-- KathrynStoltzfus - 21 Nov 2017

As a work in progress, we introduce a novel adaptive clinical trial design based on the Second-generation p-value (SGPV). The SGPV indicates when the data are compatible with the alternative hypothesis, the null hypothesis, or inconclusive. Its calculation is fast, straightforward, and can utilize the researcherís preferred interval for the treatment effect, including using standard 95% confidence intervals. It anchors the study design in clinically meaningful thresholds, as opposed to abstract statistical quantities. A trial halts when the data convincingly rule out either clinically meaningless or clinically meaningful effects.

Moreover, design settings can be adjusted to ensure required statistical performance, such as a specified Type I error rate or Power at a given treatment effect size. Our design permits sequential monitoring and seamlessly extends to examinations of multiple endpoints. In a proof of concept, the SGPV based design is compared to other leading approaches via an extensive simulation study.
Topic revision: r1 - 21 Nov 2017, KathrynStoltzfus
 

This site is powered by FoswikiCopyright © 2013-2017 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Vanderbilt Biostatistics Wiki? Send feedback