Tid: 31 januari 2018, kl 13-14

Plats: B705

Abstract

Clinical studies comparing a new treatment with a control treatment have often the aim to influence the treatment decision for future patients. Traditionally, a significance test is conducted after the study. Rejection of the null hypothesis “no treatment difference” to the advantage of the new treatment contributes to the decision to use the new treatment for future patients.

As alternative to significance tests, decision theoretic methods have been suggested both for the treatment decision for future patients as well as for designing the clinical study itself. We discuss in this talk two case studies planned for patients with rare diseases (Lyell disease and cystic fibrosis). We show how we apply decision theoretic methods for design of the study (e.g. sample size decision) and for the future treatment decision. We investigate the influence of the chosen input parameters on the optimal sample size for the decision theoretic approach.

Large parts of the talk are based on work in the InSPiRe project, which has received funding from the European Union's Seventh Framework Programme under grant agreement no FP HEALTH 2013602144.

Reference:

Miller F, Zohar S, Stallard N, Madan J, Posch M, Hee SW, Pearce M, Vågerö M, and Day S (2018). Approaches to sample size calculation for clinical trials in rare diseases. Pharmaceutical Statistics. To appear (online available).