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

Confounding in air pollution epidemiology: when does two-stage regression identify the problem?

A H Marcus1, S R Kegler

  • 1National Center for Environmental Assessment - RTP, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina, USA. marcus.allan@epa.gov

Environmental Health Perspectives
|December 19, 2001
PubMed
Summary

A novel two-stage method for assessing confounding in air pollution epidemiology may produce misleading results. This approach, designed for time-series studies of particulate matter (PM), could misinterpret direct health effects as confounding. Further validation is needed.

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

Ecological level analysis of the relationship between smoking and residential-fire mortality.

Injury prevention : journal of the International Society for Child and Adolescent Injury Prevention·2008
Same author

Experimental observations of non-Gaussian behavior and stringlike cooperative dynamics in concentrated quasi-two-dimensional colloidal liquids.

Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics·2002
Same author

Dynamics of the mitochondrial reticulum in live cells using Fourier imaging correlation spectroscopy and digital video microscopy.

Biophysical journal·2000
Same author

The conceptual structure of the integrated exposure uptake biokinetic model for lead in children.

Environmental health perspectives·1998
Same author

Some useful statistical methods for model validation.

Environmental health perspectives·1998
Same author

Nickel absorption and kinetics in human volunteers.

Proceedings of the Society for Experimental Biology and Medicine. Society for Experimental Biology and Medicine (New York, N.Y.)·1989

Area of Science:

  • Environmental epidemiology
  • Biostatistics
  • Public health

Background:

  • Time-series studies are crucial for understanding the health impacts of airborne particulate matter (PM).
  • Assessing confounding by copollutants is a significant challenge in environmental epidemiology.
  • Existing methods for confounding assessment in PM studies require careful evaluation.

Purpose of the Study:

  • To critically evaluate a recently proposed two-stage approach for assessing confounding in time-series air pollution studies.
  • To investigate the potential for misleading inferences from this method, particularly concerning particulate matter (PM) and health effects.
  • To highlight the importance of considering additional copollutants in epidemiological analyses.

Main Methods:

  • The study analyzes a two-stage regression approach applied to independent city-specific time-series data.

Related Experiment Videos

  • First stage: Fitting separate models for health effects vs. PM and confounder vs. PM within each city.
  • Second stage: Regressing city-specific slopes from the first stage to infer direct effects and confounding.

    • Main Results:

    • The proposed two-stage method's interpretation of nonzero intercepts (direct PM effect) and slopes (confounding) may be flawed.
    • A counterexample demonstrates that the presence of an additional copollutant can lead to incorrect conclusions.
    • Inferences regarding the direct pathway from PM to health effects could be overestimated or underestimated.

    Conclusions:

    • The evaluated two-stage method for confounding assessment in particulate matter (PM) epidemiology may yield unreliable results.
    • The method's assumptions can be violated by unmeasured or additional copollutants, leading to biased interpretations.
    • Researchers should exercise caution when applying this method and consider alternative or supplementary analytical strategies.