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

On uncertainty in medical testing.

Robert L Winkler1, James E Smith

  • 1Fuqua School of Business, Duke University, Durham, North Carolina 27708-0120, USA. rwinkler@duke.edu

Medical Decision Making : an International Journal of the Society for Medical Decision Making
|November 10, 2004
PubMed
Summary

Uncertainty in medical test results, including disease probability and test accuracy, can be managed using expected values in Bayesian formulas. This approach simplifies calculations for positive predictive values, correcting common errors.

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

Enhancing the Time-to-Positivity Paradigm for Guidance of Antimicrobial Therapy: A Model-Based Approach.

Computers & chemical engineering·2025
Same author

Fellowship Training After General Surgery Residency: Does it Pay?

Journal of surgical education·2025
Same author

Exploration of zileuton protective mechanisms against vancomycin-associated nephrotoxicity.

Toxicology and applied pharmacology·2025
Same author

Impact of iron limitation on in vitro and in vivo activity of ceftazidime against A. baumannii.

Diagnostic microbiology and infectious disease·2025
Same author

Drug development studies supporting zileuton as a parenteral adjuvant to attenuate antibiotic-associated nephrotoxicity.

Antimicrobial agents and chemotherapy·2025
Same author

Application of the wildland fire emissions inventory system to estimate fire emissions on forest lands of the United States.

Carbon balance and management·2024

Area of Science:

  • Medical Decision Making
  • Diagnostic Test Evaluation
  • Bayesian Statistics in Medicine

Background:

  • Significant confusion exists in medical literature regarding the handling of uncertainty in diagnostic testing.
  • This uncertainty pertains to both the pretest probability of a disease and the accuracy (sensitivity and specificity) of diagnostic tests.

Purpose of the Study:

  • To clarify methods for calculating posttest disease probabilities when faced with uncertainty in pretest probability and test characteristics.
  • To demonstrate how to compute a distribution for posttest probabilities.
  • To correct a common error in calculating positive predictive values under uncertainty.

Main Methods:

  • Utilized Bayesian inference to model disease probability.
  • Employed expected values of uncertain parameters (pretest probability, sensitivity, specificity) within the standard Bayesian formula.

Related Experiment Videos

  • Assessed the impact of parameter uncertainty on posttest probabilities and positive predictive values.

    • Main Results:

    • Under specific independence assumptions, uncertainty in pretest probability, sensitivity, and specificity does not complicate posttest probability calculations.
    • Using expected values of these parameters in the Bayesian formula yields accurate patient positive predictive values.
    • Provided a corrected method for calculating distributions of positive predictive values, addressing a previously identified error.

    Conclusions:

    • The management of uncertainty in diagnostic testing can be simplified by using expected parameter values in Bayesian calculations.
    • This approach ensures accurate estimation of positive predictive values, crucial for clinical decision-making.
    • The study offers a more robust method for assessing diagnostic test performance under uncertainty.