目的:探讨随机化检验(Randomization test)在内部预试验IPs(Internal Pilot Study)自适应设计样本量调整中对Ⅰ型错误和检验效能的影响.方法:利用蒙特-卡罗(MonteCarlo)法模拟样本量较小时的IPS样本量调整,分别采用随机化检验和t检验分析最后数据并比较二者对Ⅰ型错误、检验效能值的影响.结果:重计算的第二阶段样本量波动性较大,t检验不能很好地抑制Ⅰ型错误,随机化检验能较好的抑制Ⅰ型错误,检验效能略有降低.结论:在临床试验样本量较小的情况下,内部预试验盲态下样本量调整后随机化检验能保护Ⅰ型错误不增大,同时保证检验效能亦满足要求.
Application of randomization test in the sample size adjustment for internal pilot study
AIM: To probe into the effect of randomization test on type Ⅰ error and power when it is used to ananlyze the data from IPS (Internal Pilot Study) with samplesize adjustment. METHODS: Monte Carlo was used to simulate IPS samplesize adjustment on condition that IPS samplesize is small. Randomization test and t test were used simultaneously to analyze the final data in order to compare the different effect of two methods on type Ⅰ error and power. RESULTS: While the recalcultion sample size for the second phase is volatilely fluctuant, t test can not prevent type Ⅰ error from inflating. However, randomization test can prevent type Ⅰ error from inflating better with little lower of power. CONCLUSION: The randomization test for the final data can not only prevent type Ⅰ error from inflating, but also maintain the desired power if we re-estimate the sample size under blinded IPS when it has small sample size.
clinical trialsamplesize adjustmentinternal pilot studyrandomization testmonte carlo simulationtype Ⅰ errorpower
NETL
NSTL
万方数据
维普
