Simultaneous confidence interval construction for many-to-one comparisons of proportion differences based on...

Area of Science:

• Medical Statistics
• Biostatistics
• Clinical Trial Design

Background:

• Paired organ data analysis is crucial in ophthalmology, orthopedics, and otolaryngology.
• Comparing multiple groups with a control presents statistical challenges due to multiplicity and intra-patient correlation.

Purpose of the Study:

• To construct asymptotic simultaneous confidence intervals (SCIs) for many-to-one comparisons.
• To address the complexities of multiplicity adjustment and correlation in paired organ data analysis.

Main Methods:

• Development of asymptotic simultaneous confidence intervals (SCIs).
• Evaluation of coverage probabilities and widths using Monte Carlo simulation studies.
• Application of the methods to a real-world medical data example.

Main Results:

• The proposed SCIs effectively adjust for multiplicity and correlation in paired organ comparisons.
• Simulation studies demonstrated the performance of the developed confidence intervals.

Conclusions:

• The new SCIs provide a robust statistical tool for analyzing paired organ data in multi-group medical studies.
• This methodology enhances the reliability of statistical inference in specific clinical research areas.