Acupuncture

Belief Systems and Rituals Are Not Science

It is important to emphasize that this is a belief system. It is not, nor has it ever been, science. Belief systems and cultural traditions exist outside of the realm of science. They are not empiric. Belief systems provide believers with comfort by unifying them into groups that share the belief. Belief systems give people a sense of belonging to something bigger and better than themselves.

Most belief systems have traditional rituals that members practice together and hold sacred, thereby deepening the feeling of belonging to the larger something. Rituals are soothing, reaffirming and provide psychological comfort to believers. Humans are a social species that evolved through group living, whereby individuals depended on having a place in the group. Rituals enforce the bond between group members. As illness is a major problem for any social group, healing rituals are common among human cultures. In prescientific cultures, healing rituals arise from the belief systems of those cultures.

The notion of acupuncture as a healing ritual within a cultural belief system is not what concerns us here. People have a right to believe what they want, and that right should be respected. But again, belief systems are not science. Indeed, people who find comfort in their belief system often lament those who confuse it with a field of scientific study. One writer in The Chinese Times laments about "those who acknowledge acupuncture’s clinical value while disavowing its traditional rational. This new breed of supporter is beginning to do more to undermine long-held notions about Chinese medicine than all the sceptics who doubted its effectiveness combined!"

As Dr. Novella has pointed out, concepts like qi and meridians do not correspond with any scientifically known aspect of human physiology. If we approach the concepts of qi, meridians, Yin and Yang, we find acupuncture's pretest probability (at least by way of these traditional concepts) to have a pretest probability that approaches zero. They are not science.

"Taoist" Elizabeth Reninger acknowledges that meridians are not anatomic structures:

"Like a network of rivers nourishing a landscape, the meridians are the channels through which qi flows, to nourish and energize the human body. These channels exist within the subtle body – you won’t find them on the operating table! Collectively, they form the matrix within which the physical body functions."

We are concerned with the modern marketing of acupuncture as if it were science.

Acupuncture as Pseudoscience

NCCAM defines acupuncture as, "A family of procedures that originate in Traditional Chinese Medicine. Acupuncture is the stimulation of specific points on the body by a variety of techniques including the insertion of thin metal needles through the skin. It is intended to remove blockages in the flow of qi and restore and maintain health."

Acupuncture as it is often marketed today, bears only superficial resemblance to the ancient Chinese practices, although many departments of 'Integrative Medicine" pay lip-service to them. Some actively promote the ideas of blocked qi and meridians, others propose scientific sounding mechanisms to explain acupuncture's 'effects'. Modern practitioners must resolve the cognitive dissonance that arises when confronted with the myriad of studies that fail to show significant effects in randomized controlled trials. They do so either by denying the studies outright, rationalizing them, or touting only those that suggest minor effects ( such as the studies on nausea, and extrapolating these morsels of evidence to support acupuncture as a legitimate treatment for a wide range of medical conditions).

To argue for a mechanism for Acupuncture is to assume that Acupuncture works in the first place. However, this has not been established, and as we shall see, is not supported by the evidence. The stasis point of conjecture cannot be assumed here. Recall the fallacy of the false premise and Dr. Hall's "Tooth Fairy Science".

Let's look at some of the hand-waving explanations given by otherwise respected institutions for Acupuncture's "Tooth Fairy".

"There are many theories as to how acupuncture actually works. When acupuncture points are stimulated, it causes a dull ache or other sensations in the muscle. One theory holds that the stimulated muscle and sensory neurons send a message to the central nervous system (the brain and spinal cord), causing the release of endorphins (naturally produced pain killers) and other neurotransmitters (body chemicals that modify nerve impulses), which help block the message of pain from being delivered to the brain and have other regulatory effects as well."

Cleveland Clinic

"The anatomical points (acupuncture points) are thought to have certain electrical properties, which affect chemical neurotransmitters in the body."

Brigham and Women's Hospital (Harvard)

"It is based on the concept that the body has specific channels (meridians) through which energy (chi) flows. While this energy, or chi, is flowing smoothly, there will be no specific pain or disease, but if the chi is either deficient or excess in quantity, or is not flowing freely, there will be disease. When acupuncture needles are placed in specific points on the skin, energy is redirected to bring about a healing response."

Johns Hopkins

"The scientific explanation of how acupuncture works is through the nervous system. The channels are thought to represent the nerves that course from the spinal cord and spread throughout the body. When the nerves are stimulated by an acupuncture needle, signals are sent to the brain and chemicals, called neurotransmitters, are released. These chemicals trigger the body’s self-healing mechanisms."

Beaumont Health System

Statements such as these --from institutions that should be promoting science and critical thinking-- cast doubt on the scientific literacy of their administrators and doctors. As we discussed, qi and meridians have a near zero plausibility. However, putting needles in the skin is an active practice. It is not really like homeopathy in this regard in that homeopathic remedies have absolutely no ingredients. So, the plausibility of the practice of needling the skin to treat disease may be higher than zero. RCTs may be somewhat appropriate, but the results would have to be unequivocal. Without such evidence, acupuncture remains a pseudoscience.

Back Pain?

Now, what about back pain? It is a common claim among departments of Integrative Medicine that acupuncture helps with back pain.

Two major RCTs were then done with good methodology, but null (negative) results and bizarre conclusions by the authors. Remember, the purpose of setting up control groups ahead of time is to isolate any real effect of the treatment from all of the other variables. If the treatment effect is not better than the predetermined control, then we are dealing with the placebo effect. In science, this means that it does not work. We need to keep this in mind.

A 2005 Annals of Internal Medicine meta-analysis looked at the issue. It reported on 33 RCTs of acupuncture for low back pain. The studies varied in methodology. The study stated, "Acupuncture effectively relieves chronic low back pain. No evidence suggests that acupuncture is more effective than other active therapies." More specifically, for chronic back pain, it stated, "For the primary outcome of short-term relief of chronic pain, the meta-analyses showed that acupuncture is significantly more effective than sham treatment ...and no additional treatment." Also, "For patients with acute low back pain, data are sparse and inconclusive. Data are also insufficient for drawing conclusions about acupuncture's short-term effectiveness compared with most other therapies."

So, for short term management of chronic (but not acute) low back pain, acupuncture might make a difference. Again, the pretest probability for an active intervention such as acupuncture is not zero, so perhaps it should be studied further.

The 2006 Archives of Internal Medicine RCT looked at the role of acupuncture on chronic low back pain. Patients were randomized to 3 groups. The first received real acupuncture at designated acupuncture points on the meridians. The second received superficial needling at random, non-acupuncture points, and the third group was given nothing. The results showed that the 'nothing' group fared the worst. The other two groups did the same. There was no significant difference in back pain between the acupuncture group and the control group in the short run or in the long run. Skeptical doctors would conclude that the acupuncture was no better than the null effect. However, the authors concluded that not only did the acupuncture work, but so did the fake acupuncture!

The authors' conclusion is contrary to what one would expect given the nature of the study design. At best we can say that acupuncture points and meridians do not matter (and that they likely do not exist). At worst, we can say that acupuncture does not work for chronic back pain.

Similarly, a 2009 Archives of Internal Medicine RCT, chronic back pain patients were randomized to get real acupuncture, superficial needling at random points, or a third group which received the superficial twirling of a toothpick without breaking the skin. All patients were adequately blinded. The results were similar to the 2006 study. The groups who thought they were getting acupuncture all fared the same as the group that actually got the acupuncture. Even twirling toothpicks at random locations seemed to produce the same results. Clearly, acupuncture for chronic back pain dose not do better than the placebo methods. Therefore, we should conclude that it doesn't work.

However, as with the 2006 Archives RCT, the authors strangely felt that it was inconclusive. "Although acupuncture was found effective for chronic low back pain, tailoring needling sites to each patient and penetration of the skin appear to be unimportant in eliciting therapeutic benefits. These findings raise questions about acupuncture's purported mechanisms of action. It remains unclear whether acupuncture or our simulated method of acupuncture provide physiologically important stimulation or represent placebo or nonspecific effects." It is as though the authors tried to spin the conclusion to make it seem that acupuncture may still have some hope at being a scientifically valid treatment for chronic low back pain.

In September 2012, news agencies were abuzz with reporting on a meta-analysis published in the Archives of Medicine entitled Acupuncture for Chronic Pain, Individual Patient Data Meta-analysis. In this analysis, the authors compared various studies that looked at patients with a wide array of chronic pain. They selected studies that used traditional acupuncture compared to sham acupuncture and also compared to "usual care" (no "real" or "sham" acupuncture). The study was funded by NCCAM and the Samueli Institute, two organizations with a highly vested interest in alternative medicine. They found that pain was reported as significantly improved in patients receiving acupuncture compared to "usual care". The effects were as high as a 50% reduction on a 100-point pain scale. However, when compared to sham acupuncture, the effect was only a 5-point reduction. From this data, the authors conclude:

"Acupuncture is effective for the treatment of chronic pain and is therefore a reasonable referral option. Significant differences between true and sham acupuncture indicate that acupuncture is more than a placebo. However, these differences are relatively modest, suggesting that factors in addition to the specific effects of needling are important contributors to the therapeutic effects of acupuncture." (Underline added)

That is a very generous conclusion. A 5-point difference between "real" and "sham" acupuncture may reach a degree of clinical significance on paper, but the clinical difference is unimpressive. Another major problem, as David Gorski pointed out, is that none of the studies were double-blinded. Therefore, none of the studies meet criteria for good scientific methodology. The patients were blinded, but the examiners were not. The 5-point difference in pain between the 2 groups may be easily accounted for by this fact (see the Clever Hans Effect).

The Guardian reported on this study and included the comments of Dr. Edzard Ernst:

Edzard Ernst, emeritus professor of complementary medicine at the University of Exeter, said the study "impressively and clearly" showed that the effects of acupuncture were mostly due to placebo. "The differences between the results obtained with real and sham acupuncture are small and not clinically relevant. Crucially, they are probably due to residual bias in these studies. Several investigations have shown that the verbal or non-verbal communication between the patient and the therapist is more important than the actual needling. If such factors would be accounted for, the effect of acupuncture on chronic pain might disappear completely."

In 2013, a meta-analysis in the American Journal of Chinese Medicine looked at the trials of acupuncture for back pain. The authors were as charitable as they could be to try to rescue acupuncture, but really could not distinguish it from the placebo effect:

"Compared with no treatment, acupuncture achieved better outcomes in terms of pain relief, disability recovery and better quality of life, but these effects were not observed when compared to sham acupuncture. Acupuncture achieved better outcomes when compared with other treatments. No publication bias was detected. Acupuncture is an effective treatment for chronic low back pain, but this effect is likely to be produced by the nonspecific effects of manipulation."

In 2017, a highly touted study from Australia was published entitled, “Acupuncture for analgesia in the emergency department: a multicenter, randomized, equivalence and non-inferiority trial”. It claimed to be the largest study of acupuncture of its kind. They began with 1,964 patients, but only 528 patients met inclusion criteria and agreed to the study. They all had common painful conditions that are seen in an ER. They were randomized into 3 groups: acupuncture alone*, acupuncture plus analgesic meds, and analgesic meds alone. There was no placebo arm. The study was “single blinded”, meaning that those who collected the data were blinded to the treatment. Only 16% of the patients had clinically relevant pain relief. They found that there were no clinically significant differences between the three groups*. The authors claimed that they found that acupuncture treatment in the emergency room setting for acute pain was “noninferior” to established analgesic medication.

It should be noted that due to the “single blinding”, the patients absolutely knew what treatment they were getting. The ER doctors knew as well. This essentially is a non-blinded study. Why didn’t they blind the patients using sham acupuncture? Also, each group could request “rescue” opiates in addition to their primary treatment if their pain was inadequately controlled.

*The acupuncture only group requested “rescue” opiates twice as much as the other two groups. One should properly infer that this group as a whole had much less pain relief than the meds-only or the combined groups. The authors tacitly admitted this in the body of the paper but did not seem to note this in the conclusions. None of the groups actually had much pain relief. One wonders why the medication groups only had 16% claiming relevant relief. Also, since there was no placebo arm, there is no way of knowing how effective any of the treatments actually were. Being non-blinded, we can infer that there was a significant placebo effect in the acupuncture group given all that we know about acupuncture and the placebo effect.

This study failed to show much of anything at all. It certainly did not meet the massive burden of proof that is demanded for any implausible treatment with very little prior probability.

See the articles by skeptical bloggers Orac and Dr. Steven Novella for further details.

And as we have seen above, if we consider only the studies with the best controls and the best blinding, the effects disappear completely. Meta-analyses of doubtful claims can only contaminate good data with bad. Garbage in, garbage out.

Looking at all of these results with proper scientific methodology can only lead to one conclusion:

** Acupuncture does not work for chronic pain beyond the placebo effect. Nor does it have evidence of efficacy for post-op nausea, chemotherapy induced nausea and vomiting, fibromyalgia, rheumatoid arthritis, Parkinson's, irritable bowel syndrome, asthma, overuse syndrome, hypertension, insomnia, depression or any other condition for which proper RCTs and reviews have been done.

What's the Harm?

Acupuncture has been found to be relatively safe in experienced hands. It is still an invasive procedure involving putting foreign bodies into the skin. Direct harm has been done by way of direct side-effects (puncturing the lung for example), but these complications are rare. Moral harm can be done by promoting placebo treatments, which, as we discussed in The Placebo Effect section, constitutes lying to patients. We decided that this is never morally correct.

As with all of the pseudoscience systems we discuss, the real harm comes about when patients and providers decide to forgo proven medical treatments for serious conditions in favor of the pseudoscience. Fortunately, most of the Integrative Medicine departments of major academic centers have stressed that patients still should follow their doctor's advice. However, there are many examples of supporters of pseudoscience on the internet and in the public eye have not. This list serves as an important lesson.

Why Do People Like It?

As stated above, acupuncture's origin is that of a healing ritual within a belief system. As such, it is not science any more that the rituals of other belief systems. For those in the belief system, it is not science. When acupuncture is transplanted from the belief system into the arena of scientific medicine, it becomes pseudoscience.

As with homeopathy, studies have shown very clearly that it is not the actual implantation of needles that makes people feel better, despite the unwarranted scientific sounding rationalizations offered. People feel better because of the attention, the care, the relaxing atmosphere, the trust in the provider, the exotic nature of the 'treatment', the personal validation, the belief in the treatment and all of the other details that come with the package.

Given this understanding, it seems reasonable that we could harness some or most of this benefit for our patients without telling them that implausible, placebo-equivalent treatments are really beneficial.

Practitioners thrive on the personal anecdotes of happy patients. Thus, they practice 'experience-based medicine' rather than 'evidence-based medicine' (and certainly not 'science-based medicine'). For them, no amount of scientific evidence can change their mind. They have taken the blue pill and cannot see the Matrix for what it really is.

John Byrne, MD