Bonferroni-based correction factor for multiple, correlated endpoints

Qian D Shi, Emily S. Pavey, Rickey E. Carter

Research output: Contribution to journalArticle

39 Citations (Scopus)

Abstract

Multiple testing and its impact on the type I and type II error rates are frequently discussed in the statistical and biomedical literature. The Bonferroni adjustment is one of the most widely used approaches, yet it suffers from poor statistical performance when there are correlated test statistics. For example, it is criticized to be too conservative. Nonetheless, part of the strong appeal of the Bonferroni approach is the straightforward implementation and relatively intuitive explanation. In this manuscript, a novel adaptation to the traditional Bonferroni approach that accounts for correlated data is proposed. A simple correction factor based on intraclass correlation is applied to the standard Bonferroni method to overcome the shortcomings of the standard Bonferroni adjustment yet maintains its advantages. The method is motivated by an early phase clinical trial examining the effect of a study medication on marijuana craving, which is commonly quantified into four correlated constructs. A detailed simulation study demonstrated that the proposed approach is statistically sound and appropriate for a wide range of common settings.

Original languageEnglish (US)
Pages (from-to)300-309
Number of pages10
JournalPharmaceutical Statistics
Volume11
Issue number4
DOIs
StatePublished - Jul 2012

Fingerprint

Social Adjustment
Bonferroni
Cannabis
Clinical Trials
Adjustment
Intraclass Correlation
Type II error
Correlated Data
Multiple Testing
Appeal
Test Statistic
Error Rate
Intuitive
Simulation Study
Range of data
Craving

Keywords

  • Bonferroni correction
  • correlated data
  • multiple endpoints
  • power
  • type I error rate

ASJC Scopus subject areas

  • Pharmacology (medical)
  • Statistics and Probability
  • Pharmacology

Cite this

Bonferroni-based correction factor for multiple, correlated endpoints. / Shi, Qian D; Pavey, Emily S.; Carter, Rickey E.

In: Pharmaceutical Statistics, Vol. 11, No. 4, 07.2012, p. 300-309.

Research output: Contribution to journalArticle

@article{99ac007b50eb49658c5c33f7dc00a453,
title = "Bonferroni-based correction factor for multiple, correlated endpoints",
abstract = "Multiple testing and its impact on the type I and type II error rates are frequently discussed in the statistical and biomedical literature. The Bonferroni adjustment is one of the most widely used approaches, yet it suffers from poor statistical performance when there are correlated test statistics. For example, it is criticized to be too conservative. Nonetheless, part of the strong appeal of the Bonferroni approach is the straightforward implementation and relatively intuitive explanation. In this manuscript, a novel adaptation to the traditional Bonferroni approach that accounts for correlated data is proposed. A simple correction factor based on intraclass correlation is applied to the standard Bonferroni method to overcome the shortcomings of the standard Bonferroni adjustment yet maintains its advantages. The method is motivated by an early phase clinical trial examining the effect of a study medication on marijuana craving, which is commonly quantified into four correlated constructs. A detailed simulation study demonstrated that the proposed approach is statistically sound and appropriate for a wide range of common settings.",
keywords = "Bonferroni correction, correlated data, multiple endpoints, power, type I error rate",
author = "Shi, {Qian D} and Pavey, {Emily S.} and Carter, {Rickey E.}",
year = "2012",
month = "7",
doi = "10.1002/pst.1514",
language = "English (US)",
volume = "11",
pages = "300--309",
journal = "Pharmaceutical Statistics",
issn = "1539-1604",
publisher = "John Wiley and Sons Ltd",
number = "4",

}

TY - JOUR

T1 - Bonferroni-based correction factor for multiple, correlated endpoints

AU - Shi, Qian D

AU - Pavey, Emily S.

AU - Carter, Rickey E.

PY - 2012/7

Y1 - 2012/7

N2 - Multiple testing and its impact on the type I and type II error rates are frequently discussed in the statistical and biomedical literature. The Bonferroni adjustment is one of the most widely used approaches, yet it suffers from poor statistical performance when there are correlated test statistics. For example, it is criticized to be too conservative. Nonetheless, part of the strong appeal of the Bonferroni approach is the straightforward implementation and relatively intuitive explanation. In this manuscript, a novel adaptation to the traditional Bonferroni approach that accounts for correlated data is proposed. A simple correction factor based on intraclass correlation is applied to the standard Bonferroni method to overcome the shortcomings of the standard Bonferroni adjustment yet maintains its advantages. The method is motivated by an early phase clinical trial examining the effect of a study medication on marijuana craving, which is commonly quantified into four correlated constructs. A detailed simulation study demonstrated that the proposed approach is statistically sound and appropriate for a wide range of common settings.

AB - Multiple testing and its impact on the type I and type II error rates are frequently discussed in the statistical and biomedical literature. The Bonferroni adjustment is one of the most widely used approaches, yet it suffers from poor statistical performance when there are correlated test statistics. For example, it is criticized to be too conservative. Nonetheless, part of the strong appeal of the Bonferroni approach is the straightforward implementation and relatively intuitive explanation. In this manuscript, a novel adaptation to the traditional Bonferroni approach that accounts for correlated data is proposed. A simple correction factor based on intraclass correlation is applied to the standard Bonferroni method to overcome the shortcomings of the standard Bonferroni adjustment yet maintains its advantages. The method is motivated by an early phase clinical trial examining the effect of a study medication on marijuana craving, which is commonly quantified into four correlated constructs. A detailed simulation study demonstrated that the proposed approach is statistically sound and appropriate for a wide range of common settings.

KW - Bonferroni correction

KW - correlated data

KW - multiple endpoints

KW - power

KW - type I error rate

UR - http://www.scopus.com/inward/record.url?scp=84863724227&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84863724227&partnerID=8YFLogxK

U2 - 10.1002/pst.1514

DO - 10.1002/pst.1514

M3 - Article

VL - 11

SP - 300

EP - 309

JO - Pharmaceutical Statistics

JF - Pharmaceutical Statistics

SN - 1539-1604

IS - 4

ER -