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

Clusters galore: insights about environmental clusters from probability theory.

R Neutra1, S Swan, T Mack

  • 1California Department of Health Services, Emeryville 94608.

The Science of the Total Environment
|December 15, 1992
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

SWAT 116: the effectiveness of an infographic to improve trial recruitment-a randomised study within a trial (SWAT).

Trials·2026
Same author

SWAT 119: the effectiveness of a Thank You card to improve trial follow-up; a randomised study within a trial (SWAT).

Trials·2026
Same author

Gnaphalium in Sciatica.

The Homoeopathic physician·2023
Same author

Lac Felinum, Provings of.

The Homoeopathic physician·2023
Same author

Clinical Notes.

The Homoeopathic physician·2023
Same author

Hahnemnan the Founder of Isopathy: A Letter from Dr. Swan.

The Homoeopathic physician·2023

Most environmental cancer clusters are not caused by environmental carcinogens. Statistical significance alone is insufficient evidence, as chance can create false positives, requiring extensive investigation to confirm true environmental causes.

Area of Science:

  • Epidemiology
  • Environmental Health
  • Biostatistics

Background:

  • Environmental cancer clusters are often investigated for causal links to carcinogens.
  • Statistical significance is a common criterion for identifying potential clusters.
  • The interpretation of statistical significance in this context requires careful consideration of underlying probabilities.

Purpose of the Study:

  • To evaluate the posterior probability that a statistically significant environmental cancer cluster is causally linked to environmental carcinogens.
  • To assess the influence of prior probabilities, statistical test sensitivity, and specificity on this determination.
  • To provide a probabilistic framework for interpreting environmental cancer cluster findings.

Main Methods:

  • Application of Bayes' Theorem to calculate posterior probabilities.

Related Experiment Videos

  • Estimation of the prior probability of environmentally caused cancer clusters.
  • Calculation of the likelihood of false positives due to chance in cancer cluster detection.

    • Main Results:

    • The prior probability of environmentally caused cancer clusters is generally low.
    • Statistical tests for cancer clusters often have imperfect sensitivity and specificity.
    • By chance alone, many geographic areas (census tracts) exhibit statistically significant elevations in at least one cancer type out of 80.
    • Cancer registry data confirm these probabilistic predictions of false positives.

    Conclusions:

    • Most statistically significant environmental cancer clusters are likely not caused by environmental carcinogens.
    • A high proportion of identified clusters may represent false positives due to statistical artifacts and low prior probabilities.
    • Extensive investigation is required to identify a genuine environmental cause among numerous statistically significant clusters.