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

Power determination for geographically clustered data using generalized estimating equations

S A Hendricks1, J T Wassell, J W Collins

  • 1Center for Disease Control and Prevention, Division of Safety Research, Morgantown, WV 26505-2888, USA.

Statistics in Medicine
|September 15, 1996
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

Deep learning prediction of error and skill in robotic prostatectomy suturing.

Surgical endoscopy·2024
Same author

New Mixture.

The Southern medical record·2022
Same author

Recommendations for a standardised educational program in robot assisted gynaecological surgery: Consensus from the Society of European Robotic Gynaecological Surgery (SERGS).

Facts, views & vision in ObGyn·2019
Same author

p16<sup>INK4a</sup> , a Senescence Marker, Influences Tenofovir/Emtricitabine Metabolite Disposition in HIV-Infected Subjects.

CPT: pharmacometrics & systems pharmacology·2016
Same author

Tips and tricks for intracorporeal robot-assisted urinary diversion.

Current urology reports·2014
Same author

Lactulose and Lactobacillus plantarum, a potential complementary synbiotic to control postweaning colibacillosis in piglets.

Applied and environmental microbiology·2014
Same journal

Interpretable Bayesian Modeling for Multireader Multicase Studies: Addressing Overdispersion and Limited Sample Size in Diagnostic Enhancement Evaluation.

Statistics in medicine·2026
Same journal

Adaptive Sequential Multiple Hypotheses Testing for Concomitant Vaccine Safety Surveillance.

Statistics in medicine·2026
Same journal

Novel Distance Regression for Repeated Outcomes With Missing Data: Applications to Longitudinal and Crossover Studies of Microbiome Beta-Diversity.

Statistics in medicine·2026
Same journal

Optimal Weighted Tests for Replication Studies and the 'Two-Trials Rule' With Multiple Hypotheses.

Statistics in medicine·2026
Same journal

Identifiable Copula-Double-Cox Models: A Fully Parametric Framework for Dependent Right-Censored Survival Data.

Statistics in medicine·2026
Same journal

Moving From Individualized Risk-Based Prevention to Benefit-Based Prevention: Estimating Individualized Life-Years Gained From Prevention Services as a Basis for Eligibility.

Statistics in medicine·2026
See all related articles

Accounting for intracluster correlation is crucial in public health studies with clustered data. Failure to do so inflates Type I errors and overestimates statistical power, impacting intervention effect estimations.

Area of Science:

  • Public Health Research
  • Biostatistics
  • Occupational Health

Background:

  • Cluster-randomized trials are common in public health, involving interventions applied to groups rather than individuals.
  • Dichotomous outcomes in such designs necessitate statistical methods accounting for intracluster correlation to avoid underestimating standard errors.
  • Generalized estimating equations (GEE) offer a method to handle intracluster correlation, but sample-size determination for adequate power remains challenging.

Purpose of the Study:

  • To evaluate statistical power in a Generalized Estimating Equation (GEE) model for a public health intervention study.
  • To assess the impact of intracluster correlation on power and Type I error rates in cluster-randomized trials.
  • To inform sample-size calculations for interventions targeting lower-back injuries in nursing personnel.

Related Experiment Videos

Main Methods:

  • A simulation study was conducted to calculate power within a GEE framework.
  • The model incorporated a beta distribution for baseline injury risk and Bernoulli distributions for individual nurse risk.
  • Fixed covariates and random-intercepts specific to each nursing home were utilized to account for population characteristics and clustering.

Main Results:

  • Failure to account for intracluster correlation led to overestimated power and inflated Type I error rates by up to 20%.
  • While GEE models handled present intracluster correlation, they exhibited negatively biased estimates when no correlation existed.
  • Inflated Type I error estimates were observed from the GEE method, potentially linked to biased intracluster correlation estimates.

Conclusions:

  • Accurate estimation of intracluster correlation is vital for reliable power calculations in cluster-randomized public health studies.
  • Ignoring intracluster correlation can lead to erroneous conclusions regarding intervention effectiveness.
  • The GEE method requires careful consideration and validation, especially regarding the estimation of intracluster correlation in the absence of true correlation.