Reporting of randomized controlled trials: the CONSORT statement.

In March 2005, the Journal of Alcohol and Drug Education published an editorial (Sharma, 2005) on the TREND (Transparent Reporting of Evaluations with Nonrandomized Designs) statement for evaluating studies that use non-randomized designs (Des Jarlais, Lyles, Crepaz, and the Trend Group, 2004). A similar checklist has been developed by Moher and colleagues (2001) in the CONSORT (Consolidated Standards of Reporting Trials) statement that has been widely used in evidence-based practice of medicine and has been adopted by a large number of journals that report findings from randomized controlled trials (RCT). Through this editorial we wish to familiarize our readers, contributors, and reviewers to the CONSORT statement. The statement that originated in 1994 has been revised in 2001 and has been translated into Chinese, Dutch, French, German, Italian, Japanese, Korean, Portuguese, Russian and Spanish. Latest information about the statement can be found at the website: http://www.consortstatement.org

There are 22 points in the checklist of the CONSORT statement. In this editorial we will discuss the aspects that are relevant and important for the studies published in our journal. Item number one pertains to reporting in the title and abstract as to how the participants were allocated to interventions. This is a very important aspect and while the authors report it in the body they sometimes fail to report it in the title and the abstract, which must be diligently done. The second item is about reporting of scientific background and explanation of rationale for the study. Most of our authors do a good job in this regard. The third item is about elaboration of eligibility criteria for participants and settings and locations where the data were collected. It is quite pertinent to explicitly mention the inclusion and exclusion criteria for participants in the study and also indicate the location. The fourth item is about reporting in the methods the precise details of the intervention for each group and how and when they were actually administered. The fifth item is about elaboration of specific objectives and hypotheses in the methods section. The sixth item is about clear definition of primary and secondary outcome measures or dependent variables and, if applicable, methods used to enhance the quality of measurements (e.g. multiple measurements, training of assessors). The seventh item is about reporting sample size determination in the methods. It also includes elaboration of any interim analyses and stopping rules especially for long-term trials. The eighth item is about reporting the method used to generate the random allocation sequence including details of any restriction (e.g. blocking, stratification). This reporting helps in assessing whether the process was completely objective or if there was any source of bias. The ninth item is about reporting the method to implement the random allocation sequence (e.g. numbered containers or central telephone). It is important to clarify that the sequence was concealed until interventions were assigned. The tenth item is about reporting who generated the allocation sequence, who enrolled the participants, and who assigned the participants to the groups. The eleventh item is about whether the participants, those administering the interventions, and those assessing the outcomes were blinded to group assignment. Often times for educational trials in alcohol and drug education, those administering the intervention and those assessing the outcomes are the same people and they may not be blinded to group assignment. However, clear reporting in this regard must be done. The final item in the methods section is the reporting of statistical methods for comparing groups for primary outcome(s) and methods for additional analyses.

Item number thirteen recommends that in the results section a diagram that depicts flow of participants through each stage be provided. Such a diagram should include number of participants randomly assigned, number receiving intended treatment, number completing the stud?, protocol, and number analyzed for primary outcome(s). Any deviations from the protocol must be described with reasons. This is an important aspect that has been neglected in articles published in our Journal. Efforts must be made to diagrammatically depict the flow of participants in the trial. Item number fourteen recommends listing of the dates defining the periods of recruitment and follow-up. Item number fifteen recommends reporting of baseline demographic and clinical characteristics of each group. The Journal of Alcohol and Drug Education further recommends that such data be presented in a Table that will usually be Table 1. Item number sixteen recommends reporting the number of participants (denominator) in each group included in analyses and whether the analysis was by "intention to treat." The guidelines further recommend that the results must be reported in absolute numbers where feasible (e.g. 10/20 and not as 50%). Item number seventeen recommends reporting of estimated effect size and its precision (e.g. 95% CI) for each primary and secondary outcome. The American Psychological Association (APA) Manual also recommends similarly, "the general principle to be followed, however, is to provide the reader not only with information about statistical significance but also with enough information to assess the magnitude of the observed effect or relationship" (APA 2001, p. 26). I would encourage all the authors reporting the results of trials to pay special attention to reporting the effect sizes or the measures of the strength of association. Item number eighteen pertains to reporting of ancillary analyses. Once again the Journal of Alcohol and Drug Education recommends that these be presented in a tabular format. Item number nineteen is about reporting of all important adverse reactions and side effects in each intervention group. While this is an important aspect for clinical trials, oftentimes in educational trials this is not a major issue but nonetheless the authors must be vigilant to detect and report any adverse reactions or side effects.

In the discussion section, the first item is item number twenty about reporting of interpretation of results taking into consideration the study hypotheses, sources of potential bias or imprecision and the dangers associated with multiplicity of analyses and outcomes. Some of our authors sometimes create a separate subsection of limitations or write a separate paragraph within the Discussion section, which is an effective method to address the sources of bias or imprecision. Item number twenty-one is about reporting the generalizability (external validity) of the findings. This is again an important aspect and must be addressed by each manuscript reporting results of a trial. The final item, number twenty-two, emphasizes reporting of the results in the context of current evidence or linking the findings with the literature.

On the whole, the CONSORT statement is a useful tool for journals like our Journal of Alcohol and Drug Education. But its utility will only work if our authors and reviewers diligently and voluntarily adopt it in their work.

References

American Psychological Association (2001). Publication Manual of the American Psychological Association (5th ed.). Washington, DC: Author.

Des Jarlais, D. C., Lyles, C., Crepaz, N. & the Trend Group. (2004). Improving the reporting quality of nonrandomized evaluations of behavioral and public health interventions. American Journal of Public Health, 94, 361-366.

Moher, D., Schulz, K. E, Altman, D. G. (2001). The CONSORT statement: Revised recommendations for improving the quality of reports of parallel-group randomized trials. Lancet, 357, 1191-1194.

Sharma, M. (2005) Transparent reporting of evaluations with nonrandomized designs (TREND): A framework for evidence based public health evaluations. [Editorial]. Journal of Alcohol and Drug Education, 49 (1), 3-6.

Manoj Sharma, MBBS, CHES, Ph.D.

Editor, Journal of Alcohol & Drug Education

University of Cincinnati 

526 Teachers College PO Box 210002 Cincinnati, OH 45221-0002