Reporting Completeness of Systematic Reviews of Nutrition or Diet-Related Randomised Controlled Trials: A Meta-Research on Adherence to PRISMA Items.
Silva FM., Adegboye ARA., Curioni C., Gomes F., Collins GS., Kac G., Cook J., Ismail LC., Page MJ., Khandpur N., Lamb S., Hopewell S., Kirtley S., Bernardes S., Durão S., Vorland CJ., Rebelo F., Tibaes JRB., Fink JDS., Tunn R., Cardoso RB., Duarte CK., Stello BB., Frenzel A., Naude C., Zanella PB., Schlussel M.
Adherence to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) 2020 statement has not been explored in the nutrition field. The objectives of this meta-research were to assess the reporting completeness of systematic reviews with meta-analysis (SRs-MA) of nutrition- and diet-related randomized controlled trials (RCTs), and to identify their main reporting limitations, and potential factors associated with adherence to PRISMA. We analyzed a random sample of 100 SRs-MA of nutrition- and diet-related RCTs published in journals indexed on PubMed between July 2021 and July 2022. Pairs of reviewers read the full texts to independently extract the data. One reviewer collected data on the journal's endorsement of PRISMA guidelines. A reporting completeness (RC) score was created using the data from the evaluations regarding whether each fragmented item of PRISMA 2020, PRISMA 2020 Abstracts, PRISMA-Search, and TIDieR checklists (a total of 114 sub-items) was reported in the SRs-MA and compared the RCs between studies grouped according to predefined characteristics. SRs were published in 63 different journals; most (54%) endorsed the PRISMA guidelines. Most SRs-MA (86%) mentioned the PRISMA, and 21 attached the PRISMA checklist. The mean (±SD) RC score was 53.2 ± 7.1%. None of the PRISMA items were completely reported in all SRs-MA. Only 13 items were completely reported in more than 75% of the SRs-MA. Protocol registration (beta = 3.61; 95% CI 1.22-5.99), self-reporting of PRISMA adherence (beta = 4.21; 95% CI 0.72-7.70), and evaluation of the certainty of the body of evidence (beta = 4.99; 95% CI = 2.42-7.55) were associated with RC. The poor RC in SRs-MA of nutrition- and diet-related interventions demonstrates that research in this field requires improvements. Indicators of research integrity and transparency, such as protocol registration, self-reported adherence to PRISMA, and evaluation of the certainty of the body of evidence, were positively associated with RC.