On the efficiency of adaptive sample size design.

Area of Science:

• Clinical Trials
• Biostatistics
• Statistical Power

Background:

• Adaptive sample size designs and group sequential designs are alternatives to fixed sample size designs.
• These designs aim to improve statistical power and trial efficiency.

Purpose of the Study:

• To investigate the efficiency of adaptive sample size designs in comparison to group sequential designs.
• To demonstrate that adaptive designs can be uniformly more efficient than group sequential designs.

Main Methods:

• Construction of a uniformly more efficient adaptive sample size design based on a given group sequential design.
• Comparison of expected sample sizes and statistical power between adaptive and group sequential designs across a range of true treatment differences.

Main Results:

• A uniformly more efficient adaptive sample size design can be constructed from a group sequential design with the same maximum sample size and rejection boundary.
• The adaptive design consistently shows a smaller expected sample size than the group sequential design while maintaining statistical power.

Conclusions:

• Adaptive sample size designs are more efficient than group sequential designs, challenging popular beliefs.
• Adaptive designs offer better performance, easier implementation, and are attractive for modern clinical trials.