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

Improved odds ratio estimation by post hoc stratification of case-control data

M Neuhäuser1, H Becher

  • 1Solvay Pharma, Hannover, Germany.

Statistics in Medicine
|May 15, 1997
PubMed
Summary

This study introduces a logistic regression model for case-control studies, improving parameter estimation and reducing bias. Post hoc stratification offers a refined analysis for epidemiological research.

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

Uncovering the molecular signature of feline diffuse iris melanoma through transcriptomic analysis of disease severity.

Scientific reports·2025
Same author

Physical activity in adults with and without rheumatoid arthritis: cross-sectional results from the Survey of Health, Ageing and Retirement in Europe (SHARE).

Scandinavian journal of rheumatology·2023
Same author

A feasibility study of perioperative vitamin D supplementation in patients undergoing colorectal cancer resection.

Frontiers in nutrition·2023
Same author

Tired of feeling tired - The role of circulating inflammatory biomarkers and long-term cancer related fatigue in breast cancer survivors.

Breast (Edinburgh, Scotland)·2021
Same author

Validation of inflammatory genetic variants associated with long-term cancer related fatigue in a large breast cancer cohort.

Brain, behavior, and immunity·2018
Same author

Non-specific effects of childhood vaccinations - A case control study nested into a Health and Demographic Surveillance System in rural Burkina Faso.

Vaccine·2017

Area of Science:

  • Epidemiology
  • Biostatistics

Background:

  • Case-control studies are crucial in epidemiology for investigating disease risk factors.
  • Unmatched and frequency-matched designs present analytical challenges, particularly regarding parameter estimation and confounding.

Purpose of the Study:

  • To propose a logistic regression analysis for unmatched or frequency-matched case-control studies.
  • To enhance parameter estimation accuracy and reduce bias using conditional maximum likelihood estimation via post hoc stratification.
  • To quantify residual confounding effects within this analytical framework.

Main Methods:

  • Development of a logistic regression model incorporating conditional maximum likelihood estimation.
  • Implementation of post hoc stratification as a method to refine the analysis.

Related Experiment Videos

  • Conducting a simulation study to evaluate parameter estimates' variance and bias.
  • Application of the model to real-world case-control data for laryngeal, esophageal, and lung cancers.

    • Main Results:

    • The proposed model demonstrates smaller variance and reduced bias in parameter estimates compared to standard methods.
    • Conditional maximum likelihood estimation through post hoc stratification effectively addresses analytical challenges.
    • Residual confounding effects were successfully quantified.
    • A trade-off exists: more refined stratification reduces computation but may increase bias and reduce efficiency.

    Conclusions:

    • The proposed logistic regression analysis with conditional maximum likelihood estimation and post hoc stratification offers an improved approach for case-control studies.
    • This method enhances the reliability of parameter estimates and provides insights into residual confounding.
    • The findings have implications for epidemiological research, particularly in cancer studies.