阳性药对照临床试验中非劣效性检验的统计分析方法及相关问题

现阶段所进行的新药临床试验设计方法中,由于以安慰剂为对照的优效性临床试验存在伦理方面的问题,以阳性药为对照的非劣效性临床试验在新药临床试验中的作用越来越重要。本文就现有几种非劣效性临床试验统计学方法进行回顾比较并附以实例。经比较,可信区间法在一类错误上相对于合成法较保守,而两阶段阳性对照试验法则可通过对阳性对照药的疗效的判断尽早的终止无效的非劣效试验。最后,探索分析了现有非劣效检验统计方法需要改进的地方。

Statistical Methods for Non-inferiority Testing in Active-controlled Clinical Trials

With the concern about ethic of placebo-control clinical trials,non-inferiority testing is often considered in active controlled trials without a placebo arm.In this work,we discuss and compare several statistical methods and illustrate them with examples.We find that the Indirect Confidence Interval Comparison method is more conservative.And through the Two-stage Active Control Testing method,the uninterpretable non-inferiority testing may be avoided prior to the end of the trial.Finally, some suggestions ...