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

Clinical utility of likelihood ratios

E J Gallagher1

  • 1Department of Emergency Medicine, Albert Einstein College of Medicine, Bronx, NY 10467, USA. jgallagh@montefiore.org

Annals of Emergency Medicine
|March 20, 1998
PubMed
Summary

Likelihood ratios (LRs) offer a superior method for evaluating diagnostic test performance compared to sensitivity and specificity. LRs provide stable, clinically relevant measures for updating disease probability.

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

Novel breast cancer treatment leads to hyperglycaemia.

Diabetic medicine : a journal of the British Diabetic Association·2020
Same author

Elevated tumor LDLR expression accelerates LDL cholesterol-mediated breast cancer growth in mouse models of hyperlipidemia.

Oncogene·2017
Same author

Silencing vimentin expression decreases pulmonary metastases in a pre-diabetic mouse model of mammary tumor progression.

Oncogene·2016
Same author

The phosphorylated prodrug FTY720 is a histone deacetylase inhibitor that reactivates ERα expression and enhances hormonal therapy for breast cancer.

Oncogenesis·2015
Same author

Mammary tumor growth and pulmonary metastasis are enhanced in a hyperlipidemic mouse model.

Oncogene·2012
Same author

Resolution of insulin-requiring diabetes in a liver transplant recipient after treatment of a pheochromocytoma: case report and review of literature.

Minerva endocrinologica·2012

Area of Science:

  • Medical Diagnostics
  • Biostatistics
  • Clinical Epidemiology

Background:

  • Diagnostic test performance is often evaluated using sensitivity and specificity derived from 2x2 tables.
  • Receiver operating characteristic (ROC) curves summarize the sensitivity-specificity relationship but do not directly inform clinical decisions.
  • Predictive values offer clinically relevant information but are unstable due to disease prevalence variations.

Purpose of the Study:

  • To highlight the limitations of traditional test-performance metrics like sensitivity and specificity.
  • To introduce likelihood ratios (LRs) as a more robust and clinically applicable measure of diagnostic test accuracy.
  • To demonstrate how LRs, combined with Bayes' theorem, facilitate the revision of disease probability.

Main Methods:

  • Analysis of traditional diagnostic test performance metrics (sensitivity, specificity, predictive values).
  • Introduction and explanation of likelihood ratios (LRs) as a composite index of test performance.
  • Application of Bayes' theorem using LRs to update pretest disease odds to posttest odds.

Main Results:

  • Sensitivity and specificity, while stable, invert clinical logic and are not directly interpretable for individual patient probabilities.
  • Predictive values are interpretable but highly dependent on disease prevalence, making them unstable.
  • Likelihood ratios (LRs) offer stable, interpretable measures that directly inform probability revision.

Conclusions:

  • Likelihood ratios are a more effective tool for assessing diagnostic test performance than sensitivity and specificity alone.
  • The equation 'Pretest odds x LR = Posttest odds' demonstrates the utility of LRs in updating disease probability, aligning with clinical diagnostic strategies.
  • LRs provide a stable and mathematically sound method for integrating test results into clinical decision-making.

Related Experiment Videos