Assessment of individual bioequivalence using sufficient bootstrap procedure.

Area of Science:

• Biostatistics
• Pharmacokinetics
• Clinical Trial Design

Background:

• Assessing individual bioequivalence is crucial for drug product comparisons.
• Traditional bootstrap methods may have limitations in specific study designs.
• The 2x4 randomized crossover design is frequently used in bioequivalence trials.

Purpose of the Study:

• To propose a novel bootstrap method for individual bioequivalence assessment.
• To evaluate the performance of the proposed method in a 2x4 crossover design.
• To compare the proposed method against the traditional percentile bootstrap technique.

Main Methods:

• Development of a sufficient bootstrap method utilizing unique observations in resamples.
• Application of the method to a 2x4 randomized crossover design.
• Validation through extensive Monte Carlo simulations and a real-world experimental dataset.

Main Results:

• The proposed bootstrap method demonstrates robust performance in finite sample simulations.
• Comparison indicates the novel method is a strong competitor to classical percentile bootstrap confidence limits.
• In some scenarios, the proposed method outperforms traditional techniques for bioequivalence assessment.

Conclusions:

• The proposed unique observation bootstrap method is a valid and effective tool for individual bioequivalence analysis.
• This method provides a reliable alternative for statistical inference in 2x4 crossover bioequivalence studies.
• Further research may explore its application in other complex clinical trial designs.