Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

The partial area under the summary ROC curve.

S D Walter1

  • 1Department of Clinical Epidemiology and Biostatistics, McMaster University, Health Sciences Center 2C16, 1200 Main Street West, Hamilton, Ontario, Canada L8N 3Z5. walter@mcmaster.ca

Statistics in Medicine
|May 19, 2005
PubMed
Summary

The partial area under the curve (AUC) for summary ROC curves is sensitive to heterogeneity and complicates test comparisons, unlike the full AUC. Full AUC is generally preferred for meta-analysis of diagnostic test accuracy.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Estimating the rate of overdiagnosis with prostate cancer screening: evidence from the Finnish component of the European Randomized Study of Screening for Prostate Cancer.

Cancer causes & control : CCC·2021
Same author

Relative efficiencies of alternative preference-based designs for randomised trials.

Statistical methods in medical research·2020
Same author

A systematic survey of randomised trials that stopped early for reasons of futility.

BMC medical research methodology·2020
Same author

Corrigendum to "Studies reporting ROC curves of diagnostic and prediction data were incorporated into meta-analyses using corresponding odds ratios" [J Clin Epidemiol. 2007 May;60(5):530-4].

Journal of clinical epidemiology·2019
Same author

Randomised trials with provision for early stopping for benefit (or harm): The impact on the estimated treatment effect.

Statistics in medicine·2019
Same author

Letter to the Editor: Preference option randomized design (PORD) for comparative effectiveness research: Statistical power for testing comparative effect, preference effect, selection effect, intent-to-treat effect, and overall effect (SMMR, Vol. 28, Issue 2, 2019).

Statistical methods in medical research·2018

Area of Science:

  • Biostatistics
  • Medical Informatics
  • Diagnostic Test Evaluation

Background:

  • The area under the curve (AUC) summarizes receiver operating characteristic (ROC) curve performance for diagnostic tests.
  • AUC is utilized in meta-analyses to combine sensitivity and specificity estimates into a summary ROC (SROC) curve.
  • Partial AUC has been proposed as an alternative, focusing on specific ROC space regions.

Purpose of the Study:

  • To extend the concept of partial AUC to SROC curves in meta-analysis.
  • To investigate the impact of heterogeneity and truncation on partial AUC and its standard error.
  • To propose a scaled partial area measure for SROC curves.

Main Methods:

  • Theoretical and numerical analyses were conducted.
  • The variation in partial AUC and its standard error were examined.

Related Experiment Videos

  • A scaled partial area measure was developed.

    • Main Results:

    • Partial AUC is sensitive to inter-study heterogeneity.
    • Comparisons between diagnostic tests are more challenging with partial AUC, especially with empirical truncation.
    • The partial AUC measure lacks the symmetry property of the full AUC.

    Conclusions:

    • Partial AUC measures have several disadvantages compared to full AUC.
    • While partial AUC may have some clinical relevance, the full AUC is generally preferred for SROC meta-analyses.
    • Full AUC offers a more robust and comparable measure of overall test performance across studies.