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

Multilevel analysis of group-randomized trials with binary outcomes.

Hae-Young Kim1, John S Preisser, R Gary Rozier

  • 1Dental Research Institute, School of Dentistry, Seoul National University, Seoul, Korea.

Community Dentistry and Oral Epidemiology
|July 22, 2006
PubMed
Summary

Ignoring the intra-class correlation (ICC) in group-randomized trials (GRTs) overstates intervention effectiveness. Proper analysis of cluster trials requires accounting for ICC to avoid erroneous conclusions in public health research.

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

Evaluating the impact of conditional micro-incentives on linkage to HIV care in rural South Africa: Results from the Home-Based Intervention to Test and Start (HITS) trial.

Health policy and planning·2026
Same author

Machine Learning-Based Prediction of Dental Caries in 12-Year-Old Adolescents: A Nationwide Cross-Sectional Study in Korea.

International journal of dental hygiene·2026
Same author

Cost-Effectiveness of Differentiated Service Delivery for HIV Treatment: A Combined Mathematical Modeling Study of Four African Settings.

Open forum infectious diseases·2026
Same author

Digital-Human Public Community Care Integration for Chronic Pain in Low-Income Older Adults in a 6-Week Living Lab Setting: Quasi-Experimental Feasibility Study.

JMIR aging·2026
Same author

Treatment of increased intracranial pressure secondary to otitic hydrocephalus.

Frontiers in pediatrics·2026
Same author

How do social network models compare to all-to-all models for forecasting tuberculosis epidemics? A mathematical modeling study.

PloS one·2026

Area of Science:

  • Dental Public Health
  • Biostatistics
  • Epidemiology

Background:

  • Community-based interventions, like water fluoridation, are common in dental research.
  • Cluster trials, including group-randomized trials (GRTs), require specialized design and analysis compared to individual randomized controlled trials (RCTs).

Purpose of the Study:

  • To review analytical methods for binary outcomes in cluster trials.
  • To illustrate these methods using a school-based GRT in dental research.
  • To highlight the impact of accounting for the intra-class correlation (ICC) in data analysis.

Main Methods:

  • Examined GRT characteristics, focusing on ICC.
  • Reviewed analytical methods: group-level analysis, adjusted chi-square, and multivariable models (mixed-effects, GEE) for correlated binary data.

Related Experiment Videos

  • Applied methods to a school-based dental GRT, comparing traditional analyses with those accounting for ICC.

    • Main Results:

    • Traditional analysis methods applied to the dental GRT significantly overstated intervention effectiveness.
    • Failure to account for ICC led to inflated estimates of treatment effects.

    Conclusions:

    • Ignoring ICC in group-assignment studies can lead to incorrect conclusions.
    • Specialized analytical approaches are crucial for cluster trials in dental public health.
    • Group-level randomization in trial design warrants greater consideration.