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

Intraclass correlation coefficient and outcome prevalence are associated in clustered binary data.

M C Gulliford1, G Adams, O C Ukoumunne

  • 1Department of Public Health Sciences, King's College London, Capital House, 42 Weston St, London SE1 3QD, UK. martin.gulliford@kcl.ac.uk

Journal of Clinical Epidemiology
|February 19, 2005
PubMed
Summary

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

Chemoradiotherapy versus radiotherapy in patients with advanced nasopharyngeal cancer: phase III randomized Intergroup study 0099.

Journal of clinical oncology : official journal of the American Society of Clinical Oncology·2023
Same author

Anti-inflammatory, procollagen, and wound repair properties of topical insulin gel.

Brazilian journal of medical and biological research = Revista brasileira de pesquisas medicas e biologicas·2023
Same author

83&#x2003;Production of live calves after transfer of <i>in vitro</i>-produced embryos in synchronised wood bison (<i>Bison bison athabascae</i>).

Reproduction, fertility, and development·2022
Same author

40&#x2003;Rapid-freeze of bison semen in cholesterol-based extender.

Reproduction, fertility, and development·2022
Same author

Demand for regional level III neonatal services is not reduced during national COVID lockdowns.

Early human development·2021
Same author

Well-being, physical activity and long-term conditions: cross-sectional analysis of Health Survey for England 2016.

Public health·2020

The prevalence of an outcome can help predict the intraclass correlation coefficient (ICC). This association was observed in UK general practice and health technology assessment data, aiding in informed assumptions about ICC magnitude.

Area of Science:

  • Biostatistics
  • Health Services Research
  • Epidemiology

Background:

  • The intraclass correlation coefficient (ICC) quantifies the similarity of measurements within groups.
  • Understanding the relationship between outcome prevalence and ICC is crucial for study design and interpretation.

Purpose of the Study:

  • To investigate the association between the proportion of an outcome (prevalence) and the ICC.
  • To determine if prevalence can inform predictions of ICC values.

Main Methods:

  • Analysis of 188 ICCs from the General Practice Research Database (GPRD) in the UK.
  • Analysis of 136 ICCs from a Health Technology Assessment (HTA) review across community and health services settings.
  • Statistical modeling to assess the linear association between log ICC and log prevalence.

Related Experiment Videos

Main Results:

  • A significant linear association was found between log ICC and log prevalence in both GPRD and HTA datasets.
  • Predicted ICC values varied based on prevalence; for example, at 1% prevalence, ICC was predicted as 0.008 (GPRD) or 0.002 (HTA).
  • At 40% prevalence, predicted ICC increased to 0.075 (GPRD) or 0.046 (HTA).

Conclusions:

  • Outcome prevalence is a significant factor influencing ICC magnitude.
  • Prevalence can be used to make informed estimations of ICC in similar settings.
  • This finding has implications for sample size calculations and the interpretation of variability in health research.