Navigation
This page is maintained by the Multiplicity
Advisory Board ( Alex
Dmitrienko). Global tests
Tests for the global null hypothesis of no
treatment effect are used in clinical trials with multiple (co-primary)
outcome variables when clinical trial researchers are interested
in studying the effects of an investigational therapy on all of
the variables.
Papers and books
Dallow, N., Leonov, S., Roger. J. (2006).
O'Brien's OLS and GLS statistics in clinical trials: multivariate
approach. GlaxoSmithKline BDS Technical Report 2006-03 [ Download
report]. This technical report gave a detailed review
of global tests in a clinical trial setting, including topics
such as sample szie calculations for the global tests developed
by O'Brien (1984).
Logan, B.R., Tamhane, A.C. (2004). On O'Brien's
OLS and GLS tests for multiple endpoints. Recent Developments
in Multiple Comparison Procedures. IMS Lecture Notes and Monograph
Series. Benjamini, Y., Bretz, F., Sarkar, S. (Editors). 76-88.
This paper discussed the global tests proposed
by O'Brien (1984) and extended the OLS and GLS tests to the heteroscedastic
case (case of unequal correlation matrices).
O'Brien, P.C. (1984). Procedures for comparing
samples with multiple endpoints. Biometrics. 40, 1079-1087.
Popular global tests based on the ordinary
least squares (OLS) method for equally correlated endpoints, generalized
least squares (GLS) method for unequally correlated endpoints
and nonparametric rank-sum method were introduced in the paper.
|
