Group sequential designs for in vivo studies: Minimizing animal numbers and handling uncertainty in power analysis

Research output: Contribution to journalArticlepeer-review

32 Downloads (Pure)

Abstract

Interim analysis is the practice of performing a statistical analysis when the data have only been partially collected, for example, to save resources or to handle the uncertainty of the true effect size. Most statistical designs featuring interim analysis have been developed either in a general statistical setting or for application in clinical trials. As a result, most of them make assumptions and have conditions that in a preclinical setting are usually not met. In this paper, we present necessary changes to the most common forms of interim analysis enhanced for animal experiments, specifically for the t-test and the one-way ANOVA. Finally, we present software that allows freeware use to serve the research community to facilitate the design of experiments featuring interim analyses. The app can be found at icds.be/gsdesigner. It is in the public domain and its code can be found on github.com/ICDS-vubUZ/gsd-designer. In this GitHub folder, one can also find a tutorial for the app.

Original languageEnglish
Pages (from-to)248-254
Number of pages7
JournalResearch in Veterinary Science
Volume145
DOIs
Publication statusPublished - Jul 2022

Bibliographical note

Funding Information:
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Publisher Copyright:
© 2022 Elsevier Ltd

Copyright:
Copyright 2022 Elsevier B.V., All rights reserved.

Keywords

  • Analysis of Variance
  • Animals
  • Research Design
  • Uncertainty

Fingerprint

Dive into the research topics of 'Group sequential designs for in vivo studies: Minimizing animal numbers and handling uncertainty in power analysis'. Together they form a unique fingerprint.

Cite this