Reports of randomised control trials should begin and conclude with up-to-date systematic reviews of other relevant trials: a 25-year audit of the quality of trial reports
Year: 2024
Type: review
Abstract: Background: Science is cumulative and should cumulate scientifically. In 1999, we published our first audit of deficiencies in reports of Randomised Control Trials (RCTs) published in five major general medical journals. Over the course of the subsequent 25 years, we conducted five follow-up audits to assess the extent to which new trials had been reported as having been initiated and/or concluded with up-to-date systematic reviews of other relevant trials. Objective: We wished to assess the extent to which systematic reviews had been used in designing and reporting RCTs in five major general medical journals over the 25 years since our first audit. Design: We evaluated 175 reports of RCTs published in Annals of Internal Medicine, BMJ, JAMA, Lancet or New England Journal of Medicine, which had been published in May 1997, 2001, 2005, 2009, 2012 or 2022. Results: In trials published in May 2022, the Introduction sections of 8.8% (3/34) of reports of RCTs published in the five journals referenced up-to-date systematic reviews that had informed the trial’s design, with a further 52.9% (18/34) mentioning a previous systematic review of other trials in the topic area. In the six audits of trials combined across the 25 years, these figures were 2.9% (5/175) and 31.4% (55/175), respectively. In the Discussion sections of trials published in 2022, the RCT findings were integrated in updated systematic reviews in 2.9% (1/34) of reports and in 50% (17/34) cited previous systematic reviews without integrating the new findings. These figures for the whole 25 years were 3.4% (6/175) and 28.6% (50/175).Comparing the first and last years of this audit, 11.5% (3/26) of the trial reports published in 1997 used systematic reviews in their Introduction sections compared with 61.8% (21/34) of trial reports in 2022; while 23.1% (6/26) of the 1997 reports mentioned systematic reviews in their Discussion sections compared with 52.9% (18/34) of trial reports published in 2022. Conclusion: There was a modest increase over 25 years in the use of systematic reviews in designing and interpreting randomised trials reported in five high profile medical journals. Considerable scope remains for setting reports of RCTs in context. Until this deficiency has been addressed by the research community, people are likely to continue to suffer and sometimes to die unnecessarily because of an unreliable evidence base for health care and additional research.
Institutions: Queen's University Belfast, University of Oxford
Language: English
Subfield: Statistics, Probability and Uncertainty
Field: Decision Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W4401173238