Premium
Group sequential and discretized sample size re‐estimation designs: a comparison of flexibility
Author(s) -
Wu Xiaoru,
Cui Lu
Publication year - 2012
Publication title -
statistics in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.996
H-Index - 183
eISSN - 1097-0258
pISSN - 0277-6715
DOI - 10.1002/sim.5395
Subject(s) - sample size determination , interim , computer science , estimation , flexibility (engineering) , sample (material) , statistics , type i and type ii errors , test (biology) , statement (logic) , mathematics , paleontology , chemistry , management , archaeology , chromatography , biology , political science , law , economics , history
In clinical trials, researchers usually determine a study sample size prior to the start of the study to provide a sufficient power at a targeted treatment difference. When the targeted treatment difference deviates from the true one, the study may either have insufficient power or use more subjects than necessary. To address the difficulty in sample size planning, researchers have developed various flexible sample size designs and compared their performances. Some previous work suggests that re‐estimation designs are inefficient and that one can improve uniformly by using standard group sequential likelihood ratio tests, although more interim analyses are involved. However, researchers need to further study the statement and the minimal number of tests needed before a standard group sequential test might outperform a re‐estimation design. In this paper, we conducted simulation studies to answer these questions using various optimality criteria. Copyright © 2012 John Wiley & Sons, Ltd.