A Novel Design for Decision Rules Based on Statistical Testing Strategies of Binary Endpoints in a Definitive Go/No-Go Single-Treatment Clinical Study. (May 2014)

• Record Type: Journal Article
• Authors: Zhou, Ming; Shen, Larry Z.
• Is Part Of: Therapeutic innovation & regulatory science. Volume 48:Number 3(2014:May)
• Page Start: 327
• Page End: 335
• Publication Date: 2014-05
• Subjects: statistical testing -- decision making -- Wilson score -- binary endpoint -- sample size; Drugs -- Periodicals; Pharmacology -- Periodicals; Drugs -- Research -- Periodicals; Drugs -- Testing -- Periodicals; 615.1
• DOI: 10.1177/2168479013513583
• Languages: English
• ISSNs: 2168-4790

Abstract

In drug development, a definitive go/no-go clinical study is sometimes conducted before major investment is made to advance the development into the next stage. Such a study sometimes focuses on a key binary endpoint, such as a particular adverse event, treatment failure, or device problem. However, such a study is often constrained by budget and scope, which limit the sponsor's ability to draw clear conclusions. This paper proposes a novel design that uses Wilson confidence intervals to devise statistical testing strategies that allow the decision makers to arrive at certain definite conclusions within the constraint of the study's scope.