Reporting and accuracy of sample size calculations in orthodontic trials, focusing on standard deviation assumptions:...

This study aimed to evaluate the adequacy of reporting and the accuracy of sample size calculations, with particular emphasis on standard deviation (SD) assumptions, in randomized controlled trials (RCTs) published in 5 leading orthodontic journals in 2021 and 2022.

This meta-epidemiologic study included RCTs involving human participants from the European Journal of Orthodontics, The Angle Orthodontist, Progress in Orthodontics, American Journal of Orthodontics and Dentofacial Orthopedics, and Orthodontics and Craniofacial Research. We assessed the reporting of essential parameters for sample size calculation, including significance level (α), power (1-β), effect size, and SD. For studies with sufficient data, we recalculated sample sizes and analyzed discrepancies between assumed and recalculated values. We also compared assumed vs observed SDs and explored associations with study characteristics using descriptive statistics and regression analyses.

Of the 107 included RCTs, 62.1% of trials with continuous outcomes provided all essential parameters (41.6% for dichotomous/time-to-event outcomes). Recalculation was feasible in 77.6% of trials. Overall, the assumed sample sizes were generally smaller than the recalculated values, with approximately 55% of studies underestimating the required sample size and approximately 30% showing discrepancies greater than 10%. In addition, around two-thirds of the comparable studies showed differences of more than 30% between the assumed and observed SDs.

The findings suggest that the reporting and accuracy of sample size calculations in orthodontic RCTs could be further improved. Greater emphasis on transparent reporting, adherence to established guidelines such as Consolidated Standards of Reporting Trials (CONSORT), and appropriate statistical support may contribute to strengthening the quality of orthodontic clinical research.