TY - JOUR

T1 - Protocol designed subgroup analyses in multiarmed clinical trials

T2 - Multiplicity aspects

AU - Biesheuvel, Egbert H.E.

AU - Hothorn, Ludwig A.

N1 - Funding Information: We are grateful to the two referees and the guest editor for very helpful comments, and Dr. W. Maurer, Novartis AG, Basel, Switzerland, for the data of the second example. This work was partly supported for the second author by Volkswagen Stiftung project II/78 956.

PY - 2003

Y1 - 2003

N2 - Subgroup analyses are quite common in clinical trials although the subgroup findings are usually presented without adjustment for the multiple analyses, which increases the chance of false positive findings. This paper is focused on the multiplicity aspects of subgroup analyses in multiarmed randomized clinical trials and provides some guidance to practitioners.

AB - Subgroup analyses are quite common in clinical trials although the subgroup findings are usually presented without adjustment for the multiple analyses, which increases the chance of false positive findings. This paper is focused on the multiplicity aspects of subgroup analyses in multiarmed randomized clinical trials and provides some guidance to practitioners.

KW - Multiplicity

KW - Multivariate t-distribution

KW - Subgroup analyses

UR - http://www.scopus.com/inward/record.url?scp=0141924405&partnerID=8YFLogxK

U2 - 10.1081/BIP-120024201

DO - 10.1081/BIP-120024201

M3 - Article

C2 - 14584714

AN - SCOPUS:0141924405

VL - 13

SP - 663

EP - 673

JO - Journal of Biopharmaceutical Statistics

JF - Journal of Biopharmaceutical Statistics

SN - 1054-3406

IS - 4

ER -