中药临床试验的设计经常涉及到样本量的计算问题。样本量不足则检验效能过低，不易检出已存在的阳性结果，影响试验结果的准确性；样本量过大会造成资源浪费和执行困难，而且不适宜的样本量会有悖于伦理道德。本文比较了大样本试验和小样本试验的优缺点，并从多个角度探讨临床试验样本量设计的优化思路，为中药临床试验设计的样本量选择提供依据。

Sample size of Chinese medicine is a matter in design of clinical trails. The optimal sample size in a trial can assure the testing power to detect statistical significance. But using too many participants in a trial is expensive and is hard to carry out for researchers. Conversely，if the trial is underpowered for too few participants， it will be statistically inconclusive and may make the whole protocol a failure. So both too small and too large sample size are unethical. In this study we compare the advantages and disadvantages of different sample size，and present some method of design optimization，thus to provide a basis for the design of clinical trail of Chinese medicine.

R285.6

广东省重大科技专项(2012A080202017)；广东省科技计划项目(2010A30100009)。