Improving the Quality of Reporting of Randomized Controlled Trials (RCT)

improving-qualityThe controlled clinical trial (CCT), more than any other methodology, can have a strong and immediate impact on patient care. Ideally, the assessment report will provide the reader with the necessary information on the design, processing, analysis and generalizability of the results. This information will help the reader to judge the internal and external validity of the study. A complete and accurate also in the interest of directors and reviewers, in regard to the manuscripts submitted to journals. For the ECC can benefit patients, the report must meet the maximum standards set.

Although a report meets all the items on the checklist and flow diagram contribution is also important how the reporting of RCTs (i.e. format) are presented. For us, the ideal format is one that includes a combination of five subsections in the text of the report and use the checklist during reporting.

Three of the sub-sections fall under the section “Methods” section of the report of a trial: protocol mapping and masking. For example, under the subheading “assignment”, the authors describe the unit of randomization (e.g., patient-level). The other two subsections are included when the authors report the results: flow and monitoring of participants, and analysis. Subsections flow and monitoring of participants are used together when the details of flow chart are described. These five subsections are for the reader to know in each case where to find the most important information. Journals receiving reports ECC should require comprehensive checklist. For example, authors should specify whether or not the clinical study report describes the unit of randomization, and describes where it is documented in the report. In certain cases, changes in the structure of the reports, for example be required when there are differences due to the use of more complex methods.

The advantage of the CONSORT format is that you must not make excessive changes in terms or scope of the manuscript readability and clarity and organization of the report of a trial, with the inclusion of five new subsections, while at the same time, information submitted to the directors and reviewers maximized, using the checklist. This prevents some of the criticisms of the report formats that were proposed in previous occasions.

Some authors, editors, reviewers and even, may find our recommendations for reporting RCTs difficult and even restrictive. Similar views were received when it was proposed that the summaries would provide more information. Both efforts separately as in the CONSORT group were produced based on the need to provide readers with valuable information and enough content on the design, analysis and form of RCTs.

We would be remiss not stigmatize evaluate if the CONSORT approach has had the desired impact. This evaluation should include the design we propose, with the improvements in the reporting of RCTs. You need to evaluate both the process and results, such as the readability of the report, its extension, and other standards of quality assessment. In the coming months we will work on the design and implementation of the evaluation.

The meeting agreed unanimously that the reporting of RCTs and research in general, are often incomplete. We have already cited examples of inadequate reporting and the consequences that this entails. For this reason, we decided that our deliberations we were disseminated as much as possible, hoping that the CONSORT group serve for the reporting of RCTs are more complete and comprehensive. We are aware that these recommendations need to be reviewed, as will having more experience. We invite all journal editors and directors of clinical trials using the checklist and CONSORT box process. Both are available for all managers’ concerned magazines wishing to disseminate this information to its reviewers. Interested readers can also find the checklist and diagram on the website of JAMA.

Financial support for this study was provided by Abbott, Abbott Park, Ill., and the Council of Biology Editors, Northbrook and laboratories. We also thank all members of the group of standards for information on testing and the Asilomar Working Group drafting of Recommendations for the presentation of the Report of Clinical Trials in the Biomedical Literature, which has helped us to reach this point. And to express our sincere thanks to those who reviewed the manuscript.