A two-stage design for comparative clinical trials

The heteroscedastic solution

Elena M. Buzaianu, Pinyuen Chen

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

The problem of comparing several experimental treatments to a control arises frequently in clinical trials. Various multi-stage randomized phase II/III designs have been proposed for the purpose of selecting one or more promising experimental treatments and comparing them with a control, while controlling overall Type I and Type II error rates. In this paper, a hybrid selection and testing design for comparing the means of several experimental normal populations among themselves and with the mean of a control normal population is proposed. It is assumed that the variances of the experimental and the control normal populations are unknown and unequal. A Stein-type two-sample selection approach is used at both the selection and testing stages to solve the heteroscedastic problems caused by the unknown variances. The hybrid two-stage design allows for dropping the poorly performing treatments early on the basis of interim analysis results and for early termination if none of the experimental treatments seems promising. Numerical computations are given to show the advantage of the proposed procedure over a pure selection procedure. An example is provided to illustrate the use of the new procedure.

Original languageEnglish (US)
Pages (from-to)151-177
Number of pages27
JournalSankhya: The Indian Journal of Statistics
Volume80B
DOIs
StatePublished - Jan 1 2018

Fingerprint

Two-stage Design
Clinical Trials
Normal Population
Interim Analysis
Early Termination
Type II error
Sample Selection
Unknown
Testing
Selection Procedures
Unequal
Numerical Computation
Error Rate
Clinical trials

Keywords

  • Clinical trials
  • Hybrid selection and testing design
  • Ranking and selection
  • Stein-type two-stage design

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

A two-stage design for comparative clinical trials : The heteroscedastic solution. / Buzaianu, Elena M.; Chen, Pinyuen.

In: Sankhya: The Indian Journal of Statistics, Vol. 80B, 01.01.2018, p. 151-177.

Research output: Contribution to journalArticle

@article{65154c6569ff402eb66d1dc0ad955766,
title = "A two-stage design for comparative clinical trials: The heteroscedastic solution",
abstract = "The problem of comparing several experimental treatments to a control arises frequently in clinical trials. Various multi-stage randomized phase II/III designs have been proposed for the purpose of selecting one or more promising experimental treatments and comparing them with a control, while controlling overall Type I and Type II error rates. In this paper, a hybrid selection and testing design for comparing the means of several experimental normal populations among themselves and with the mean of a control normal population is proposed. It is assumed that the variances of the experimental and the control normal populations are unknown and unequal. A Stein-type two-sample selection approach is used at both the selection and testing stages to solve the heteroscedastic problems caused by the unknown variances. The hybrid two-stage design allows for dropping the poorly performing treatments early on the basis of interim analysis results and for early termination if none of the experimental treatments seems promising. Numerical computations are given to show the advantage of the proposed procedure over a pure selection procedure. An example is provided to illustrate the use of the new procedure.",
keywords = "Clinical trials, Hybrid selection and testing design, Ranking and selection, Stein-type two-stage design",
author = "Buzaianu, {Elena M.} and Pinyuen Chen",
year = "2018",
month = "1",
day = "1",
doi = "10.1007/s13571-017-0147-9",
language = "English (US)",
volume = "80B",
pages = "151--177",
journal = "Sankhya: The Indian Journal of Statistics",
issn = "0972-7671",
publisher = "Indian Statistical Institute",

}

TY - JOUR

T1 - A two-stage design for comparative clinical trials

T2 - The heteroscedastic solution

AU - Buzaianu, Elena M.

AU - Chen, Pinyuen

PY - 2018/1/1

Y1 - 2018/1/1

N2 - The problem of comparing several experimental treatments to a control arises frequently in clinical trials. Various multi-stage randomized phase II/III designs have been proposed for the purpose of selecting one or more promising experimental treatments and comparing them with a control, while controlling overall Type I and Type II error rates. In this paper, a hybrid selection and testing design for comparing the means of several experimental normal populations among themselves and with the mean of a control normal population is proposed. It is assumed that the variances of the experimental and the control normal populations are unknown and unequal. A Stein-type two-sample selection approach is used at both the selection and testing stages to solve the heteroscedastic problems caused by the unknown variances. The hybrid two-stage design allows for dropping the poorly performing treatments early on the basis of interim analysis results and for early termination if none of the experimental treatments seems promising. Numerical computations are given to show the advantage of the proposed procedure over a pure selection procedure. An example is provided to illustrate the use of the new procedure.

AB - The problem of comparing several experimental treatments to a control arises frequently in clinical trials. Various multi-stage randomized phase II/III designs have been proposed for the purpose of selecting one or more promising experimental treatments and comparing them with a control, while controlling overall Type I and Type II error rates. In this paper, a hybrid selection and testing design for comparing the means of several experimental normal populations among themselves and with the mean of a control normal population is proposed. It is assumed that the variances of the experimental and the control normal populations are unknown and unequal. A Stein-type two-sample selection approach is used at both the selection and testing stages to solve the heteroscedastic problems caused by the unknown variances. The hybrid two-stage design allows for dropping the poorly performing treatments early on the basis of interim analysis results and for early termination if none of the experimental treatments seems promising. Numerical computations are given to show the advantage of the proposed procedure over a pure selection procedure. An example is provided to illustrate the use of the new procedure.

KW - Clinical trials

KW - Hybrid selection and testing design

KW - Ranking and selection

KW - Stein-type two-stage design

UR - http://www.scopus.com/inward/record.url?scp=85050480871&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85050480871&partnerID=8YFLogxK

U2 - 10.1007/s13571-017-0147-9

DO - 10.1007/s13571-017-0147-9

M3 - Article

VL - 80B

SP - 151

EP - 177

JO - Sankhya: The Indian Journal of Statistics

JF - Sankhya: The Indian Journal of Statistics

SN - 0972-7671

ER -