Biometrics Built Better

Re-Randomization Tests in Clinical Trials: A Simulation Case Study & Practical Guidance

Learn how re-randomization tests can strengthen clinical trial analysis with a simulation case study and practical statistical guidance.

Share this article

At the Young Statisticians’ Meeting in Cambridge, Veramed statisticians Huw Wilson and Alice Fletcher presented their poster “Re-Randomization Tests in Clinical Trials: A Simulation Case Study & Practical Guidance.” The poster explores how re-randomization testing can help determine whether an observed treatment effect is likely to be genuine or the result of a Type I (false-positive) error arising from the randomization process.

Using a simulation-based case study, Huw and Alice demonstrate how re-randomization tests can complement traditional statistical analyses by evaluating thousands of alternative treatment allocations consistent with the original trial design. The poster also highlights practical considerations for implementing and validating these methods in clinical trials.

What the Poster Covers

  • How re-randomization tests are used in clinical trials to assess treatment effects.
  • A simulation case study using covariate-adaptive randomization.
  • Interpretation of re-randomization p-values alongside standard statistical analyses.
  • Results from scenarios with and without a true treatment difference.
  • Best practices for simulation generation, validation, and quality control.
  • Practical guidance on when re-randomization testing may add value.

Click on the poster below for an introduction to re-randomization testing and practical insights into its application in modern clinical trials. Huw and Alice will present a more detailed discussion of this work at the upcoming PHUSE EU Conference in Glasgow.

Share this article

Stay Connected

Sign up to receive the latest insights, case studies, and Veramed news—straight to your inbox.

Related Posts