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

Biological models and statistical interactions: an example from multistage carcinogenesis

J Siemiatycki, D C Thomas

    International Journal of Epidemiology
    |December 1, 1981
    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

    A disease-associated gene desert directs macrophage inflammation through ETS2.

    Nature·2024
    Same author

    Richer than Gold: the fungal biodiversity of Reserva Los Cedros, a threatened Andean cloud forest.

    Botanical studies·2023
    Same author

    Long-term effects of Covid-19 on the kidney.

    QJM : monthly journal of the Association of Physicians·2021
    Same author

    Inter-rater reliability of occupational exposure assessment in a case-control study of female breast cancer.

    Journal of occupational and environmental hygiene·2021
    Same author

    Benefits of a Physician-Scientist-Tailored Curriculum at a Medical Student-Run Free Clinic.

    Medical science educator·2021
    Same author

    Analysis of the B cell receptor repertoire in six immune-mediated diseases.

    Nature·2019

    Statistical interaction in epidemiology does not always reflect biological interaction. Even independent risk factors can appear additive or multiplicative, showing that biological inference requires careful reasoning, not just statistical models.

    Area of Science:

    • Epidemiology
    • Biostatistics
    • Carcinogenesis

    Background:

    • Statistical interaction between risk factors is often assessed in epidemiologic studies.
    • It is tempting to infer the nature of biological interactions from statistical findings.

    Purpose of the Study:

    • To investigate the relationship between statistical and biological interactions of risk factors.
    • To demonstrate that statistical models can be misleading when inferring biological processes.

    Main Methods:

    • Utilized the multistage model of carcinogenesis as a biological framework.
    • Developed simple hypothetical examples to illustrate statistical modeling outcomes.
    • Analyzed scenarios where carcinogenic factors act independently.

    Related Experiment Videos

    Main Results:

    • Demonstrated that independently acting carcinogenic factors can exhibit various statistical interaction patterns.
    • Showed that some pairs of independent factors may fit an additive statistical model.
    • Showed that some pairs of independent factors may fit a multiplicative statistical model, while others fit neither.

    Conclusions:

    • Statistical interaction in epidemiologic data does not reliably infer biological interaction.
    • Elucidating biological interactions requires imaginative and prudent inductive and deductive reasoning.
    • Mechanical application of statistical models is insufficient for understanding biological interactions.