CONSORT and clinical trial reporting: room for improvement

CONSORT (CONsolidated Standards of Reporting Trials) is a guideline that has been created and revised over several decades to improve reporting of randomized controlled trials (RCTs). This work has been accomplished by an international group of methodologists, guideline developers, knowledge translation specialists, and journal editors. The CONSORT statement was first published in 1996 followed by revisions in 2001 and 2010.1-3 The work done by this group consists of a CONSORT statement, a checklist, the checklist explanation and elaboration document, and a CONSORT flow diagram to summarize how the study population was recruited and handled throughout the study (available at http://www.consort-statement.org/). In January 2007 an editorial was published in TRANSFUSION introducing CONSORT to journal readers and reviewers.4 Our readership was encouraged to become familiar with CONSORT's 22-item checklist and the flow diagram as these tools provided a roadmap for complete and transparent reporting of RCTs. In 2009, a second editorial was published indicating that the journal TRANSFUSION endorsed the use of CONSORT and made it a requirement for authors to follow CONSORT guidelines when submitting a RCT manuscript to the journal.5 At this time the instructions for authors were changed to reflect the new requirement for CONSORT compliance along with a link to the CONSORT website.6 Publishing RCTs in TRANSFUSION has been one of our goals, and although the numbers have remained relatively stable over the years, the opportunity for more RCT publications has expanded with our increasing emphasis on evidence-based transfusion practices and patient blood management. However, publishing clinical trials comes with the responsibility of accepting publications only when they are of high methodologic quality and reported in compliance with CONSORT. Since 2009 there have been a number of changes to CONSORT including a revision (2010) to the checklist increasing the number of items to 25.3 One of the additions in the 2010 revision was the requirement to include information on trial registration. Registration is required to ensure a transparent and nonbiased reporting of clinical trials. Trial registration should always occur before the first patient is enrolled in a study.7 A complete list of changes and additions to the 2010 CONSORT checklist has been published for those readers who are interested in specific details about the evolution of this document. The original CONSORT statement was designed to provide guidance when reporting a parallel-arm RCT. Over the years a number of extensions of the CONSORT statement had been developed to provide more specific guidance for other types of trial designs including: cluster trials, noninferiority and equivalency trials, pragmatic trials, and N-of-1 studies. Extensions have also been developed for other types of interventions including herbal medicinal interventions, nonpharmacologic treatment interventions, and acupuncture. Finally CONSORT extensions have been developed for other types of data: CONSORT–PRO (patient-reported outcomes), harms, and abstracts). The checklists and flow diagrams (if applicable) for each of these extensions contain items specific to the type of study design, intervention, or data and can be accessed at the website http://www.consort-statement.org/extensions. The CONSORT group continues to work on other extensions to make this list more comprehensive. Many scientific journals have adopted the requirement to comply with CONSORT when submitting a clinical study manuscript. This requirement has led to a number of publications examining compliance to the CONSORT statement for RCTs published in the scientific literature. In the clinical research focus section of this issue of TRANSFUSION, Jaffer and colleagues8 have evaluated compliance with CONSORT in our Journal. RCTs that were published in TRANSFUSION between January 2007 and December 2014 were identified through a literature search. They reviewed the information in 62 articles meeting the inclusion criteria (56 primary study reports; six pilot RCTs), to determine how well the information in these publications complied with the requirements specified in CONSORT. The authors found that many of the studies published in TRANSFUSION were not compliant for one or more of the CONSORT items. The items with compliance in less than 50% of the articles fell into the categories of trial registration (registration number and name of the trial registry) and information about study implementation (who generated the random allocation sequence, who enrolled study participants, and who assigned participants to a study intervention). Study participant flow and follow-up information had the highest frequency of compliance, which was greater than 90%. TRANSFUSION is not alone in the need to improve compliance with CONSORT. There are numerous studies over the years that have examined various aspects of CONSORT compliance in different medical specialties including surgery,9-12 traumatic brain injury,13 dentistry,14, 15 and solid organ transplants.16 All of these reports consistently showed a lack of CONSORT compliance often in areas of randomization, concealment, implementation, blinding, trial registration, and protocol availability. Endorsement of the CONSORT guidelines by journals does not guarantee compliance.17 Consistent with all areas of medicine, change is likely to require a multipronged approach including tools to assist with the change, education, and consistent patience and commitment to help authors with this transition. The American Journal of Orthodontics and Dentofacial Orthopedics (AJO-DO) hopes to improve CONSORT compliance with the use of educational videos, links to the CONSORT website, and a more active role by reviewers and editors to guide authors through the process.17 Others have suggested that authors be required to complete and upload the CONSORT checklist with their submission.18 To quote Adie and coworkers,19 "the bar can be raised only through the cooperation of all stakeholders."11 The ultimate responsibility for submitting an article that is written in compliance with CONSORT rests with the submitting authors; however, it is also important that reviewers and editors identify the deficiencies related to CONSORT and only accept studies for publication if the CONSORT requirements have been met. To improve CONSORT compliance for TRANSFUSION, we have made changes within Manuscript Central for authors and reviewers requiring confirmation that the CONSORT guidelines have been followed and that trial registration information is included in the manuscript. These questions will serve as a reminder to authors that TRANSFUSION has endorsed CONSORT and compliance is required, and will remind reviewers to confirm compliance with CONSORT as part of their review process. Also, we encourage researchers to use the CONSORT tools when designing their research projects not just when reporting the results. It will not be possible to comply with all CONSORT requirements if the requirements were not considered when the study was designed! The authors have disclosed no conflicts of interest. Nancy M. Heddle, MSc, FCSMLS(D)1 e-mail: heddlen@mcmaster.ca Paul M. Ness2 1Department of Medicine McMaster University Hamilton, ON, Canada 2Department of Pathology, Medicine and Oncology Johns Hopkins University School of Medicine Baltimore, MD

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.