English Chinese (Simplified) Japanese Korean Spanish

"Difficulties Facing Modern Acupuncture Research in America" By Alison Warren

By Alison Warren, L.Ac, Dipl. OM, ABT

     Acupuncture research in America is not yet as common as perhaps it should be, and when it is done, it is often subject to much criticism by our western counterparts.  Usually this translates into accusations of not being scientific enough, and to be honest, I agree.  I believe there is much improvement to be made in our research designs, which hold the potential to bring about change in America’s apprehension. 
     There is a hierarchy of validity among the different categories of research.  To many scientists, experiments are considered the gold standard, as they are meant to establish cause and effect by maximizing control over variables.  Among the different types of experiments, a double blind randomized control is held in the highest regard.  There are many other means of gathering data, each with their own advantages and disadvantages.   Correlation studies tend to be perceived as a step down from experiments, and observational studies are usually seen as quasi-experimental.  This perception is a bit skewed however, as many scientists would not disregard a study that yields a high correlation between two variables.  The majority of studies about acupuncture are going to be of experimental design.  The following are a few main potential complications if our field of research.

     We’ve all been taught the qualities of good research.  It should be systematic, controlled, valid, reliable, and replicable, but the focus of this paper is where this quality becomes compromised.  One inherent problem with our methodology of research is the lack of specificity.   We should understand the broad range of variability among individuals better than any health profession.  While these individual differences allow this medicine to shine by considering these variables in diagnosis and thus customizable treatments, they can become a hindrance when it comes to research results.  It is my belief, however, that this potential hindrance can be significantly minimized by designing research studies with greater specificity.  For example, there are mixed results about the efficacy of acupuncture to treat knee pain.  Well, what kind of knee pain?  There are so many different etiologies of knee pain with varying severity, chronicity, and prognosis – hence the need to determine what injuries were being evaluated.  If a study evaluating the effectiveness of acupuncture under the broad topic of “knee pain” includes ACL tears, MCL tears, meniscus injuries, patellofemoral pain syndrome, and arthritis, it is more than reasonable to expect varying results.  Furthermore, were these subjects in good health? Were they young or elderly?  Have they had prior surgery for the problem?  Have there been repeated injuries to the knee?  When one begins to dissect the details (which should be our forte), the potential problems begin to become clear, but so do the solutions.  We can see then that these are factors that we need to take into account.  Here is where we can raise the specificity of a study.  I think a better question to ask, for example, would be “How effective is acupuncture in treating anterior knee pain due to patellofemoral pain syndrome?”   In answering this question I can then design a study controlling for the above mentioned parameters.  Decide what age range I would like to study, what severity range on the pain scale, how long they’ve been experiencing the pain, etc.  An 85 yr old obese diabetic male with PFPS who has had previous knee surgery and multiple injuries to the knee is going to respond very differently than a 35 yr old active male in good health with recent onset knee pain.  The more specific we can be, the better we can measure results. 
     The topic of sham acupuncture is a pet peeve among many acupuncturists.  I myself find it difficult not to go on a ranting rampage about the topic.  One thing that needs to be cleared from the minds of both eastern and western scientists once and for all is this: There is no such thing as sham acupuncture.  Uneducated and poorly guided acupuncture perhaps, but not “sham”.  As such, it should not be used as the control group for comparison.  By definition, the control group should not be receiving the variable being measured.  If group A is receiving an acupuncture knee protocol and group B is receiving sham acupuncture…they are both receiving acupuncture.  If you stick a needle anywhere in the body, there will be an effect.   How much an effect and whether it is the desired effect is another story, but will produce an effect nonetheless.  It is my opinion that this is due, in part, to the fact that we too often fail to articulate to western researchers and doctors how this medicine actually works.  We can’t criticize a doctor who puts a needle in the toe and doesn’t think it will effect they eye if we have not explained the somatovisceral pathways of the meridians.  Likewise, we cannot fault a doctor inserting a needle in a painful site distal to the area of complaint if we do not explain the value of a shi points (and the fact that they too are acupuncture points).  It is my belief that this would clear up many misconceptions and cynicism about acupuncture and in doing so, help to refine the way in which the research is carried out.
  What then, should we use for comparison?   If we revisit our knee pain example, a better study may involve 2 groups who have been seeking the same allopathic care (treated with NSAIDs for example), and then divided into group A, who receives acupuncture, and group B who continues their NSAIDs.  It’s important to be clear what you are comparing.   Acupuncture versus physical therapy, acupuncture versus an analgesic, acupuncture versus no treatment, are among the numerous examples which could be implemented

     On a final note, it would also behoove our profession to elucidate the training we receive in needle technique.  When inserting an acupuncture needle, there is a significant difference between an acupuncturist with 4 years of training when compared to a practitioner with 100 hours of training.  As such, it is important who is designated to perform the acupuncture protocol in the study. 
The above mentioned points are just a few minor details that we can keep in mind when designing a research protocol.  They are certainly not all-inclusive, but just a few adjustments that can help improve the value of acupuncture research in America.