Sample size for comparing correlated concordance rates

Sin Ho Jung, Huiman X. Barnhart, Insuk Sohn, Sandra S. Stinnett, David K. Wallace

Research output: Contribution to journalArticle

Abstract

We consider collecting the measurements of a gold standard and two methods with error from each site of subjects. We propose an asymptotic test statistic to compare the concordance rates of two methods with the gold standard and a closed-form sample size formula. Through simulations, we show that the test statistic accurately controls the type I error in small sample sizes, and the sample size formula accurately maintains the power in various settings. The proposed test statistic and the sample size formula can be easily modified for other indices for agreement such as sensitivity and specificity. A real eye study is taken as an example.

Original languageEnglish (US)
Pages (from-to)359-369
Number of pages11
JournalJournal of Biopharmaceutical Statistics
Volume18
Issue number2
DOIs
StatePublished - Mar 1 2008
Externally publishedYes

Fingerprint

Concordance
Sample Size
Test Statistic
Gold
Asymptotic Test
Type I error
Small Sample Size
Specificity
Closed-form
Sensitivity and Specificity
Simulation
Standards

Keywords

  • Cluster size
  • Correlation coefficients
  • GEE
  • Sample size recalculation
  • Sensitivity
  • Specificity

ASJC Scopus subject areas

  • Pharmacology (medical)
  • Pharmacology, Toxicology and Pharmaceutics(all)

Cite this

Sample size for comparing correlated concordance rates. / Jung, Sin Ho; Barnhart, Huiman X.; Sohn, Insuk; Stinnett, Sandra S.; Wallace, David K.

In: Journal of Biopharmaceutical Statistics, Vol. 18, No. 2, 01.03.2008, p. 359-369.

Research output: Contribution to journalArticle

Jung, Sin Ho ; Barnhart, Huiman X. ; Sohn, Insuk ; Stinnett, Sandra S. ; Wallace, David K. / Sample size for comparing correlated concordance rates. In: Journal of Biopharmaceutical Statistics. 2008 ; Vol. 18, No. 2. pp. 359-369.
@article{891947403d38494f835de912888e2598,
title = "Sample size for comparing correlated concordance rates",
abstract = "We consider collecting the measurements of a gold standard and two methods with error from each site of subjects. We propose an asymptotic test statistic to compare the concordance rates of two methods with the gold standard and a closed-form sample size formula. Through simulations, we show that the test statistic accurately controls the type I error in small sample sizes, and the sample size formula accurately maintains the power in various settings. The proposed test statistic and the sample size formula can be easily modified for other indices for agreement such as sensitivity and specificity. A real eye study is taken as an example.",
keywords = "Cluster size, Correlation coefficients, GEE, Sample size recalculation, Sensitivity, Specificity",
author = "Jung, {Sin Ho} and Barnhart, {Huiman X.} and Insuk Sohn and Stinnett, {Sandra S.} and Wallace, {David K.}",
year = "2008",
month = "3",
day = "1",
doi = "10.1080/10543400701697216",
language = "English (US)",
volume = "18",
pages = "359--369",
journal = "Journal of Biopharmaceutical Statistics",
issn = "1054-3406",
publisher = "Taylor and Francis Ltd.",
number = "2",

}

TY - JOUR

T1 - Sample size for comparing correlated concordance rates

AU - Jung, Sin Ho

AU - Barnhart, Huiman X.

AU - Sohn, Insuk

AU - Stinnett, Sandra S.

AU - Wallace, David K.

PY - 2008/3/1

Y1 - 2008/3/1

N2 - We consider collecting the measurements of a gold standard and two methods with error from each site of subjects. We propose an asymptotic test statistic to compare the concordance rates of two methods with the gold standard and a closed-form sample size formula. Through simulations, we show that the test statistic accurately controls the type I error in small sample sizes, and the sample size formula accurately maintains the power in various settings. The proposed test statistic and the sample size formula can be easily modified for other indices for agreement such as sensitivity and specificity. A real eye study is taken as an example.

AB - We consider collecting the measurements of a gold standard and two methods with error from each site of subjects. We propose an asymptotic test statistic to compare the concordance rates of two methods with the gold standard and a closed-form sample size formula. Through simulations, we show that the test statistic accurately controls the type I error in small sample sizes, and the sample size formula accurately maintains the power in various settings. The proposed test statistic and the sample size formula can be easily modified for other indices for agreement such as sensitivity and specificity. A real eye study is taken as an example.

KW - Cluster size

KW - Correlation coefficients

KW - GEE

KW - Sample size recalculation

KW - Sensitivity

KW - Specificity

UR - http://www.scopus.com/inward/record.url?scp=40449109385&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=40449109385&partnerID=8YFLogxK

U2 - 10.1080/10543400701697216

DO - 10.1080/10543400701697216

M3 - Article

VL - 18

SP - 359

EP - 369

JO - Journal of Biopharmaceutical Statistics

JF - Journal of Biopharmaceutical Statistics

SN - 1054-3406

IS - 2

ER -