Trials will consider manuscripts on any aspect of the design, performance, and findings of randomised controlled trials in any discipline related to health care, and will also encourage the publication of protocols. Trialists will be able to provide the necessary detail for a true and complete scientific record. They will be able to communicate not only all outcome measures, as well as varying analyses and interpretations, but also in-depth descriptions of what they did and honest reflections about what they learnt.
An especially important role for Trials is to make trial protocols available to everyone. A published protocol establishes intellectual property, allows discussion of methodological issues at greater length than is usually allowable, and can be referred to for further methodological detail when reporting the main trial results. Also, this public record of a trial reduces the risk of non-publication and may help to reduce the risk of inappropriate replication of trials. We will encourage authors to accompany their protocol with a discussion of why the trial was set up the way it was, indicating why alternatives were not adopted.
Authors will be expected to follow the recommendations of the CONSORT Statement [6
will give adequate space to present results in suitable detail. After evidence that trials reported as 'short reports' (500–600 words) had serious reporting deficiencies [7
] the Lancet
stopped using that format, but some journals continue. Indeed some journals specifically note that trials with 'negative' findings can be reported in brief, as if somehow those findings are less important. But even full-length reports of say 2000–4000 words frequently fail to include key information about methods and results. As Donald Mainland observed in 1938:
"Limitation of journal space and the expense of publishing numerous or elaborate tables usually prohibit full publication of data, except where these are few in number; and yet incompleteness of evidence is not merely a failure to satisfy a few highly critical readers. It not infrequently makes the data that are presented of little or no value." [8
The results of randomised trials should be reported, regardless of what the findings were. We support the view expressed by Austin Bradford Hill almost 50 years ago:
"A negative result may be dull but often it is no less important than the positive; and in view of that importance it must, surely, be established by adequate publication of the evidence." [9
A similar view was expressed by Iain Chalmers in 1990:
"Failure to publish an adequate account of a well-designed clinical trial is a form of scientific misconduct which can lead to those caring for patients to make inappropriate treatment decisions." [10
And more recently the International Committee of Medical Journal Editors (2004) wrote:
"In return for the altruism and trust that make clinical research possible, the research enterprise has an obligation to conduct research ethically and to report it honestly." [11
Compounding the problems of non-publication of some studies, especially those with non-significant results, it is clear that authors may report their results selectively, emphasising those that are statistically significant [5
will enable and encourage complete reporting of trial results, so that all the collected information can be made available. For this reason, Trials
will consider for publication detailed, extended versions of reports of RCTs that have already been published in a conventional, short form.
Bearing in mind the dangers of over-summarising study data, we encourage graphical displays that show results for individual trial participants. We will also encourage authors to make available all (or some of) the raw data from the trials [12
]. Such detailed reports, and especially the raw data, will be of great value for subsequent systematic reviewers, whose efforts are regularly devalued by the inadequate and even dishonest way in which trials are currently reported.
Trials will thus facilitate the publication of a series of linked publications from a single trial, beginning with the study protocol:
"Electronic publication of a protocol could be simply the first element in a sequence of 'threaded' electronic publications, which continues with reports of the resulting research (published in sufficient detail to meet some of the criticisms of less detailed reports published in print journals), followed by deposition of the complete data set. Not only would this approach allow alternative explorations of the data, it would help to address some of the growing concerns about research misconduct." [13
A key element here, of course, is the availability of a unique trial identifier associated with trial registration [14
]. Registration is a condition for publication in Trials
of any article based on a specific trial.
Trials will be a forum for discussing other aspects of specific randomised trials, as an important element in the full publication of a trial or as an educational resource, or both. Such information may provide insight into scientific or practical problems (e.g. recruitment, choice of outcome measures, resolution of logistic challenges, data monitoring) and thus may have value for those planning and conducting future trials.
In addition, we believe that there is scope for new and better ways to report the findings of trials. Trials will develop and refine innovative approaches to improving communication about trials, in particular to make the article's message comprehensible to a variety of reader groups. Here we anticipate taking advantage of the ingenuity of authors and the fine advisory team that we have assembled.
Trials will have a special section that will present commentaries on, and critiques of, trial reports published in other journals, with neither the time nor word constraints of letters in those journals. We will invite the authors of the original article to respond, and will indicate if they choose not to do so.