Integrating the predictiveness of a marker with its performance as a classifier

Margaret S. Pepe, Ziding Feng, Ying Huang, Gary Longton, Ross Prentice, Ian M. Thompson, Yingye Zheng

Research output: Contribution to journalArticle

183 Citations (Scopus)

Abstract

There are two popular statistical approaches to biomarker evaluation. One models the risk of disease (or disease outcome) with, for example, logistic regression. A marker is considered useful if it has a strong effect on risk. The second evaluates classification performance by use of measures such as sensitivity, specificity, predictive values, and receiver operating characteristic curves. There is controversy about which approach is more appropriate. Moreover, the two approaches can give contradictory results on the same data. The authors present a new graphic, the predictiveness curve, which complements the risk modeling approach. It assesses the usefulness of a risk model when applied to the population. Although the predictiveness curve relates to classification performance measures, it also displays essential information about risk that is not displayed by the receiver operating characteristic curve. The authors propose that the predictiveness and classification performance of a marker, displayed together in an integrated plot, provide a comprehensive and cohesive assessment of a risk marker or model. The methods are demonstrated with data on prostate-specific antigen and risk factors from the Prostate Cancer Prevention Trial, 1993-2003.

Original languageEnglish (US)
Pages (from-to)362-368
Number of pages7
JournalAmerican Journal of Epidemiology
Volume167
Issue number3
DOIs
StatePublished - Feb 2008

Fingerprint

ROC Curve
Data Display
Prostate-Specific Antigen
Prostatic Neoplasms
Biomarkers
Logistic Models
Sensitivity and Specificity
Population

Keywords

  • Biological markers
  • Classification analysis
  • Diagnostic tests, routine
  • Epidemiologic methods
  • Predictive value of tests
  • Prostate-specific antigen
  • Risk assessment
  • Risk model

ASJC Scopus subject areas

  • Epidemiology

Cite this

Pepe, M. S., Feng, Z., Huang, Y., Longton, G., Prentice, R., Thompson, I. M., & Zheng, Y. (2008). Integrating the predictiveness of a marker with its performance as a classifier. American Journal of Epidemiology, 167(3), 362-368. https://doi.org/10.1093/aje/kwm305

Integrating the predictiveness of a marker with its performance as a classifier. / Pepe, Margaret S.; Feng, Ziding; Huang, Ying; Longton, Gary; Prentice, Ross; Thompson, Ian M.; Zheng, Yingye.

In: American Journal of Epidemiology, Vol. 167, No. 3, 02.2008, p. 362-368.

Research output: Contribution to journalArticle

Pepe, MS, Feng, Z, Huang, Y, Longton, G, Prentice, R, Thompson, IM & Zheng, Y 2008, 'Integrating the predictiveness of a marker with its performance as a classifier', American Journal of Epidemiology, vol. 167, no. 3, pp. 362-368. https://doi.org/10.1093/aje/kwm305
Pepe MS, Feng Z, Huang Y, Longton G, Prentice R, Thompson IM et al. Integrating the predictiveness of a marker with its performance as a classifier. American Journal of Epidemiology. 2008 Feb;167(3):362-368. https://doi.org/10.1093/aje/kwm305
Pepe, Margaret S. ; Feng, Ziding ; Huang, Ying ; Longton, Gary ; Prentice, Ross ; Thompson, Ian M. ; Zheng, Yingye. / Integrating the predictiveness of a marker with its performance as a classifier. In: American Journal of Epidemiology. 2008 ; Vol. 167, No. 3. pp. 362-368.
@article{882525a3168d4105902d3f7446b18f03,
title = "Integrating the predictiveness of a marker with its performance as a classifier",
abstract = "There are two popular statistical approaches to biomarker evaluation. One models the risk of disease (or disease outcome) with, for example, logistic regression. A marker is considered useful if it has a strong effect on risk. The second evaluates classification performance by use of measures such as sensitivity, specificity, predictive values, and receiver operating characteristic curves. There is controversy about which approach is more appropriate. Moreover, the two approaches can give contradictory results on the same data. The authors present a new graphic, the predictiveness curve, which complements the risk modeling approach. It assesses the usefulness of a risk model when applied to the population. Although the predictiveness curve relates to classification performance measures, it also displays essential information about risk that is not displayed by the receiver operating characteristic curve. The authors propose that the predictiveness and classification performance of a marker, displayed together in an integrated plot, provide a comprehensive and cohesive assessment of a risk marker or model. The methods are demonstrated with data on prostate-specific antigen and risk factors from the Prostate Cancer Prevention Trial, 1993-2003.",
keywords = "Biological markers, Classification analysis, Diagnostic tests, routine, Epidemiologic methods, Predictive value of tests, Prostate-specific antigen, Risk assessment, Risk model",
author = "Pepe, {Margaret S.} and Ziding Feng and Ying Huang and Gary Longton and Ross Prentice and Thompson, {Ian M.} and Yingye Zheng",
year = "2008",
month = "2",
doi = "10.1093/aje/kwm305",
language = "English (US)",
volume = "167",
pages = "362--368",
journal = "American Journal of Epidemiology",
issn = "0002-9262",
publisher = "Oxford University Press",
number = "3",

}

TY - JOUR

T1 - Integrating the predictiveness of a marker with its performance as a classifier

AU - Pepe, Margaret S.

AU - Feng, Ziding

AU - Huang, Ying

AU - Longton, Gary

AU - Prentice, Ross

AU - Thompson, Ian M.

AU - Zheng, Yingye

PY - 2008/2

Y1 - 2008/2

N2 - There are two popular statistical approaches to biomarker evaluation. One models the risk of disease (or disease outcome) with, for example, logistic regression. A marker is considered useful if it has a strong effect on risk. The second evaluates classification performance by use of measures such as sensitivity, specificity, predictive values, and receiver operating characteristic curves. There is controversy about which approach is more appropriate. Moreover, the two approaches can give contradictory results on the same data. The authors present a new graphic, the predictiveness curve, which complements the risk modeling approach. It assesses the usefulness of a risk model when applied to the population. Although the predictiveness curve relates to classification performance measures, it also displays essential information about risk that is not displayed by the receiver operating characteristic curve. The authors propose that the predictiveness and classification performance of a marker, displayed together in an integrated plot, provide a comprehensive and cohesive assessment of a risk marker or model. The methods are demonstrated with data on prostate-specific antigen and risk factors from the Prostate Cancer Prevention Trial, 1993-2003.

AB - There are two popular statistical approaches to biomarker evaluation. One models the risk of disease (or disease outcome) with, for example, logistic regression. A marker is considered useful if it has a strong effect on risk. The second evaluates classification performance by use of measures such as sensitivity, specificity, predictive values, and receiver operating characteristic curves. There is controversy about which approach is more appropriate. Moreover, the two approaches can give contradictory results on the same data. The authors present a new graphic, the predictiveness curve, which complements the risk modeling approach. It assesses the usefulness of a risk model when applied to the population. Although the predictiveness curve relates to classification performance measures, it also displays essential information about risk that is not displayed by the receiver operating characteristic curve. The authors propose that the predictiveness and classification performance of a marker, displayed together in an integrated plot, provide a comprehensive and cohesive assessment of a risk marker or model. The methods are demonstrated with data on prostate-specific antigen and risk factors from the Prostate Cancer Prevention Trial, 1993-2003.

KW - Biological markers

KW - Classification analysis

KW - Diagnostic tests, routine

KW - Epidemiologic methods

KW - Predictive value of tests

KW - Prostate-specific antigen

KW - Risk assessment

KW - Risk model

UR - http://www.scopus.com/inward/record.url?scp=38849184152&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=38849184152&partnerID=8YFLogxK

U2 - 10.1093/aje/kwm305

DO - 10.1093/aje/kwm305

M3 - Article

VL - 167

SP - 362

EP - 368

JO - American Journal of Epidemiology

JF - American Journal of Epidemiology

SN - 0002-9262

IS - 3

ER -