Comparison of Three Methods for Power and Sample Size Determination in Bioequivalence Test

摘要

Bioequivalence test is becoming more frequently used in clinical trials. Investigators are now asking statisticians for advice on how to plan and analyze studies using bioequivalence test, which includes questions on sample size determination (SSD). SSD is a crucial aspect of experimental design. Here, three methods were discussed for evaluating power and sample size in bioequivalence test with two-independent samples design and two-period crossover design. Based on summarizing the factors which affected power and sample size in traditional significance test and bioequivalence test such as α, β,θ,Δ,δ and σ, especially introduced the meaning ofαandβin detail, we explore some of the questions surrounding SSD for bioequivalence test in this paper. We found that the power will be negative calculated by simplified Jen-pei Liu and Shein-Chung Chow’s method when n is small or CV is large, which is undoubtedly unreasonable. What’s more, we confirmed computer simulation is a great way to work out sample size and it can make up for the defections of the other two methods.