Diagnosis using predictive probabilities without cut-offs

Young Ku Choi, Wesley O. Johnson, Mark Thurmond

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

Standard diagnostic test procedures involve dichotomization of serologic test results. The critical value or cut-off is determined to optimize a trade off between sensitivity and specificity of the resulting test. When sampled units from a population are tested, they are allocated as either infected or not according to the test outcome. Units with values high above the cut-off are treated the same as units with values just barely above the cut-off, and similarly for values below the cut-off. There is an inherent information loss in dichotomization. We thus develop a diagnostic screening method based on data that are not dichotomized within the Bayesian paradigm. Our method determines the predictive probability of infection for each individual in a sample based on having observed a specific serologic test result and provides inferences about the prevalence of infection in the population sampled. Our fully Bayesian method is briefly compared with a previously developed frequentist method. We illustrate the methodology with serologic data that have been previously analysed in the veterinary literature, and also discuss applications to screening for disease in humans. The method applies more generally to a variation of the classic parametric 2-population discriminant analysis problem. Here, in addition to training data, additional units are sampled and the goal is to determine their population status, and the prevalence(s) of the subpopulation(s) from which they were sampled.

Original languageEnglish (US)
Pages (from-to)699-717
Number of pages19
JournalStatistics in Medicine
Volume25
Issue number4
DOIs
StatePublished - Feb 28 2006

Fingerprint

Unit
Screening
Infection
Serologic Tests
Population
Information Loss
Diagnostic Tests
Bayesian Methods
Discriminant Analysis
Bayes Theorem
Specificity
Critical value
Diagnostics
Routine Diagnostic Tests
Trade-offs
Paradigm
Optimise
Methodology
Sensitivity and Specificity
Training

Keywords

  • Diagnostic testing
  • Discriminant analysis
  • No gold standard test
  • Prevalence estimation
  • Serologic data

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

Diagnosis using predictive probabilities without cut-offs. / Choi, Young Ku; Johnson, Wesley O.; Thurmond, Mark.

In: Statistics in Medicine, Vol. 25, No. 4, 28.02.2006, p. 699-717.

Research output: Contribution to journalArticle

Choi, Young Ku ; Johnson, Wesley O. ; Thurmond, Mark. / Diagnosis using predictive probabilities without cut-offs. In: Statistics in Medicine. 2006 ; Vol. 25, No. 4. pp. 699-717.
@article{02bee6ffbfe2488bbc924203dce9efb0,
title = "Diagnosis using predictive probabilities without cut-offs",
abstract = "Standard diagnostic test procedures involve dichotomization of serologic test results. The critical value or cut-off is determined to optimize a trade off between sensitivity and specificity of the resulting test. When sampled units from a population are tested, they are allocated as either infected or not according to the test outcome. Units with values high above the cut-off are treated the same as units with values just barely above the cut-off, and similarly for values below the cut-off. There is an inherent information loss in dichotomization. We thus develop a diagnostic screening method based on data that are not dichotomized within the Bayesian paradigm. Our method determines the predictive probability of infection for each individual in a sample based on having observed a specific serologic test result and provides inferences about the prevalence of infection in the population sampled. Our fully Bayesian method is briefly compared with a previously developed frequentist method. We illustrate the methodology with serologic data that have been previously analysed in the veterinary literature, and also discuss applications to screening for disease in humans. The method applies more generally to a variation of the classic parametric 2-population discriminant analysis problem. Here, in addition to training data, additional units are sampled and the goal is to determine their population status, and the prevalence(s) of the subpopulation(s) from which they were sampled.",
keywords = "Diagnostic testing, Discriminant analysis, No gold standard test, Prevalence estimation, Serologic data",
author = "Choi, {Young Ku} and Johnson, {Wesley O.} and Mark Thurmond",
year = "2006",
month = "2",
day = "28",
doi = "10.1002/sim.2365",
language = "English (US)",
volume = "25",
pages = "699--717",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "4",

}

TY - JOUR

T1 - Diagnosis using predictive probabilities without cut-offs

AU - Choi, Young Ku

AU - Johnson, Wesley O.

AU - Thurmond, Mark

PY - 2006/2/28

Y1 - 2006/2/28

N2 - Standard diagnostic test procedures involve dichotomization of serologic test results. The critical value or cut-off is determined to optimize a trade off between sensitivity and specificity of the resulting test. When sampled units from a population are tested, they are allocated as either infected or not according to the test outcome. Units with values high above the cut-off are treated the same as units with values just barely above the cut-off, and similarly for values below the cut-off. There is an inherent information loss in dichotomization. We thus develop a diagnostic screening method based on data that are not dichotomized within the Bayesian paradigm. Our method determines the predictive probability of infection for each individual in a sample based on having observed a specific serologic test result and provides inferences about the prevalence of infection in the population sampled. Our fully Bayesian method is briefly compared with a previously developed frequentist method. We illustrate the methodology with serologic data that have been previously analysed in the veterinary literature, and also discuss applications to screening for disease in humans. The method applies more generally to a variation of the classic parametric 2-population discriminant analysis problem. Here, in addition to training data, additional units are sampled and the goal is to determine their population status, and the prevalence(s) of the subpopulation(s) from which they were sampled.

AB - Standard diagnostic test procedures involve dichotomization of serologic test results. The critical value or cut-off is determined to optimize a trade off between sensitivity and specificity of the resulting test. When sampled units from a population are tested, they are allocated as either infected or not according to the test outcome. Units with values high above the cut-off are treated the same as units with values just barely above the cut-off, and similarly for values below the cut-off. There is an inherent information loss in dichotomization. We thus develop a diagnostic screening method based on data that are not dichotomized within the Bayesian paradigm. Our method determines the predictive probability of infection for each individual in a sample based on having observed a specific serologic test result and provides inferences about the prevalence of infection in the population sampled. Our fully Bayesian method is briefly compared with a previously developed frequentist method. We illustrate the methodology with serologic data that have been previously analysed in the veterinary literature, and also discuss applications to screening for disease in humans. The method applies more generally to a variation of the classic parametric 2-population discriminant analysis problem. Here, in addition to training data, additional units are sampled and the goal is to determine their population status, and the prevalence(s) of the subpopulation(s) from which they were sampled.

KW - Diagnostic testing

KW - Discriminant analysis

KW - No gold standard test

KW - Prevalence estimation

KW - Serologic data

UR - http://www.scopus.com/inward/record.url?scp=31744433060&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=31744433060&partnerID=8YFLogxK

U2 - 10.1002/sim.2365

DO - 10.1002/sim.2365

M3 - Article

C2 - 16220514

AN - SCOPUS:31744433060

VL - 25

SP - 699

EP - 717

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 4

ER -