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

Integrating stratum-specific likelihood ratios with the analysis of ROC curves

J C Peirce1, R G Cornell

  • 1Department of Medical Education and Research, Good Samaritan Regional Medical Center, Phoenix, Arizona 85006.

Medical Decision Making : an International Journal of the Society for Medical Decision Making
|April 1, 1993
PubMed
Summary
This summary is machine-generated.

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

Using the free-to-total prostate-specific antigen ratio to detect prostate cancer in men with nonspecific elevations of prostate-specific antigen levels.

Journal of general internal medicine·2000
Same author

The paradox of physicians and administrators in health care organizations.

Health care management review·2000
Same author

Response to the commentaries of Kaluzny, Arndt, and Bigelow: personal reflections on health care organization as viewed through the lens of complexity theory.

Health care management review·2000
Same author

Talking about treatment.

Annals of internal medicine·2000
Same author

Free PSA for detecting prostate cancer.

JAMA·1998
Same author

Understanding medical systems.

Annals of internal medicine·1998

Adding stratum-specific likelihood ratios (SSLRs) to receiver operating characteristic (ROC) curve analysis provides valuable diagnostic information. This method quantifies spectrum bias and improves the interpretation of diagnostic test results.

Area of Science:

  • Medical diagnostics
  • Biostatistics
  • Clinical epidemiology

Background:

  • Receiver operating characteristic (ROC) curves and ROC area under the curve (AUC) are standard for evaluating diagnostic tests.
  • Stratum-specific likelihood ratios (SSLRs) can be derived from ROC data, but their utility alongside ROC analysis is not fully established.

Purpose of the Study:

  • To determine if incorporating SSLRs into ROC curve analysis yields additional useful information.
  • To assess the ability of SSLRs to quantify spectrum bias and improve diagnostic discrimination.

Main Methods:

  • Analysis of four diverse datasets using ROC curves and derived SSLRs with 95% confidence intervals.
  • Evaluation of diagnostic performance for acute myocardial infarction (AMI) using serum creatine kinase (SCK), strep throat diagnosis, and computed tomographic (CT) image assessment.

Related Experiment Videos

Main Results:

  • SCK for AMI in the coronary care unit (CCU) yielded highly discriminating strata, while SCK for chest pain in the emergency center (EC) showed limited discrimination.
  • SSLR analysis effectively quantified spectrum bias by identifying cutpoints higher than reported normal values.
  • The number of strata for strep throat diagnosis was reduced, highlighting potential limitations with smaller sample sizes or data pooling.

Conclusions:

  • Deriving SSLRs as part of ROC curve analysis provides additional, valuable information for diagnostic test evaluation.
  • SSLRs enhance the interpretation of diagnostic accuracy and aid in understanding test performance across different patient populations or disease spectra.