Journal Article

Impact of an Online Writing Aid Tool for Writing a Randomized Trial Report: The COBWEB (Consort-Based WEB Tool) Randomized Controlled Trial

Giraudeau, Bruno, Porcher, Raphael, Boutron, Isabelle, Centre d’Investigation Clinique [Tours] CIC 1415 (CIC) ; Centre Hospitalier Régional Universitaire de Tours (CHRU Tours)-Institut National de la Santé et de la Recherche Médicale (INSERM), Columbia Mailman School of Public Health ; Columbia University [New York], Altman, Douglas G, Equipe 5 : METHODS - Méthodes de l’évaluation thérapeutique des maladies chroniques (CRESS - U1153) ; Université Paris Descartes - Paris 5 (UPD5)-Centre de Recherche Épidémiologie et Statistique Sorbonne Paris Cité (CRESS (U1153 / UMR_A_1125 / UMR_S_1153)) ; Institut National de la Recherche Agronomique (INRA)-Université Paris Diderot - Paris 7 (UPD7)-Université Paris Descartes - Paris 5 (UPD5)-Université Sorbonne Paris Cité (USPC)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Institut National de la Recherche Agronomique (INRA)-Université Paris Diderot - Paris 7 (UPD7)-Université Sorbonne Paris Cité (USPC)-Institut National de la Santé et de la

2015

Overview

BACKGROUND: Incomplete reporting is a frequent waste in research. Our aim was to evaluate the impact of a writing aid tool (WAT) based on the CONSORT statement and its extension for non-pharmacologic treatments on the completeness of reporting of randomized controlled trials (RCTs). METHODS: We performed a 'split-manuscript' RCT with blinded outcome assessment. Participants were masters and doctoral students in public health. They were asked to write, over a 4-hour period, the methods section of a manuscript based on a real RCT protocol, with a different protocol provided to each participant. Methods sections were divided into six different domains: 'trial design', 'randomization', 'blinding', 'participants', 'interventions', and 'outcomes'. Participants had to draft all six domains with access to the WAT for a random three of six domains. The random sequence was computer-generated and concealed. For each domain, the WAT comprised reminders of the corresponding CONSORT item(s), bullet points detailing all the key elements to be reported, and examples of good reporting. The control intervention consisted of no reminders. The primary outcome was the mean global score for completeness of reporting (scale 0-10) for all domains written with or without the WAT. RESULTS: Forty-one participants wrote 41 different manuscripts of RCT methods sections, corresponding to 246 domains (six for each of the 41 protocols). All domains were analyzed. For the primary outcome, the mean (SD) global score for completeness of reporting was higher with than without use of the WAT: 7.1 (1.2) versus 5.0 (1.6), with a mean (95 % CI) difference 2.1 (1.5-2.7; P <0.01). Completeness of reporting was significantly higher with the WAT for all domains except for blinding and outcomes. CONCLUSION: Use of the WAT could improve the completeness of manuscripts reporting the results of RCTs. TRIAL REGISTRATION: Clinicaltrials.gov ( http://clinicaltrials.gov NCT02127567 , registration date first received April 29, 2014).

• Publisher: CCSD,BioMed Central,BioMed Central Ltd,Springer Nature B.V
• Subject: Adult / Biomedicine / Consent / Epidemiology / Female / Humans / Intervention / Life Sciences / Male / Medical Writing - standards / Medicine / Medicine & Public Health / Periodicals as Topic - standards / Public health / Publishing - standards / Randomized Controlled Trials as Topic - standards / Research Article / Research Design - standards / Software / Writing
• ISBN: 0003613492000