Abstract
Given the key role that pilot and feasibility (PAF) trials play in addressing the challenges of surgical trials, adequate reporting completeness is essential. Our aim was to assess completeness of reporting and evaluate if the items of the Consolidated Standards of Reporting Trials (CONSORT) extension for PAF trials have been reported in surgical PAF trials.
This is a metaresearch study reported according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Studies were included if they were pilot or feasibility randomized trials evaluating a surgical intervention. The primary outcome was overall adherence to the CONSORT statement extension to randomized PAF trials checklist. A Poisson regression was performed to explore the association between research factors and reporting completeness.
MEDLINE, Embase, and Cochrane Central Register of Controlled Trials were searched from January 1-December 31, 2011 and 2021.
After screening 1991 citations, 38 studies from 2011 to 34 studies from 2021 were included. The mean CONSORT reporting score across all included studies was 21.5 (standard deviation 6.3). After excluding items that were not applicable to all studies, a mean of 20.1 (standard deviation 6.1) of 34 items (0.59) were reported. Studies published in 2021 (vs 2011) did not have a greater number of CONSORT items reported (incidence rate ratio [IRR] 1.01, 95% confidence interval [CI] 0.89–1.15). Studies registered in a clinical trial registry (IRR 1.29, 95% CI 1.12–1.48) and randomizing more than 50 patients (IRR 1.16, 95% CI 1.04–1.30) were associated with more CONSORT items reported.
The reporting completeness of surgical PAF trials is poor and has not improved after the publication of the CONSORT extension.
•The reporting completeness of surgical pilot and feasibility trials is poor.•Reporting completeness has not improved since publication of the CONSORT extension.•Registration in trial databases was associated with greater reporting completeness.•Sample sizes more than 50 were associated with greater reporting completeness.•Limitations include sampling bias, lack of statistical power, and heterogeneity.