https://witold.xyz/Witold WięcekBayesian statistics and health research 2026-10-02T23:33:56+01:00 Witold Więcek https://witold.xyz/ Jekyll © 2026 Witold Więcek /assets/img/favicons/favicon.ico /assets/img/favicons/favicon-96x96.png Market shaping and trial design belong together2026-07-20T00:00:00+01:00 2026-10-02T23:33:11+01:00 https://witold.xyz/posts/market-shaping-and-trial-design/ Witold Więcek Originally posted at Clinical Trials Abundance Substack. (1) Innovative trial designs In a nutshell, innovative trial designs (see glossary in a box below) can reduce costs and speed up testing of interventions. They achieve higher efficiency by creating shared trial infrastructures, borrowing information across populations, and adapting testing to information that is learned during the exp... Allocating sample across control and treatment2026-04-15T00:00:00+01:00 2026-06-11T11:55:11+01:00 https://witold.xyz/posts/control-arm-sizing-note/ Witold Więcek Summary In a randomised controlled trial, should your treatment arm(s) and control be allocated the same number of observations? What are the efficiency gains from optimal allocation? In a simple treatment vs control comparison, it is usually OK to have 1:1 allocation, unless there are large differences in cost. If you have an RCT with K treatment arms, in most cases you should consider ... FDA Guidance on Bayesian Clinical Trials2026-01-28T00:00:00+00:00 2026-01-28T00:00:00+00:00 https://witold.xyz/posts/bayesian-fda-guidance/ Witold Więcek So, “FDA is Bayesian now!”? Published originally on Andrew’s blog. The Food & Drug Administration just released a guidance for industry document on “Use of Bayesian Methodology in Clinical Trials of Drug and Biological Products”: This draft guidance provides guidance to sponsors and applicants submitting investigational new drug applications (INDs), new drug applications (NDAs), bio... Bayesian interim analyses in clinical trials2024-03-07T00:00:00+00:00 2024-03-07T19:05:07+00:00 https://witold.xyz/posts/bayesian-interim-analyses/ Witold Więcek These are some basic notes that explain the idea of Bayesian interim analysis in clinical trials in plain language, contrasting it to the frequentist idea. These notes are very simplistic by design, but I provide links at the end that go into more detail and give examples. Let’s say I am running a vaccine trial. In a typical situation, the statistical analysis is supposed to be triggered at so... Publication bias in meta-analysis2024-02-27T00:00:00+00:00 2024-03-04T16:22:19+00:00 https://witold.xyz/posts/pub-bias-meta-analysis/ Witold Więcek Some notes on assessing publication bias. This is my own working in public note, so I will be going a bit into the weeds and skipping over basics. If you stumbled on this note, it may make sense to read a general textbook before diving in. For a quick overview you can try Mathias Harrer’s chapter on publication bias. I will also not get into p-hacking and other research malpractices that every...