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

A median-based test under informative dropout: the one-sample case

W M Liang1, M B Brown

  • 1China Medical College, Taiwan.

Controlled Clinical Trials
|October 7, 1997
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

Perspectives on enhancing human performance in the heat: Is the solution to simply "just add water"?

Sports medicine and health science·2025
Same author

Exploring the Effects of Giraffe Skin Disease Limb Lesions on Locomotion.

Ecology and evolution·2025
Same author

Mechanistic insight into heat enhanced permeation of diclofenac and piroxicam in combination with chemical penetration enhancers across skin.

European journal of pharmaceutical sciences : official journal of the European Federation for Pharmaceutical Sciences·2024
Same author

Evaluating the effects of giraffe skin disease and wire snare wounds on the gaits of free-ranging Nubian giraffe.

Scientific reports·2023
Same author

The effect of dilution of fusidic acid cream and betamethasone dipropionate cream in complex extemporaneous mixes on formulation performance.

International journal of pharmaceutics·2022
Same author

Correction to: Enhanced Follicular Delivery of Finasteride to Human Scalp Skin Using Heat and Chemical Penetration Enhancers.

Pharmaceutical research·2020
Same journal

On the generation and ownership of alpha in medical studies.

Controlled clinical trials·2004
Same journal

An analysis of the effect of funding source in randomized clinical trials of second generation antipsychotics for the treatment of schizophrenia.

Controlled clinical trials·2004
Same journal

Symptom recording in a randomised clinical trial: paper diaries vs. electronic or telephone data capture.

Controlled clinical trials·2004
Same journal

Statistical comparison of random allocation methods in cancer clinical trials.

Controlled clinical trials·2004
Same journal

Analyzing bronchodilation with emphasis on disease type, age and sex.

Controlled clinical trials·2004
Same journal

Geographic variability in patient characteristics, treatment and outcome in an International Trial of Magnesium in acute myocardial infarction.

Controlled clinical trials·2004
See all related articles

This study introduces a new median-based statistic for clinical trials with dropouts who have poorer efficacy. The proposed method offers a more reliable significance test than the standard paired t-test when dropouts are not random.

Area of Science:

  • Biostatistics
  • Clinical Trial Design
  • Medical Research Methodology

Background:

  • Clinical trials often involve continuous outcomes and patient dropouts.
  • Dropouts may exhibit different efficacy compared to completers, potentially biasing results.
  • Ignoring dropout data or using standard methods like the paired t-test can lead to misleading conclusions when dropouts are not random.

Purpose of the Study:

  • To develop a robust statistical method for analyzing continuous outcomes in clinical trials with non-random dropouts.
  • To propose a median-based statistic that accounts for dropouts with poorer efficacy.
  • To compare the performance of the proposed statistic against the paired t-test under various dropout scenarios.

Main Methods:

  • Assumed Gaussian distribution for efficacy measure (change in outcome).

Related Experiment Videos

  • Assumed dropouts have worse efficacy than the median.
  • Proposed a median-based t-like statistic using the sample median and estimating variance from the complete half-sample.
  • Conducted simulations to compare the proposed statistic with the paired t-test.

    • Main Results:

    • The median-based statistic provides a conservative and reliable bound for testing treatment significance.
    • The paired t-test fails to maintain its significance level with non-random dropouts, except in uniform dropout cases.
    • Simulations demonstrated the superiority of the median-based approach in scenarios with dropouts exhibiting poorer efficacy.

    Conclusions:

    • The proposed median-based statistic is a more appropriate tool for clinical trials with dropouts who have poorer efficacy.
    • The paired t-test should be avoided in such trials due to its unreliability.
    • This method enhances the integrity of statistical analysis in clinical trials with incomplete data.