Testing for Efficacy in Primary and Secondary Endpoints by Partitioning Decision Paths

• 成果类型: Article
• 署名作者: Liu, Yi; Hsu, Jason
• 署名单位: University System of Ohio; Ohio State University
• 刊物名称: JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
• ISSN/ISSBN: 0162-1459
• DOI: 10.1198/jasa.2009.tm08538
• 发表日期: 2009
• 页码: 1661-1670
• 关键词: dose-response; clinical-trials; multiple

摘要

Testing for efficacy in multiple endpoints has emerged as an important statistical problem. The Food and Drug Administration (FDA) will issue a guidance on Multiple Endpoints in the near future. When there are primary and secondary endpoints, efficacy in the secondary endpoint is relevant only if efficacy in the primary endpoint has been shown. Thus, there are defined paths to decision making. Current approaches to this problem are based on closed testing, testing all possible intersection hypotheses, and collating the results. For decision making to follow predefined paths, strategic choices of test statistics and critical values must be made. As the number of doses and endpoints increase, such strategic choices become increasingly difficult. Partition testing is an alternative to closed testing. It provides insight into confidence sets for stepwise tests, and can be more powerful than closed testing. It also can simplify problem formulation when decision making follows specific paths. For the primary-secondary endpoints problem, we show that partition testing has advantages. When used to implement what we call the decision path principle, partition testing not only drastically reduces the number of hypotheses to be tested, but also guides decision making along predefined paths. With Our way of setting critical values, it has higher probabilities than gatekeeping methods of correctly inferring efficacious primary endpoints as being efficacious, while maintaining the same level of strong FWER control. These advantages are illustrated with a real data example and by simulation.