Systematic Reviews
Copyright ©The Author(s) 2021. Published by Baishideng Publishing Group Inc. All rights reserved.
World J Gastroenterol. Nov 21, 2021; 27(43): 7572-7581
Published online Nov 21, 2021. doi: 10.3748/wjg.v27.i43.7572
Minimum sample size estimates for trials in inflammatory bowel disease: A systematic review of a support resource
Morris Gordon, Svetlana Lakunina, Vasiliki Sinopoulou, Anthony Akobeng
Morris Gordon, Svetlana Lakunina, Vasiliki Sinopoulou, Biomedical Evidence Synthesis and Translation to Practice Unit, School of Medicine, Preston PR1 7BH, United Kingdom
Anthony Akobeng, Division of Gastroenterology, Sidra Med and Res Ctr, Doha 26999, Qatar
Author contributions: Gordon M conceived the study, contributed to design, analysis and writing; Lakunina S led completion, analysis and write up; Sinopoulou V and Akobeng A contributed to analysis, reviewed and edited the write up.
Conflict-of-interest statement: None to declare.
PRISMA 2009 Checklist statement: The authors have read the PRISMA 2009 Checklist, and the manuscript was prepared and revised according to the PRISMA 2009 Checklist.
Open-Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Morris Gordon, MBChB, PhD, Professor, Biomedical Evidence Synthesis and Translation to Practice Unit, School of Medicine, HA340, Harrington Building, 135A Adelphi St, Preston PR1 7BH, United Kingdom. mgordon@uclan.ac.uk
Received: March 4, 2021
Peer-review started: March 4, 2021
First decision: June 3, 2021
Revised: June 30, 2021
Accepted: October 31, 2021
Article in press: October 31, 2021
Published online: November 21, 2021
Processing time: 259 Days and 22 Hours
Abstract
BACKGROUND

Of 25% of randomised controlled trials (RCTs) on interventions for inflammatory bowel disease (IBD) have no power calculation.

AIM

To systematically review RCTs reporting interventions for the management of IBD and to produce data for minimum sample sizes that would achieve appropriate power using the actual clinical data.

METHODS

We included RCTs retrieved from Cochrane IBD specialised Trial register and CENTRAL investigating any form of therapy for either induction or maintenance of remission against control, placebo, or no intervention of IBD in patients of any age. The relevant data was extracted, and the studies were grouped according to the intervention used. We recalculated sample size and the achieved difference, as well as minimum sample sizes needed in the future.

RESULTS

A total of 105 trials were included. There was a large discrepancy between the estimated figure for the minimal clinically important difference used for the calculations (15% group differences observed vs 30% used for calculation) explaining substantial actual sample size deficits. The minimum sample sizes indicated for future trials based on the 25 years of trial data were calculated and grouped by the intervention.

CONCLUSION

A third of intervention studies in IBD within the last 25 years are underpowered, with large variations in the calculation of sample sizes. The authors present a sample size estimate resource constructed on the published evidence base for future researchers and key stakeholders within the IBD trial field.

Keywords: Inflammatory bowel disease; Crohn’s disease; Ulcerative colitis; Gastroen- terology; Statistics; Sample size

Core Tip: This work has identified a large variation in the estimated minimal clinically important difference (MCID) between study groups in inflammatory bowel disease trials in the literature, with no standard to support study designers or reviewers. We have provided a resource to support sample size estimation based on observed MICD in the literature over the last 25 years.