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

Bayesian estimation of intervention effect with pre- and post-misclassified binomial data.

James D Stamey1, John W Seaman, Dean M Young

  • 1Department of Statistical Science, Baylor University, Waco, Texas, USA. James_Stamey@baylor.edu

Journal of Biopharmaceutical Statistics
|January 16, 2007
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

Bayesian variable selection for logistic regression with a differentially misclassified binary covariate.

Communications in statistics: Simulation and computation·2025
Same author

Bayesian inference for prediction of survival probability in prime-boost vaccination regimes.

Statistics in medicine·2023
Same author

Stability analysis using mixed models: A critique of tolerance interval methods and a probabilistic solution.

Pharmaceutical statistics·2023
Same author

Topical Anti-TNFα Agent Licaminlimab (OCS-02) Relieves Persistent Ocular Discomfort in Severe Dry Eye Disease: A Randomized Phase II Study.

Clinical ophthalmology (Auckland, N.Z.)·2022
Same author

Topical Ocular Anti-TNFα Agent Licaminlimab in the Treatment of Acute Anterior Uveitis: A Randomized Phase II Pilot Study.

Translational vision science & technology·2022
Same author

A Bayesian Hierarchical Spatial Model to Correct for Misreporting in Count Data: Application to State-Level COVID-19 Data in the United States.

International journal of environmental research and public health·2022

This study introduces a Bayesian approach to address diagnostic test biases from regression to the mean. The Bayesian method outperforms traditional likelihood methods, especially for highly accurate diagnostic tests.

Area of Science:

  • Biostatistics
  • Medical Diagnostics
  • Epidemiology

Background:

  • Diagnostic tests can be fallible, leading to misclassification and biased intervention effect estimates.
  • Regression to the mean, caused by imperfect test accuracy, complicates the interpretation of intervention outcomes.
  • Existing likelihood methods struggle with highly accurate tests (near 100% sensitivity or specificity) and often require external validation data.

Purpose of the Study:

  • To develop a robust statistical approach for analyzing intervention effects when using fallible diagnostic tests.
  • To overcome the limitations of existing likelihood methods, particularly in scenarios with high test accuracy.
  • To propose a Bayesian framework that mitigates bias from misclassification and regression to the mean.

Main Methods:

Related Experiment Videos

  • A novel Bayesian approach is proposed to model disease status and intervention effects.
  • The method is compared against the maximum likelihood estimator (MLE) using simulation studies.
  • The Bayesian approach accommodates multiple information sources and can function without validation data.

Main Results:

  • The proposed Bayesian approach demonstrates superior performance compared to the MLE method.
  • Performance improvements are particularly significant when diagnostic tests exhibit high sensitivity or specificity.
  • A real-world data example with near-perfect specificity illustrates the method's practical utility.

Conclusions:

  • The Bayesian approach provides a more accurate and flexible alternative for analyzing intervention effects with fallible diagnostic tests.
  • This method effectively addresses biases introduced by test misclassification and regression to the mean.
  • The findings are especially relevant for studies employing highly accurate diagnostic tools in medical research.