Related Experiment Videos
Exact repeated confidence intervals for Bernoulli parameters in a group sequential clinical trial
1Roosevelt University, Chicago, Illinois.
Controlled Clinical Trials
|February 1, 1993
Summary
This study introduces exact repeated confidence intervals (RCIs) for Bernoulli trial analyses. These intervals offer a flexible way to evaluate treatment success probabilities at interim stages, improving upon fixed significance tests.
Area of Science:
- Biostatistics
- Clinical Trial Methodology
- Statistical Inference
Background:
- Group sequential trials require methods for interim data analysis.
- Traditional repeated significance tests use rigid stopping rules.
- Evaluating treatment efficacy requires flexible confidence intervals.
Purpose of the Study:
- To present methods for constructing exact repeated confidence intervals (RCIs).
- To apply RCIs to single and dual Bernoulli treatments in group sequential trials.
- To offer an alternative to repeated significance tests for interim analyses.
Main Methods:
- Development of exact repeated confidence intervals (RCIs).
- Application to success probability (p) for a single Bernoulli treatment.
- Extension to the difference in success probabilities (delta = p1-p2) for two independent Bernoulli treatments.
Main Results:
- Exact RCIs can be constructed for key treatment efficacy measures.
- These RCIs provide continuous evaluation of accumulating data.
- Methods are adaptable for relative risk and odds ratio calculations.
Conclusions:
- Exact RCIs enhance data evaluation in group sequential trials.
- RCIs offer a more informative approach than fixed stopping criteria.
- The presented methods support robust interim analysis in clinical research.