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

General relative risk regression models for epidemiologic studies.

S H Moolgavkar1, D J Venzon

  • 1Fred Hutchinson Cancer Research Center, Division of Public Health Sciences, Seattle, WA 98104.

American Journal of Epidemiology
|November 1, 1987
PubMed
Summary

For case-control studies, only one relative risk function ensures analysis is independent of binary covariate coding. Non-multiplicative models require caution, favoring likelihood-based methods for accurate inference.

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

Systems biological and mechanistic modelling of radiation-induced cancer.

Radiation and environmental biophysics·2007
Same author

Consideration of copollutants.

Science (New York, N.Y.)·2007
Same author

A generalized Luria-Delbrück model.

Mathematical biosciences·2005
Same author

From mechanisms to risk estimation--bridging the chasm.

Advances in space research : the official journal of the Committee on Space Research (COSPAR)·2005
Same author

The use of biologically based cancer risk models in radiation epidemiology.

Radiation protection dosimetry·2003
Same author

Air pollution and hospital admissions for chronic obstructive pulmonary disease in three metropolitan areas in the United States.

Inhalation toxicology·2003

Area of Science:

  • Epidemiology
  • Biostatistics

Background:

  • Case-control studies are crucial for investigating disease etiology.
  • Relative risk functions are key tools in analyzing epidemiological data.
  • Inference methods must be robust to covariate coding and model assumptions.

Purpose of the Study:

  • To evaluate three parametric families of relative risk functions for case-control data.
  • To identify relative risk functions with inference independent of binary covariate coding.
  • To highlight potential issues with asymptotic covariance matrix methods in non-multiplicative models.

Main Methods:

  • Discussion of three parametric families of relative risk functions.
  • Assessment of covariate coding independence for each family.
  • Illustration of inference issues using examples.

Related Experiment Videos

  • Comparison of asymptotic covariance matrix methods versus likelihood-based procedures.

    • Main Results:

    • Only one of the three discussed relative risk function families ensures inference is independent of binary covariate coding.
    • Inference based on the asymptotic covariance matrix can be misleading for non-multiplicative relative risk functions, especially with smaller sample sizes.
    • Likelihood-based procedures are recommended for analyses involving non-multiplicative relative risk functions.

    Conclusions:

    • The choice of relative risk function in case-control studies significantly impacts inference validity.
    • Researchers should prioritize relative risk functions that are invariant to covariate coding.
    • Likelihood-based inference is essential for robust analysis when non-multiplicative relative risks are employed.