Design considerations in crossover trials with a single interim analysis and serial patient entry

Area of Science:

• Biostatistics
• Clinical Trial Design

Background:

• Crossover trials with serial patient entry pose analytical challenges.
• Standard analyses may not fully leverage data from patients with incomplete treatment periods.

Purpose of the Study:

• To present a two-stage sequential design for crossover trials with a single interim analysis.
• To provide a method for controlling the type I error rate in such designs.
• To evaluate the design's efficiency compared to standard crossover trial analyses.

Main Methods:

• A two-stage sequential design incorporating a single interim analysis.
• Utilizes a linear statistic combining data from patients observed for one or both treatment periods.
• Type I error rate is controlled by partitioning across the two analyses with derived critical values.

Main Results:

• The proposed design allows for a single interim analysis in serial crossover trials.
• It maintains statistical control by partitioning the type I error rate.
• Investigates sample size inflation factors and potential savings in required responses.

Conclusions:

• The two-stage sequential design offers an efficient approach for crossover trials with serial patient entry.
• It provides a statistically sound method for interim analysis in these complex trial designs.
• The procedure demonstrates potential for optimizing resource allocation in clinical studies.