Sample size increase during a survival trial when interim results are promising. Issue 14 (18th July 2017)

• Record Type: Journal Article
• Authors: Liu, Yanning; Lim, Pilar
• Is Part Of: Communications in statistics. Volume 46:Issue 14(2017)
• Page Start: 6846
• Page End: 6863
• Publication Date: 2017-07-18
• Subjects: Group sequential design -- promising zone -- sample size re-estimation -- survival trials; Mathematical statistics -- Periodicals; Mathematics; Statistics; 519.2
• DOI: 10.1080/03610926.2015.1137596
• Languages: English
• ISSNs: 0361-0926

Abstract

ABSTRACT: This paper is to extend Mehta and Pocock (2011 ) to provide a way in doing sample size increase in survival trials. Sample space is divided by observed test statistic at interim into three zones: unfavorable, promising, and favorable, within which sample size (required number of events) has a proper increase if falling into the promising zone and otherwise remains unchanged. Simulations with scenarios in the presence/absence of censoring, with/without adaptation, and allowing fourfolds versus twofolds of increase in sample size are compared.