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 simple method for analyzing data from a randomized trial with a missing binary outcome.

Stuart G Baker1, Laurence S Freedman

  • 1Biometry Research Group, Division of Cancer Prevention, National Cancer Institute, USA. sb16i@nih.gov

BMC Medical Research Methodology
|May 8, 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

Total Sugars Misreporting in Self-Reported 24-Hour Recalls in the Study of Latinos: Nutrition and Physical Activity Assessment Study.

The Journal of nutrition·2026
Same author

Prostate cancer screening: Decision-making when MRI results are inconclusive for cancer (PI-RADS 3).

Urologic oncology·2026
Same author

Adjusting for non-compliance and contamination with a more plausible assumption.

Journal of medical screening·2026
Same author

The efficient stored specimen design for evaluating multiple screening technologies: Application to multicancer detection tests.

Journal of medical screening·2025
Same author

Minimum test tradeoff applied to uterine artery pulsatility index for antepartum prediction of preeclampsia.

Pregnancy hypertension·2025
Same author

The carbon isotope ratio of alanine is a biomarker of added sugar and sugar-sweetened beverage intakes: a pooled analysis of 4 studies.

The American journal of clinical nutrition·2025
Same journal

Methods for incorporating test result information within the high-dimensional propensity score framework: application in UK electronic health record data.

BMC medical research methodology·2026
Same journal

Sparse multi-way DMDC for longitudinal classification in high dimension low sample size data.

BMC medical research methodology·2026
Same journal

Tree-based exploratory identification of predictive biomarkers in non-randomized data.

BMC medical research methodology·2026
Same journal

Comparative evaluation of interrupted time series analytical methods for healthcare quality improvement research: a Monte Carlo simulation study.

BMC medical research methodology·2026
Same journal

Methodological advances in claims-based dementia algorithms: integrating medication and clinical data for medicare populations.

BMC medical research methodology·2026
Same journal

An interpretable XGboost algorithm for predicting 30-day mortality in acute pancreatitis using routine biomarkers.

BMC medical research methodology·2026
See all related articles

This study introduces a new method to bound bias in randomized trials with missing binary outcomes. The approach leverages randomization to estimate maximum bias, proving useful when less than 15% of data are missing.

Area of Science:

  • Biostatistics
  • Clinical Trials Methodology
  • Statistical Inference

Background:

  • Randomized trials frequently encounter missing binary outcome data.
  • Existing methods for handling missing data do not fully utilize randomization to bound bias, especially when missingness is not at random.

Purpose of the Study:

  • To propose a novel statistical approach for bounding bias in randomized trials with missing binary outcomes.
  • To explicitly use the randomization distribution to quantify maximum potential bias when missingness depends on outcome and treatment.

Main Methods:

  • Developed a method calculating anticipated maximum bias by multiplying two factors: potential confounding bias and an upper bound factor.
  • The upper bound factor is determined by the proportion of missing data within each randomization group.

Related Experiment Videos

  • Demonstrated that if missingness is below 15% per group, the upper bound factor is less than 0.18.

    • Main Results:

    • Applied the methodology to the Polyp Prevention Trial data.
    • Calculated an anticipated maximum bias of 0.025, which would not alter the trial's original conclusion of no treatment effect.
    • Observed low missingness (7% and 9%) resulted in a small upper bound factor (0.10) after covariate adjustment.

    Conclusions:

    • The proposed method is straightforward to implement in statistical software.
    • This bias-bounding approach is especially valuable and informative when the percentage of missing data in each treatment arm is below 15%.