It is well known all over the world that the randomized controlled trial (RCT), is the only study design for experiments involving human subjects that can create evidence of the highest quality regarding the efficacy and safety of a new treatment. In Japan, however, most clinical researchers have shown no interest in RCT and thus the number of good RCTs is quite small. Although the recent development of evidence-based medicine has played a role in promoting RCT among clinical researchers, the number of researchers who can understand what an RCT is, design one properly and analyze data appropriately is also quite small. This situation is also completely applicable to the world of complementary and alternative medicine. In this paper, we shall introduce the reader to the concepts behind RCT and a statistical way of thinking that is indispensable for properly conducting RCT to glean scientific evidence of the efficacy and safety of complementary and alternative medicine.