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

Exposure-dependent misclassification of exposure in interaction analyses.

M Lundberg1, J Hallqvist, F Diderichsen

  • 1Department of Public Health Sciences, Karolinska Institutet, Stockholm, Sweden.

Epidemiology (Cambridge, Mass.)
|September 1, 1999
PubMed
Summary

Exposure misclassification can bias interaction analysis, potentially leading to underestimation of effects. Lower exposure prevalence reduces the magnitude of this bias, impacting epidemiological research findings.

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

Correspondence: The First Biosimilar for Biologics in Allergy: CT-P39 (Omlyclo-Omalizumab-Igec) Is Available for Asthma, Chronic Spontaneous Urticaria, and Chronic Rhinosinusitis With Nasal Polyps.

Allergy·2026
Same author

The Unmet Need of Olfactory Testing in Inflammatory Disorders of the Upper Airways-An EAACI Position Paper.

Allergy·2026
Same author

Person-centred care in the context of higher education - a discourse analysis based on interviews with programme directors.

BMC medical education·2024
Same author

Adolescents' experiences of Help Overcoming Pain Early-A school based person-centred intervention for adolescents with chronic pain.

Paediatric & neonatal pain·2023
Same author

Unmet challenges in septoplasty-nordic studies from a uniform healthcare and geographical area.

Frontiers in surgery·2022
Same author

Person-centred care on the move - an interview study with programme directors in Swedish higher education.

BMC medical education·2022

Area of Science:

  • Epidemiology
  • Biostatistics
  • Statistical modeling

Background:

  • Exposure misclassification is a common issue in epidemiological studies.
  • Accurate estimation of effect modification (interaction) is crucial for understanding disease etiology.
  • The impact of exposure misclassification on interaction estimates requires careful consideration.

Purpose of the Study:

  • To analyze the consequences of exposure misclassification on effect estimates in interaction analysis.
  • To develop a mathematical equation for the potentially biased estimate.
  • To identify conditions leading to overestimation or underestimation of interaction effects due to misclassification.

Main Methods:

  • Theoretical analysis of exposure misclassification dependent on a second exposure but independent of outcome.

Related Experiment Videos

  • Development of a mathematical framework to quantify bias in interaction effect estimates.
  • Evaluation of factors influencing the direction and magnitude of bias.

    • Main Results:

    • Exposure misclassification can theoretically lead to overestimation of interaction effects.
    • Underestimation of interaction effects is more likely due to the nature of multidimensional misclassification.
    • Bias magnitude is inversely related to the prevalence of the misclassified exposure, stratified by the second exposure and outcome.

    Conclusions:

    • Exposure misclassification can significantly bias interaction effect estimates in epidemiological studies.
    • The direction of bias (over- or underestimation) depends on specific misclassification patterns.
    • Lower exposure prevalence may mitigate the impact of misclassification bias on interaction analyses.