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

Variable selection and Bayesian model averaging in case-control studies.

V Viallefont1, A E Raftery, S Richardson

  • 1INSERM-U. 170, 16 av P. Vaillant-Couturier, 94 807 Villejuif Cedex, France.

Statistics in Medicine
|December 18, 2001
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

Clinical clusters during acute illness predict long-term mortality in older patients.

BMC medicine·2025
Same author

Cationic Polystyrene Latex Nanocarriers for Immunostimulatory Long Double-Stranded RNA Delivery to Ovarian Cancer Cells.

Journal of biomedical materials research. Part B, Applied biomaterials·2024
Same author

BIC extensions for order-constrained model selection.

Sociological methods & research·2022
Same author

Accelerometer-based estimation of oxygen uptake in adults with Down syndrome: vector magnitude vs. vertical axis.

Journal of intellectual disability research : JIDR·2022
Same author

Statistical properties of sketching algorithms.

Biometrika·2022
Same author

The molecular landscape and associated clinical experience in infant medulloblastoma: prognostic significance of second-generation subtypes.

Neuropathology and applied neurobiology·2020
Same journal

Interpretable Bayesian Modeling for Multireader Multicase Studies: Addressing Overdispersion and Limited Sample Size in Diagnostic Enhancement Evaluation.

Statistics in medicine·2026
Same journal

Adaptive Sequential Multiple Hypotheses Testing for Concomitant Vaccine Safety Surveillance.

Statistics in medicine·2026
Same journal

Novel Distance Regression for Repeated Outcomes With Missing Data: Applications to Longitudinal and Crossover Studies of Microbiome Beta-Diversity.

Statistics in medicine·2026
Same journal

Optimal Weighted Tests for Replication Studies and the 'Two-Trials Rule' With Multiple Hypotheses.

Statistics in medicine·2026
Same journal

Identifiable Copula-Double-Cox Models: A Fully Parametric Framework for Dependent Right-Censored Survival Data.

Statistics in medicine·2026
Same journal

Moving From Individualized Risk-Based Prevention to Benefit-Based Prevention: Estimating Individualized Life-Years Gained From Prevention Services as a Basis for Eligibility.

Statistics in medicine·2026
See all related articles

Statistical variable selection in case-control studies overstates findings. Bayesian model averaging offers a more accurate assessment of risk factors by accounting for model uncertainty, improving reliability in epidemiological research.

Area of Science:

  • Epidemiology
  • Biostatistics
  • Statistical Modeling

Background:

  • Case-control studies frequently use statistical variable selection for covariate and confounder identification.
  • Traditional methods like stepwise logistic regression ignore model uncertainty, potentially underestimating risks.

Purpose of the Study:

  • To evaluate the impact of variable selection on statistical inference in case-control studies.
  • To introduce and assess Bayesian model averaging (BMA) as a method to address model uncertainty.

Main Methods:

  • A simulation study mimicking actual case-control studies was conducted.
  • Compared p-values from stepwise selection with results from Bayesian model averaging.
  • Applied and compared methods using a case-control study of cervical cancer.

Related Experiment Videos

Main Results:

  • Variable selection methods significantly overstate the strength of evidence; only 49% of variables declared significant (p<0.05) were true risk factors in simulations.
  • Bayesian model averaging provides a posterior probability of a variable being a risk factor.
  • BMA demonstrated reasonable calibration in simulated scenarios.

Conclusions:

  • Standard variable selection in case-control studies inflates statistical significance.
  • Bayesian model averaging is a robust approach to incorporate model uncertainty.
  • BMA offers a more reliable method for identifying risk factors in epidemiological research.