Sample size calculations for comparative clinical trials with over-dispersed Poisson process data.

Area of Science:

• Biostatistics
• Clinical Trial Design
• Epidemiology

Background:

• Accurate sample size calculation is crucial for the validity and efficiency of comparative clinical trials.
• Traditional methods may not adequately address the complexities of Poisson or over-dispersed Poisson process data.
• Heterogeneity in event rates and time-varying effects present challenges in sample size determination.

Purpose of the Study:

• To develop and present a novel formula for sample size calculations in comparative clinical trials.
• To specifically address trials involving Poisson or over-dispersed Poisson process data.
• To create a formula that accommodates time heterogeneity, inter-patient heterogeneity, and time-varying treatment effects.

Main Methods:

• Development of a new sample size formula based on asymptotic approximations.
• Utilized a two-sample non-parametric test to compare empirical event rate functions between treatment groups.
• The formula's capability to incorporate various sources of heterogeneity and time-dependent effects was established.

Main Results:

• A new, versatile formula for sample size calculations in relevant clinical trials has been successfully developed.
• The formula demonstrates the ability to account for time heterogeneity, inter-patient variability, and time-varying treatment effects.
• An illustrative application to a chronic granulomatous disease trial is provided, demonstrating practical utility.

Conclusions:

• The proposed formula offers a more robust approach to sample size calculation for comparative clinical trials with complex event rate data.
• This methodology enhances the precision of sample size determination, leading to more reliable trial outcomes.
• The formula's applicability is demonstrated, suggesting its value for future clinical trial planning and statistical analysis.