Clinical Trials on Acupuncture
Tim H. Tanaka, Ph.D.
The Pacific Wellness Institute, Toronto, Ontario, Canada
Dept. Acupuncture Moxibustion, Tsukuba University of Technology, Japan
Updated January 18, 2011
The vast majority of research papers in acupuncture until recent years are clinical case reports, which are considered low level of evidence based on the EBM concept.4
That is because case reports are uncontrolled studies, involving descriptions of the patient’s history, administered treatments and outcome. Even if the study was based on successful results on hundreds or thousands of patients; we really cannot tell how much, if any, administered therapeutic intervention contributed to improvements of the patient’s condition, without observation of a comparative patient population (otherwise known as a control group). For this reason, case reports are usually not accepted by reputable peer-reviewed journals for publication (unless the subject is involved in very rare cases).
A study by Haruto Kinosita of Japan in 19715 is probably the first randomized control clinical trial of acupuncture ever conducted. A clinical trial of acupuncture appeared in an English database in 19736 and since then, a growing number of controlled clinical trials have been conducted in the US and Europe. The greatest number of acupuncture studies have been conducted and published in China. However, the vast majority of the studies are non-controlled studies.
In previously conducted controlled trials, virtually all research trials suffered from methodological limitations. There are some difficulties when conducting acupuncture trials. For example, while EBM aims to establish the most appropriate treatment for large populations with the same clinical condition, the traditional concept of acupuncture is to establish treatment principles for each individual patient rather than disease populations. Another difficulty is the issue of an adequate control group and blinding both practitioners and patients when conducting a clinical trial of acupuncture. Unlike the double-blind drug trials, it is hard to blind the practitioners and patients in an acupuncture study. There have been many acupuncture trials using “sham” acupuncture, however, “sham” acupuncture used in many previous studies are not considered physiologically inert (it means there are potential therapeutic effects). Readers of those previous studies need to be fully aware of this fact, when interpreting the outcome of studies, in particular if a study found no statistical difference between the acupuncture group and the “sham” acupuncture group. In recent years, different types of placebo acupuncture needles have been invented.7-9 Single-blind 10 11 and even double-blind 12 studies are claimed to be possible using those placebo needles, however there are some considerable therapeutic limitations and restrictions involved with those needles.
Keeping in mind various issues surrounding acupuncture studies, it is still essential to conduct research and produce quality evidence. However, there is a need for utilizing new or modified research designs that can accommodate the unique, individually oriented nature of non-drug intervention such as acupuncture. Some research methodologies that are considered suitable for acupuncture studies are using stratified randomization according to each patients’ traditional East Asian medicinal diagnosis patterns. In addition, study design such as n-of-1 RCT, which considers an individual’s unique characteristics, as well as their responses to interventions.13 14 Thus far this experimental design has primarily been used in psychology but its usage has been recently expanded and utilized in acupuncture research trials.15 16 An outcome of the n-of-1 RCT is considered high quality evidence. However the experimental design can be utilized only for patients that exhibit certain chronic conditions. It is not a suitable design to evaluate the efficacy of acupuncture for acute or episodic types of conditions.
Finally, the most important issue when interpreting outcomes from clinical research studies is the fact that acupuncture protocols used in many previous studies are far from those commonly used in clinical practice. When researchers plan the study, acupuncture protocols are determined not only by its potential efficacy but also by its experimental design, budgets, and other methodological factors. This generally results in compromised acupuncture treatments which become simple, short, and standardized protocol. The results derived from compromised acupuncture should not be automatically assumed to be the same as clinical acupuncture.