Overview of multiple testing methodology and recent development in clinical trials.

摘要: Multiplicity control is an important statistical issue in clinical trials where strong of the type I error rate required. Many multiple testing methods have been proposed and applied to address multiplicity issues trials. This paper provides application oriented comprehensive overview commonly used procedures recent developments methodology Commonly are test non-hierarchical hypotheses gatekeeping can be hierarchically ordered while controlling overall rate. The recently developed graphical approach has flexibility integrate hierarchical into one framework. A procedure with "no-dead-end" opportunity fully recycle α across hypothesis families. Two hypothetical trial examples illustrate applications these procedures. advantages disadvantages different briefly discussed.