Kernel-based adaptive randomization toward balance in continuous and discrete covariates

Fei Jiang, Yanyuan Ma, Guosheng Yin

Research output: Contribution to journalArticle

Abstract

Covariate balance among different treatment arms is critical in clinical trials, as confounding effects can be effectively eliminated when patients in different arms are alike. To balance the prognostic factors across different arms, we propose a new dynamic scheme for patient allocation. Our approach does not require discretizing continuous covariates to multiple categories, and can handle both continuous and discrete covariates naturally. This is achieved through devising a statistical measure to characterize the similarity between a new patient and all the existing patients in the trial. Under the similarity weighting scheme, we develop a covariate-adaptive biased coin design and establish its theoretical properties, thus improving the original Pocock-Simon design. We conduct extensive simulation studies to examine the design operating characteristics and we illustrate our method with a data example. The new approach is thereby demonstrated to be superior to existing methods in terms of performance.

Original languageEnglish (US)
Pages (from-to)2841-2856
Number of pages16
JournalStatistica Sinica
Volume28
Issue number4
DOIs
StatePublished - Oct 2018

Fingerprint

Randomisation
Covariates
kernel
Biased Coin Design
Prognostic Factors
Adaptive Design
Alike
Confounding
Operating Characteristics
Clinical Trials
Weighting
Simulation Study
Randomization
Kernel
Similarity
Design

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

@article{13b88eebc4a9478fad5cf35b211ac0cf,
title = "Kernel-based adaptive randomization toward balance in continuous and discrete covariates",
abstract = "Covariate balance among different treatment arms is critical in clinical trials, as confounding effects can be effectively eliminated when patients in different arms are alike. To balance the prognostic factors across different arms, we propose a new dynamic scheme for patient allocation. Our approach does not require discretizing continuous covariates to multiple categories, and can handle both continuous and discrete covariates naturally. This is achieved through devising a statistical measure to characterize the similarity between a new patient and all the existing patients in the trial. Under the similarity weighting scheme, we develop a covariate-adaptive biased coin design and establish its theoretical properties, thus improving the original Pocock-Simon design. We conduct extensive simulation studies to examine the design operating characteristics and we illustrate our method with a data example. The new approach is thereby demonstrated to be superior to existing methods in terms of performance.",
author = "Fei Jiang and Yanyuan Ma and Guosheng Yin",
year = "2018",
month = "10",
doi = "10.5705/ss.202016.0538",
language = "English (US)",
volume = "28",
pages = "2841--2856",
journal = "Statistica Sinica",
issn = "1017-0405",
publisher = "Institute of Statistical Science",
number = "4",

}

Kernel-based adaptive randomization toward balance in continuous and discrete covariates. / Jiang, Fei; Ma, Yanyuan; Yin, Guosheng.

In: Statistica Sinica, Vol. 28, No. 4, 10.2018, p. 2841-2856.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Kernel-based adaptive randomization toward balance in continuous and discrete covariates

AU - Jiang, Fei

AU - Ma, Yanyuan

AU - Yin, Guosheng

PY - 2018/10

Y1 - 2018/10

N2 - Covariate balance among different treatment arms is critical in clinical trials, as confounding effects can be effectively eliminated when patients in different arms are alike. To balance the prognostic factors across different arms, we propose a new dynamic scheme for patient allocation. Our approach does not require discretizing continuous covariates to multiple categories, and can handle both continuous and discrete covariates naturally. This is achieved through devising a statistical measure to characterize the similarity between a new patient and all the existing patients in the trial. Under the similarity weighting scheme, we develop a covariate-adaptive biased coin design and establish its theoretical properties, thus improving the original Pocock-Simon design. We conduct extensive simulation studies to examine the design operating characteristics and we illustrate our method with a data example. The new approach is thereby demonstrated to be superior to existing methods in terms of performance.

AB - Covariate balance among different treatment arms is critical in clinical trials, as confounding effects can be effectively eliminated when patients in different arms are alike. To balance the prognostic factors across different arms, we propose a new dynamic scheme for patient allocation. Our approach does not require discretizing continuous covariates to multiple categories, and can handle both continuous and discrete covariates naturally. This is achieved through devising a statistical measure to characterize the similarity between a new patient and all the existing patients in the trial. Under the similarity weighting scheme, we develop a covariate-adaptive biased coin design and establish its theoretical properties, thus improving the original Pocock-Simon design. We conduct extensive simulation studies to examine the design operating characteristics and we illustrate our method with a data example. The new approach is thereby demonstrated to be superior to existing methods in terms of performance.

UR - http://www.scopus.com/inward/record.url?scp=85054553558&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054553558&partnerID=8YFLogxK

U2 - 10.5705/ss.202016.0538

DO - 10.5705/ss.202016.0538

M3 - Article

AN - SCOPUS:85054553558

VL - 28

SP - 2841

EP - 2856

JO - Statistica Sinica

JF - Statistica Sinica

SN - 1017-0405

IS - 4

ER -