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

Analyzing excessive no changes in clinical trials with clustered data.

Shou-En Lu1, Yong Lin, Wei-Chung Joe Shih

  • 1Division of Biometrics, University of Medicine and Dentistry of New Jersey, School of Public Health, New Brunswick, New Jersey 08903, USA.

Biometrics
|March 23, 2004
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

Problem-solving therapy versus supportive psychotherapy for Veterans with moderate suicide risk and chronic pain: A pilot randomized clinical trial.

Behaviour research and therapy·2026
Same author

Per- and Polyfluoroalkyl Substances (PFAS) Exposure Profiles and Their Predictors in a Study of US Volunteer Firefighters.

American journal of industrial medicine·2026
Same author

A cross-sectional analysis of low-cost non-cancer registry recruitment sources for a hybrid type 2 trial evaluating a digital intervention for melanoma survivors.

mHealth·2026
Same author

Actigraphy-based differences in sleep in those receiving opioid versus non-opioid therapy postsurgical removal of impacted third molars: a pilot study from the opioid analgesic reduction study (OARS).

Sleep medicine·2026
Same author

Illness perceptions and behavioural responses as mechanisms of change in problem-solving treatment for Veterans with Gulf War Illness.

British journal of health psychology·2026
Same author

Analgesic Differences in Males and Females After Third Molar Surgery: A Subgroup Analysis of the OARS Randomized Clinical Trial.

JAMA network open·2025

This study introduces a novel mixture model for clinical trial data with many zero measurements, improving analysis of clustered efficacy measures. The new model accounts for measurement errors and offers a more sensitive approach than traditional methods.

Area of Science:

  • Biostatistics
  • Clinical Trials
  • Statistical Modeling

Background:

  • Clinical trials often collect efficacy measures from multiple sites per patient (e.g., tooth sites, spine sites).
  • Conventional analysis using patient averages can dilute results due to a high proportion of zero or unchanged measurements.
  • Existing methods like the two-part model are suitable for independent data but may not fully address clustered data with excessive zeros.

Purpose of the Study:

  • To develop and evaluate a statistical model for analyzing clustered clinical trial data characterized by excessive zero measurements.
  • To improve the sensitivity and accuracy of efficacy measure analysis in such settings.
  • To incorporate potential measurement errors into the modeling framework.

Main Methods:

Related Experiment Videos

  • Utilized a mixture of distributions model to handle clustered data with a high prevalence of zero values.
  • Incorporated considerations for potential measurement errors within the statistical framework.
  • Demonstrated that the proposed mixture model encompasses the two-part model as a special case.

    • Main Results:

    • The proposed mixture model effectively addresses the challenge of excessive zeros in clustered efficacy data.
    • The model provides a more sensitive analysis compared to conventional per-patient averaging methods.
    • The inclusion of measurement error enhances the robustness of the findings.

    Conclusions:

    • A mixture distribution model offers a superior approach for analyzing clustered clinical trial data with many zero efficacy measures.
    • This method provides a more nuanced understanding of treatment effects by accounting for data structure and measurement variability.
    • The proposed model advances statistical methodologies for complex clinical trial designs.