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

Bringing context back into epidemiology: variables and fallacies in multilevel analysis

A V Diez-Roux1

  • 1Division of General Medicine, College of Physicians and Surgeons, School of Public Health, Columbia University, New York, USA.

American Journal of Public Health
|March 10, 1998
PubMed
Summary

Epidemiologic research often overlooks group factors. This study advocates for multilevel analysis, integrating individual and group variables to better understand population health and disease patterns.

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

Dietary intake, plasma homocysteine, and repetitive element DNA methylation in the Multi-Ethnic Study of Atherosclerosis (MESA).

Nutrition, metabolism, and cardiovascular diseases : NMCD·2014
Same author

Cumulative life course and adult socioeconomic status and markers of inflammation in adulthood.

Journal of epidemiology and community health·2008
Same author

Airborne particulate matter exposure and urinary albumin excretion: the Multi-Ethnic Study of Atherosclerosis.

Occupational and environmental medicine·2007
Same author

Neighborhood poverty and the resurgence of tuberculosis in New York City, 1984-1992.

American journal of public health·2001
Same author

Area characteristics and individual-level socioeconomic position indicators in three population-based epidemiologic studies.

Annals of epidemiology·2001
Same author

Commentary: causes of incidence and causes of cases--a Durkheimian perspective on Rose.

International journal of epidemiology·2001

Area of Science:

  • Epidemiology
  • Social Epidemiology
  • Public Health

Background:

  • Current epidemiologic research often relies on methodologic individualism, focusing solely on individual characteristics to explain population health.
  • This approach may limit understanding of complex health and disease distributions within communities.

Purpose of the Study:

  • To highlight the necessity of incorporating group- or macro-level variables into epidemiologic studies.
  • To advocate for multilevel analyses that consider multiple determinants of health outcomes.

Main Methods:

  • Discusses the conceptual framework for contextual or multilevel analyses in epidemiology.
  • Addresses methodological challenges, including selection of contextual units and variables, and model specification.

Related Experiment Videos

Main Results:

  • Multilevel analysis integrates individual and group-level factors to explain health and disease.
  • It necessitates developing theoretical models that span different levels of determination.

Conclusions:

  • Multilevel analysis offers a more comprehensive approach to understanding population health.
  • It emphasizes the significant role of macro-level factors in shaping health and disease patterns.