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

Adjusting sample size for anticipated dropouts in clinical trials

J E Overall1, G Shobaki, C Shivakumar

  • 1Department of Psychiatry and Behavioral Science, University of Texas Medical School, Houston 77030, USA.

Psychopharmacology Bulletin
|May 2, 1998
PubMed
Summary

To maintain statistical power in clinical trials, researchers should increase sample sizes to account for anticipated dropouts. This strategy supports intent-to-treat analyses by ensuring sufficient data despite participant attrition.

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

Tabular foundation model predicts alternative lengthening of telomeres (ALT) and identifies SMARCAL1 as a target in ALT-driven cancers.

bioRxiv : the preprint server for biology·2026
Same author

Hybrid delivery of cluster-set resistance training for individuals previously treated for lung cancer: the results of a single-arm feasibility trial.

Pilot and feasibility studies·2023
Same author

Study protocol: investigating the feasibility of a hybrid delivery of home-based cluster set resistance training for individuals previously treated for lung cancer.

Pilot and feasibility studies·2022
Same author

Determining the appropriate use of Technology Enabled Care Services (TECS) to manage upper-limb trauma injuries during the COVID-19 pandemic: A multicentre retrospective observational study.

Journal of plastic, reconstructive & aesthetic surgery : JPRAS·2022
Same author

Where next for the design, delivery, and evaluation of community-based physical activity prescription? Emerging lessons from the United Kingdom.

Applied physiology, nutrition, and metabolism = Physiologie appliquee, nutrition et metabolisme·2021
Same author

The impact of signposting and group support pathways on a community-based physical activity intervention grounded in motivational interviewing.

Journal of public health (Oxford, England)·2021

Area of Science:

  • Biostatistics
  • Clinical Trial Design
  • Statistical Modeling

Background:

  • Sample size calculations for clinical trials often neglect dropout impacts on intent-to-treat (ITT) analysis power.
  • Dropouts can significantly reduce the statistical power of controlled clinical trials.

Purpose of the Study:

  • To propose empirical dropout correction coefficients for adjusting sample sizes in ANOVA and ANCOVA.
  • To evaluate the impact of different analyses, data structures, and dropout rates on sample size adjustments.

Main Methods:

  • Developed and applied empirical dropout correction coefficients to sample size calculations.
  • Compared Analysis of Variance (ANOVA) and Analysis of Covariance (ANCOVA) under various conditions.
  • Considered compound symmetry and autoregressive correlational structures for repeated measurements.

Related Experiment Videos

  • Evaluated dropout rates of 20% and 30%.

    • Main Results:

    • Analysis of Covariance (ANCOVA) is recommended for correcting baseline and time-related differences.
    • With an autoregressive structure and ANCOVA, increasing sample size by the expected number of dropouts is supported.
    • This strategy maintains the power of ITT analyses, including those with incomplete data.

    Conclusions:

    • The common practice of inflating sample size for anticipated dropouts is validated for ITT analyses.
    • The proposed method is effective for maintaining statistical power in various analytical approaches, including mixed-model regression.
    • Empirical correction coefficients provide a robust method for sample size adjustments in the presence of dropouts.