Evaluating traditional Chinese medicine using modern clinical trial design and statistical methodology: application to a randomized controlled acupuncture trial

Abstract

Traditional Chinese medicine (TCM), used in China and other Asian counties for thousands of years, is increasingly utilized in Western countries. However, due to inherent differences in how Western medicine and this ancient modality are practiced, employing the so-called Western medicine-based gold standard research methods to evaluate TCM is challenging. This paper is a discussion of the obstacles inherent in the design and statistical analysis of clinical trials of TCM. It is based on our experience in designing and conducting a randomized controlled clinical trial of acupuncture for post-operative dental pain control in which acupuncture was shown to be statistically and significantly better than placebo in lengthening the median survival time to rescue drug. We demonstrate here that PH assumptions in the common Cox model did not hold in that trial and that TCM trials warrant more thoughtful modeling and more sophisticated models of statistical analysis. TCM study design entails all the challenges encountered in trials of drugs, devices, and surgical procedures in the Western medicine. We present possible solutions to some but leave many issues unresolved.

Figure 1

Kaplan–Meier (KM) survival curves and log–log curves across acupuncture treatment groups: (A1) and (A2) are the KM curves for the duration of first pain free time (Y₁) across the true (group ‘X’) vs. the placebo acupuncture group (group‘Z’) and across the true (group‘X’) vs. the sham acupuncture group (group‘Y’), respectively; (B1) and (B2) are the corresponding KM curves for the time to rescue drug (Y₂) across acupuncture groups; (A3) plots the Log[–log(estimated survival)] vs log(survival time) curves for the duration of first pain-free time (Y₁) across three treatment groups; and (B3) plots the corresponding log-log curves for the duration time to rescue drug (Y₂) across three treatment groups.