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

Group sequential large sample T2-like chi2 tests for multivariate observations.

John M Lachin1, Samuel W Greenhouse, Oliver M Bautista

  • 1The Biostatistics Center, The George Washington University, 6110 Executive Boulevard, Suite 750, Rockville, MD 20852, USA. jml@biostat.bsc.gwu.edu

Statistics in Medicine
|October 21, 2003
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

Differences Between Glycemia Estimates from Hemoglobin A1c and Continuous Glucose Monitoring and Their Association with Complete Blood Counts.

Diabetes technology & therapeutics·2025
Same author

Differential Longitudinal Effects of Glucose-Lowering Medications on Glucagon and C-peptide Responses in the Glycemia Reduction Approaches in Diabetes: A Comparative Effectiveness Study (GRADE).

Diabetes care·2025
Same author

Glycemia reduction in type 2 diabetes-Hypoglycemia outcomes: A randomized clinical trial.

PloS one·2024
Same author

Relationship Between Average Glucose Levels and HbA1c Differs Across Racial Groups: A Substudy of the GRADE Randomized Trial.

Diabetes care·2024
Same author

History of the Diabetes Control and Complications Trial and Its Follow-up Epidemiology of Diabetes Interventions and Complications Study: Studies That Changed the Treatment of Type 1 Diabetes.

Diabetes care·2024
Same author

Epicardial and intra-thoracic adipose tissue and cardiovascular calcifications in type 1 diabetes (T1D) in epidemiology of diabetes Interventions and Complications (EDIC): A pilot study.

American journal of preventive cardiology·2024

This study introduces a new method for calculating group sequential test boundaries for complex statistical models, improving interim analysis in clinical trials. The approach handles statistics without independent increments, enhancing the reliability of early trial results.

Area of Science:

  • Statistics
  • Biostatistics
  • Clinical Trial Methodology

Background:

  • K degree of freedom chi2 tests are common for multivariate responses and group comparisons.
  • Existing group sequential methods often require an independent increments variance-covariance structure, limiting their application.
  • Many widely used statistics, such as Liang-Zeger GEE and Wei-Lachin multivariate Wilcoxon tests, lack this structure.

Purpose of the Study:

  • To describe the computation of group sequential boundaries for interim analyses of emerging results.
  • To address the challenge of calculating boundaries for K degree of freedom tests expressed as quadratic forms in asymptotically multivariate normal statistics.
  • To provide a method applicable to statistics that do not inherently possess an independent increments variance-covariance structure.

Related Experiment Videos

Main Methods:

  • Deriving the covariance matrix elements for successive K degree of freedom chi2 statistics using quadratic form distribution theorems.
  • Estimating the covariance matrix by augmenting data from successive interim analyses.
  • Computing sequential test boundaries using methods like Slud and Wei or Lan and DeMets' alpha-spending function with surrogate information.

Main Results:

  • A method for computing group sequential boundaries for complex statistics without independent increments is presented.
  • The covariance matrix of successive K degree of freedom chi2 statistics is derived and estimated.
  • The approach was demonstrated with an example analyzing repeated cholesterol measurements in a clinical trial.

Conclusions:

  • The developed method enables robust interim analyses for a broader range of statistical models in clinical trials.
  • Accurate computation of sequential test boundaries is crucial for efficient and reliable trial decision-making.
  • This work extends the applicability of group sequential testing to commonly used, complex statistical methods.