Incorporating the sample correlation into the testing of two endpoints in clinical trials

Area of Science:

• Biostatistics
• Clinical Trial Methodology
• Statistical Inference

Background:

• Multiple endpoints in clinical trials require careful statistical adjustment to control type-1 error rates.
• Traditional methods like the Bonferroni correction can be overly conservative, reducing statistical power.
• Accurate estimation of correlation between endpoints is crucial for effective multiple testing procedures.

Purpose of the Study:

• To introduce an improved Bonferroni method for testing two primary endpoints in clinical trials.
• To develop a data-adaptive critical value that incorporates the sample correlation coefficient.
• To provide a more powerful and less conservative approach to multiple endpoint testing.

Main Methods:

• The methodology utilizes Student's t-test statistics for comparing means under normality assumptions.
• A confidence interval is constructed for the unknown population correlation coefficient.
• The proposed method estimates the type-1 error rate using the lower confidence limit of the correlation.

Main Results:

• The improved Bonferroni method demonstrates a less conservative upper bound for the type-1 error rate compared to the traditional Bonferroni method.
• The data-adaptive critical value effectively incorporates sample correlation, enhancing test precision.
• Performance comparisons show advantages over existing multiple testing procedures.

Conclusions:

• The proposed data-adaptive Bonferroni method offers a statistically sound and more powerful alternative for analyzing multiple primary endpoints in clinical trials.
• This improved method enhances the reliability of clinical trial results by optimizing the balance between type-1 error control and statistical power.
• The findings suggest a significant advancement in statistical strategies for clinical trial design and analysis.