines are prepared, visit the website for the EQUATOR (Enhancing the QUAlity and Transparency Of health Research) Network, which maintains an ever-expanding list of reporting guidelines at www.equator-network.org. This section was prepared as part of the work of the All-Ireland Hub for Trials Methodology Research, which is funded by the UK Medical Research Council. The Hub is helping improve the quality of clinical trials, through the conduct and dissemination of research relevant to the methods for trials and other evaluations in health and social care.
DOI: 10.1111/jebm.12025
Language: English
Subfield: Statistics, Probability and Uncertainty
Field: Decision Sciences
Domain: Social Sciences
Open Access status: bronze
OpenAlex ID: W4251223674
