Does Acupuncture Work? What the Evidence Shows

By Dov Michaeli, MD, PhD | Published 2/19/2018 4

woman getting acupuncture 696 x 464

What do we know about acupuncture? The short answer: Very little. The longer answer: Not much, but the potential is intriguing.

Clinical studies, especially the well-designed RCTs (Randomized Controlled Trials) by definition deal with populations, not with personal anecdotes. So please, read the article to the end and spare me the poison pens, the likes of “are you a real doctor?“, “This treatment/supplement saved my life“, or “Shame on you. You sold out to the medical-pharmaceutical conspiracy!” “It made my semen thicker since I started taking it.” These are not made up quotes; I have actually received such comments.


A bit about acupuncture

Acupuncture is a technique in which practitioners insert thin needles in the skin at specific points in the body. The practice originated in China, most likely around 100 BC.

Early acupuncture beliefs relied on concepts that are common in traditional Chinese medicine, such as a life force energy called qi. Qi was believed to flow from the body’s primary organs (such as the heart, liver, GI tract, kidneys) to the “superficial” body tissues of the skin, muscles, tendons, bones, and joints, through channels called meridians. Acupuncture points where needles are inserted are mainly (but not always) found at locations along the meridians. Acupuncture points not found along a meridian are called extraordinary points and those with no designated site are called “A-shi” points.

Interestingly, the belief in qi and the practice of acupuncture slowly died out in China, not so much because of lack of scientific proof, but because it was associated with the lower classes and its practitioners were in less prestigious occupations such as alchemy, shamanism, and midwifery. But then politics came to the rescue. The popularity of acupuncture rebounded in 1949 when Mao Tse Tung took power and sought to unite China behind traditional cultural values, including traditional Chinese medicine.


Is there a scientific basis for acupuncture?

When you examine the meridians of the different schools of thought of acupuncture, you find absolutely no neurological basis for them. There are no nerves that follow the meridians. What about the muscular system? It provides no anatomical basis for the practice either.

In a way, looking for a scientific basis for ancient theories is a bit unfair and certainly irrelevant. The ancient Chinese, just like the ancient Greeks, did not have the concept of empiricism in science. In fact, the circulatory system was discovered by William Harvey only in the 17th century. Until then, it was believed that the blood was moved by the lungs. Why? Because Galen, a second-century physician to the gladiator school in Pergamum said so—and for 1,500 years nobody bothered to check!

How about Qi? In the 18th and 19th centuries, vitalism was discussed by various biologists. There were arguments between those who felt that the known mechanics of physics would eventually explain the difference between life and non-life and the vitalists who thought that life could not be reduced to a mechanistic process. So is vitalism finally dead? If you read or listen to New Age-speak you realize that the concept of “life-force” or “energy” is still alive and well.


Examining the evidence

Before digging into the scientific literature on the clinical benefits of acupuncture for specific conditions, let me insert a word of caution: consider the source. A review of studies on acupuncture found that trials originating in China, Japan, Hong Kong, and Taiwan were uniformly favorable to acupuncture, as were ten out of eleven studies conducted in Russia.

The same study found that, methodologically, Chinese studies were inferior to Western trials. And lest you believe that “fake news” was invented by president Trump, scientist and journalist Steven Salzberg identifies acupuncture and Chinese medicine generally as a focus for “fake medical journals”</a,> such as the Journal of Acupuncture and Meridian Studies and Acupuncture in Medicine.   

 Nature magazine, a top-tier biomedical journal, described acupuncture and traditional Chinese medicine in general as “fraught with pseudoscience” with the majority of its treatments having no logical mechanism of action.

Are there any pearls to be found in the garbage?

I am not going to deal with conditions that the Cochrane Reviews, the gold standard of clinical knowledge, found to be ineffective, such as Alzheimer’s disease, insomnia, obesity, irritable bowel syndrome, whiplash, dry eyes, tinnitus, among many others. The main use of acupuncture has to do with specific causes of pain.

The Acupuncture Trialists’ Collaboration (not exactly a disinterested group) published the results of their meta-analysis, an update of a previous analysis. They found acupuncture to be moderately superior to sham acupuncture for non-specific musculoskeletal pain, osteoarthritis, chronic headache, and shoulder pain. They also found that the effects of acupuncture decreased by about 15% after one year. A review in Autonomic Neuroscience, among several other reviews, failed to show a difference between sham and real acupuncture.

  • Low back pain is touted by many practitioners and anecdotal testimonials as a condition treatable by acupuncture. The most recent clinical practice guideline, published in 2017, found low to moderate evidence that acupuncture was effective for chronic low back pain, and limited evidence that it was effective for acute low back pain. The same report found that the strength of the evidence for both conditions was low to moderate. Let me add that low back pain is a self-limiting condition. Personally, I suffered from low back pain with disc involvement for about 40 years and, except for a few severe attacks when I had to take ibuprofen, I have managed it with daily Yoga stretches, abdominal exercises, and maintaining my weight. And by all means, I recommend that back pain sufferers try to avoid the surgeon’s knife. The track record of surgical treatment is not brilliant, to say the least.
  • Osteoarthritic pain. A 2010 Cochrane review found that acupuncture shows statistically significant benefit over sham acupuncture in the treatment of peripheral joint osteoarthritis; however, these benefits were found to be so small that their clinical significance was doubtful, and “probably due at least partially to placebo effects from incomplete blinding”. Many other reviews either concur or come to an even more negative conclusion.
  • Rheumatoid arthritis pain. This 2012 review found the evidence for pain relief with acupuncture as “spare and inconclusive”.

Related Content:  The Best Ways To Treat Back Pain Without Drugs

Are there really no studies that show benefit?

Some positive reports are showing up and the good news is that they are of good quality.

Dr. Dawn Hershman of Columbia Medical Center in New York and her colleagues ran a well-designed (probably the best so far in the acupuncture field) study, narrowly focused on hormone receptor-positive (HR+) early stage breast cancers. HR+ tumors rely on estrogen to fuel their growth. To reduce the risk of cancer recurrence, patients take aromatase inhibitors after surgery. These drugs, which block estrogen production, are also used to prevent breast cancer in postmenopausal women at high risk for the disease and to treat HR-positive metastatic breast cancer. In a statement to the National Cancer Institute blog, Dr. Hershman described the problem: “About 50% of patients on these medications complain of some joint pain or stiffness, and about half of those patients describe the pain as severe.”

The 226 women in the trial were all taking a third-generation aromatase inhibitor—anastrozole (Arimidex®), letrozole (Femara®), or exemestane (Aromasin®) after surgery for early-stage HR-positive breast cancer. They were randomly assigned to receive true acupuncture, sham acupuncture (placebo), or no treatment. Sham acupuncture involves shallow insertion of short, thin needles at non-acupuncture points.

Roughly half of the study participants (110) received true acupuncture twice a week for 6 weeks, followed by once-a-week maintenance sessions for 6 more weeks. The other half were in one of two control groups: 59 received sham acupuncture on the same schedule as the true acupuncture group, and the remaining 57 received no treatment.

Study investigators followed the women for another 12 weeks after treatments ended. Patients reported on their pain before, during, and after treatment using various methods, including a questionnaire on which women could indicate a rating of “worst pain” on a scale of 0 to 10.

And the results: The NCI blog quotes Dr. Hershman saying,

“After 6 weeks, we saw a mean two-point reduction of worst pain [in the true-acupuncture group compared with worst pain before treatment], which is a major reduction, and these effects were maintained after 12 weeks.”

Moreover, she added that “even at 24 weeks, women in the true-acupuncture group had less pain overall than women in either [control] group.”

A 2-point drop may sound trivial, but in fact, it is clinically very significant. It may make the difference between patients “not taking it anymore” and discontinuing the treatment, and being able to tolerate the pain and staying on the life-saving drug.


What’s wrong with placebo anyway?

So now we have objective evidence of the efficacy of acupuncture over and above the placebo effect. But what’s wrong with placebo in the first place? There is ample evidence showing that neurohormonal changes occur in the brains of patients who react to a placebo treatment. So why does it matter to a clinician whether the effect is due to a painkiller affecting pain pathways in the brain, or due to endorphins generated by the brain and accomplishing the same effect? It doesn’t.

In fact, I would propose that it is time to “re-litigate” many of the negative trials that showed no significant effect over placebo. If the patients felt better, why deprive them of relief? Instead, let’s investigate why they got pain relief. It may turn out that they are more prone to releasing endorphins, or dopamine, secondary to activation of the somatosensory cortex, which is known to ‘light up’ due to any skin stimulation, being light touch or a needle stick. The somatosensory cortex has extensive connections to various regions in the brain, including the limbic system (aka the reward system). If you enjoyed a massage, or just a hug with a loved one, you activated your limbic system via the somatosensory cortex. If this is “just” a placebo effect, I’d still take it.


Dov Michaeli, MD, PhD

Dov Michaeli, M.D., Ph.D. (now retired) was a professor and basic science researcher at the University of California San Francisco. In addition to his clinical and research responsibilities, he also taught biochemistry to first-year medical students for many years.

During this time he was also the Editor of Lange Medical Publications, a company that developed and produced medical texts that were widely used by health professionals around the world.

He loves to write about the brain and human behavior as well as translate knowledge and complicated basic science concepts into entertainment for the rest of us.

He eventually left academia to enter the world of biotech. He served as the Chief Medical Officer of biotech companies, including Aphton Corporation. He also founded and served as the CEO of Madah Medica, an early-stage biotech company that developed products to improve post-surgical pain control.

Now that he is retired, he enjoys working out for two hours every day. He also follows the stock market, travels the world, and, of course, writes for TDWI.


  • I tried acupuncture after having severe pain on my hands/arms (feeling as if I was going handicapped). I couldn’t do much anymore, I could no longer wash my car, wash dishes, comb my daughter long hair, put on my bra, anyway, you get the point. I was in much pain, so my sister surprised me and took me for a visit to an acupuncturist (for she knew I wouldn’t go on my own). I have to say that after around 4 -5 visits I was feeling much better and was able to so a lot more things again. now when I experience sciatica or other pain I go to acupuncture because it helped me tremendously.

  • >What’s wrong with placebo anyway?

    Exactly! I understand on one hand why people write off the placebo effect, but on the other hand, I think “if it works, it works” is perfectly valid.

  • Hi Dov Michaeli, I’ve been looking into some very interesting research involving acupuncture’s effect on purinergic signalling and regulation of nitric oxide. This may be behind many of acupuncture’s health effects.

  • Acupuncture has worked for me for the following maladies.
    1. Sciatica
    2. Shoulder pain
    3. Headaches
    4. Plantar fasciitis
    5. General fatigue

    I find the therapy quite helpful.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Comment will held for moderation