https://doi.org/10.37527/2023.73.S1
1University Of Oxford, Oxford, United Kingdom, 2Universidade Federal de Ciências da Saúde de Porto Alegre, Porto Alegre, Brazil, 3Cochrane Nutrition, Cape Town, South Africa, 4Pan-American Health Organization, World Health Organization, Washington, DC, USA.
Introduction: As part of a project to consolidate reporting guidance for randomised controlled trials (RCTs), protocols of RCTs and systematic reviews of nutritional interventions, it was important to understand the main reporting limitations and study characteristics potentially associated with adherence to SPIRIT and TIDIeR reporting guidelines. Objectives: To assess reporting completeness in a sample of published protocols of diet and nutrition-related RCTs and explore aspects related to reporting completeness. Methods: We conducted a meta-research of 200 nutrition or diet-related RCT protocols published in 2019 and 2021. Data extraction was performed independently by two reviewers and included bibliometric information, general study characteristics, and SPIRIT and TIDieR items. We calculated the frequency of answers for each question and the score of proportional reporting completeness. We investigated potential associations between selected publication aspects and reporting completeness. Results: The majority of protocols included adults and elderly (36.5%), supplementation as intervention (48.0%), placebo as comparator (44.5%), and evaluated clinical status (40.0%). Most protocols described a parallel RCT (94%) with a superiority framework (70.5%). Overall reporting completeness was 52.04% (±10.78). Adherence to SPIRIT items ranged from 0% (item 18a) to 98.5% (item 10). Adherence to TIDieR items ranged from 35.5% (item 7) to 98.5% (item 1). Reporting completeness was higher in protocols published in 2021 (59.9±13.3%) compared to those published in 2019 (56.3±10.4%); in Methodological journals’ publications (60.8±12.3%) compared to Nutrition (56.6±10.5%) or general Medical journals’ ones (49.7±12.7%); in those registered (58.1±12.1%) compared to those without registration disclosure (49.3±7.5%); and in protocols with self-declared compliance to SPIRIT (62.5±11.2%) compared to those that did not mention it (54.7±11.7%). The lowest and highest reporting completeness scores were observed among protocols with nutrition education (51.9±9.2%) and foods (62.2±9.2%) as interventions, respectively. Conclusions: Reporting completeness in a random sample of nutrition or diet-related RCT protocols was low. Year of publication, protocol registration, journal’s scientific field, type of nutritional intervention, and self-reported adherence to SPIRIT seems to be related to reporting completeness.
Keywords: research integrity, research transparency, randomized controlled trials, nutritional interventions, reporting guidelines.