Communicable E56: Frequentist vs Bayesian for clinical trial analysis – 99% probability you’ll want to listen to this
Notas do episódio
In this episode of Communicable, Emily McDonald and Josh Davis are joined by Roger Lewis (USA) and Ian Marschner (Australia) to compare and contrast Bayesian and frequentist statistical approaches. The panel discusses the fundamental principles of both methods, common misconceptions, and the extent to which they are often more similar than many realise. Together, they explore their use in clinical trial design, analysis, and reporting, including adaptive trials and sequential learning. Additional topics include sample size misconceptions, regulatory versus clinical thresholds, and the challenges of interpreting post hoc reanalyses of negative trials.
This episode was edited by Kathryn Hostettler and the executive producer of Communicable is Angela Huttner.
Further reading:
- Berry SM, et al. Bayesian Adaptive Methods for Clinical Trials (Chapman & Hall/CRC Biostatistics Series). Boca Raton (FL): CRC Press; 2010.
- FDA Guidance Document: Use of Bayesian Methodology in Clinical Trials of Drug and Biological Products FDA, 2026, https://www.fda.gov/regulatory-information/search-fda-guidance-documents/use-bayesian-methodology-clinical-trials-drug-and-biological-products
- Lee TC, et al. Contextualizing the use of corticosteroids in severe Pneumocystis jirovecii pneumonia through a Bayesian lens. CMI Comms 2025, https://www.cmi-comms.org/article/S2950-5909(25)00082-4/fulltext
- Livingston EH and Lewis RJ. JAMA Guide to Statistics and Methods, https://jamaevidence.mhmedical.com/Book.aspx?bookId=2742
- Marschner I. Confidence distributions for treatment effects in clinical trials: Posteriors without priors. Stat Med 2024, doi: 10.1002/sim.10000.
- Whitehead J. The design and analysis of sequential clinical trials. Revised 2nd ed. Chichester: John Wiley & Sons; 1997.