A recent meta-analysis evaluated the efficacy of acupuncture for treatment of irritable bowel syndrome (IBS) and arrived at bizarrely positive conclusions.

The authors state that they searched 4 electronic databases for double-blind, placebo-controlled trials investigating the efficacy of acupuncture in the management of IBS. Studies were screened for inclusion based on randomization, controls, and measurable outcomes reported.

Six RCTs were included in the meta-analysis, and 5 articles were of high quality.  The pooled relative risk for clinical improvement with acupuncture was 1.75 (95%CI: 1.24-2.46, P = 0.001). Using two different statistical approaches, the authors confirmed the efficacy of acupuncture for treating IBS and concluded that acupuncture exhibits clinically and statistically significant control of IBS symptoms.

As IBS is a common and often difficult to treat condition, this would be great news! But is it true? We do not need to look far to find the embarrassing mistakes and – dare I say it? – lies on which this result was constructed.

The largest RCT included in this meta-analysis was neither placebo-controlled nor double blind; it was a pragmatic trial with the infamous ‘A+B versus B’ design. Here is the key part of its methods section: 116 patients were offered 10 weekly individualised acupuncture sessions plus usual care, 117 patients continued with usual care alone. Intriguingly, this was the ONLY one of the 6 RCTs with a significantly positive result!

The second largest study (as well as all the other trials) showed that acupuncture was no better than sham treatments. Here is the key quote from this trial: there was no statistically significant difference between acupuncture and sham acupuncture.

So, let me re-write the conclusions of this meta-analysis without spin, lies or hype: These results of this meta-analysis seem to indicate that:

  1. currently there are several RCTs testing whether acupuncture is an effective therapy for IBS,
  2. all the RCTs that adequately control for placebo-effects show no effectiveness of acupuncture,
  3. the only RCT that yields a positive result does not make any attempt to control for placebo-effects,
  4. this suggests that acupuncture is a placebo,
  5. it also demonstrates how misleading studies with the infamous ‘A+B versus B’ design can be,
  6. finally, this meta-analysis seems to be a prime example of scientific misconduct with the aim of creating a positive result out of data which are, in fact, negative.

24 Responses to Acupuncture for IBS: the evidence shows effectiveness, but only if one is very liberal with the truth

  • I have saved this (the four point rebuttal in particular) to post along with my comments on acupuncture. I invariably get a tidal wave of, “acupuncture DOES work” with studies such a this cited and pleas for me to “educate yourself”. Acupuncture is the woo that has gained the most credibility, and even many who are somewhat skeptical believe it to have some value.

    The battle is long and difficult.

  • Yes Irene.
    It will indeed be a long and difficult battle. Scores of (real) physicians take pride in their own prowess in this allegedly ancient but oh-so inexplicable art. Interestingly, most of those I know have grown tired of applying acupuncture in practice but are reluctant to denounce it altogether. They use what little speculation there is about gate-theory or endorphin release or whatever, to smother their inevitable cognitive dissonance from finding it not so useful. It is apparently quite common for those who want to keep up appearances to go over to electro-acupuncture, which is of course in effect TENS with a dash of TCM-hype.

    I think you and many other regulars here know the following, but I want to throw this in for the benefit of other readers.
    Edzards writings are of course a treasure trove for us but what I have found to be invaluable when preaching the sad truth of acupuncture is to recount the verifiable 3/4 century long history of “modern” acupuncture as uncovered by men such as Ben Kavoussi and Paul Unschuld (Rose Shapiro: “Suckers”, Page 76-82; Interview with P. Unschuld in German) .
    I also point out salient facts such as that there was no way of producing the thin stiff (“filiform”) steel needles used in the modern, less traumatic version of acupuncture until the 20th century so any 2000 or even 200 year history is just not plausible.

    But best of all, and the real reason I ventured to comment on this, is to spread the word about my favourite acupuncture “killer”, the book “Thirty years in Moukden” by Dugald Christie, a Scottish surgeon and missionaire. In this book, Mr. Christie gives an horrifying eye-witness account of TCM as it presented to a western surgeon in a large chinese city in 1883 to 1913. His reports of the traditional chinese medical practice and in particular the torture techniques even then called acu-puncture, are seriously shocking. It turns out that acupuncture was in those days little more than a variant of the western bloodletting culture and any real therapeutic value of the ancient art was at best incidental. The west and east were never really isolated and influences readily traveled both ways via the silk-road for example, even in ancient times. TCM was in no way as marvelous as we are led to believe. As a benign example of acupuncture only a century ago, sick children were pierced to draw blood with sharp lances (not filiform needles, they did not exist) under the fingernails. I guess that stopped them complaining.

    Mr. Christie’s book can be obtained for free as electronic downloads in many formats and even read online. For example here It is worth every minute spent reading it at least twice.

    • Why should people denounce it?
      I sympathise with your point of view,however, if acupuncture stimulates a placebo effect and people are told that this is the mechanism (complicated as it is!) to why it can be beneficial i do not see the woo.

      It is interesting that sticking stainless steel needles into the innocent tissue creating tissue damage that was not there originally (however small) to improve a dysfunction is so readily accepted is bizarre in my opinion, but who am i to force others to my line of thinking. If thats how they choose to bring about placebo the so be it.

      There are many ways to skin a cat. Everyone believes that the way THEY find a balance is the best way for everyone else to achieve what they have. They assume we want what they want. If we educate and let people chose their own mechanisms for bringing about change is that so bad?

      It is the certainty of the language and tone used to justify, perpetuate and denounce treatments that i find more distressing than the treatment itself.For instance even the language used by Irene above i.e: the battle is long and difficult.

      What battle? i dont want to fight. Do i have to use a sword? I think ill go put the coffee machine on instead.

      • the data do not actually show that ‘acupuncture stimulates a placebo effect’. they merely demonstrate that it is associated with a placebo effect. this is no justification for using acupuncture: if we administer a therapy that works beyond placebo with empathy, we generate a placebo effect as well – PLUS EFFECTS BEYOND THAT!!!
        in other words, it is unethical to use ineffective treatments just because they also generate placebo effects.

        • Plus side effects beyond the effects.
          When studies show that patients who received acupuncture have fewer days with symptoms, use less medication with side effects, make fewer visits to doctors, and take fewer sick days off work, what therapy can produce the same results, aside from sham acupuncture?

      • @ Neill – as far as “sticking stainless steel needles into the innocent tissue…” issue: it’s not really that bizarre, if you look at where acupuncture falls in context. Traditionally, acupuncture wouldn’t be the first “go to” therapy. First would be adjusting diet and exercise. Then massage/manual therapy. If things haven’t been nudged back into balance, more drastic measures like acupuncture would be used. Many consider acupuncture like surgery – if things have progressed to the point where it’s needed…it’s a wonderful tool to have. But it is more invasive.

        So you have to look at studies like this in perspective. First, IBS isn’t really a ‘thing’ in the Chinese medical model (what the west calls IBS would be symptoms of conditions like Liver Qi Stagnation, Spleen Qi Deficiency, or one of many others). I’m sure the researchers know this, and it was factored into the study.

        Second – acupuncture is one tool in the toolchest. A patient with Spleen Qi Deficiency wouldn’t go for “acupuncture” – they would go for Chinese medicine. Which may or may not involve the use of needles. So a study like this would help determine to what extent the needles would be helpful, and how it would fit in with diet, exercise, manual therapy, herbs, etc.

        Thinking that acupuncture alone would be a solution for anything isn’t groundbreaking news. That’s been known for thousands of years. It’s a tool. More important is that the practitioner know how and when to use the tool, and it’s strengths and limitations.

        • That should have said “…acupuncture alone wouldn’t be a solution…”. But hopefully the typo was obvious.

          • I agree, in that context suddenly the behaviour isn’t as bizarre.
            It is the understanding of what the strengths and limitations of each therapy are. Interpretation of research and experience bias I suppose is what leads to the difficulty in deciding ethical and unethical.

  • What if we don’t have a therapy that will generate an effect beyond placebo?
    Is it ethical to use therapies that are associated with placebo effect to try to bring about placebo?

    • I would say NO.
      and besides, can you name a condition/symptom for which there is no effective therapy [I don’t mean a cure but a treatment that alleviates symptoms]?

      • For many patients with IBS, insomnia, cough, pain, etc., there is either no effective therapy, or the side effects are worse than the symptoms. That’s why they turn to alternatives.

  • No I Can’t. But at the same time I cant name a therapy that I know to be 100% effective.
    In most conditions/symptoms are there not anomalies where effective therapies do not alleviate symptoms? I suppose that would then be an incorrect diagnosis?
    I have to disagree and believe there are situations where it is ethical to use acupuncture if thats what an individual consents to. As for how i would decide on ethical/unethical I don’t know. I suppose Im lucky that its not up to me.

  • Thanks, Edzard, for the post. And thanks, Björn, for the book recommendation. I read the first 60 pages of the color PDF and was enthralled. I haven’t done acupuncture for 8 months now, but keep reading research, systematic reviews, etc. A few times I wondered if I was being too hard on my old profession, as some of the reviews sure make it sound like it may be better than placebo for things like IBS, back pain, shoulder pain. I have several former patients who would like me to start practicing again. However, I’m enjoying my new direction and am still aware that the best evidence shows acupuncture to be a “theatrical placebo.” With Dr. Ernst’s encouragement I am working (slowly) on a book about my experiences with TCM.

    Neil, the field of medical ethics is what we are looking at. Important considerations for the practitioner include giving proper informed consent and respecting patient autonomy. The old way of doing things (such as for many MDs of the 1950s) is considered ‘paternalistic’ and means that the patient shouldn’t bother to evaluate the diagnosis, prescription, and therapy for its scientific basis but should just do what the Dr. says. In today’s world, it is ethically required that the practitioner discuss the options, likelihood of success, possible complications, potential adverse effects to watch out for, and help enable to the patient to make choices based on established facts and the best research. In other words, the days of “I don’t think anything’s wrong with Mrs. Smith, so I’ll sell her a sugar pill for $100 and tell her it is a strong medicine to help her sleep” should be over.

    However, homeopaths, chiropractors, and acupuncturists continue this tradition, even though many of them are self-deceived and think they have good intentions (some of them don’t realize how anti-science they have become to guard their pet theories). The catch is that someone with a special degree and license in a modality should make every reasonable effort to stay up on the best research regarding it. In homeopathy, this research shows it is no better than a sugar pill (and that it is, indeed, a sugar pill). In acupuncture, it shows that the diagnosis techniques are unreliable and that the point prescriptions do not matter–it is mainly the confidence and compassion with which the treatment is administered which determines the subjective improvements the patient will report. Add to the the big number of people with self-limited conditions on the verge of resolving on their own who, at the worst of their pain, go to an alternative practitioner and spend a few hundred dollars for placebo treatments, then credit those treatments with the improvements which would have happened anyhow (this is particularly the case with low back pain).

    The acupuncture research which matters most to me is the retracting needles which don’t even pierce the skin. When there is no difference between putting needles through the skin (which has some risk and discomfort) and pretending to put needles in the skin (which has no risk of infection and less discomfort), it would be ethically preferable to just tell patients that they have had acupuncture done on highly effective traditional acupuncture points which will resolve their disease when, in fact, they have had toothpicks twirled on the surface of the skin at random points while getting attention in a comfortable atmosphere.

    But somehow that seems unethical, doesn’t it? Even if it works just as well and has less risk of complications… Going down the placebo path ultimately means that the most confidently proclaimed fake treatment by the most charismatic practitioner charging as much money as possible (as more expensive, exotic placebos work better) will lead to better reports of improvement (in subjective symptoms) from patients. Throughout time, from Mesmer to John of God to today’s Naturopathic homeopath, every confident practitioner has had some happy patients who blame their improvement on the intervention, as the improvements happened after the intervention. And the ones who died (from undiagnosed cancer, heart failure, etc.), usually don’t complain much.

    • After reading your reply to my comments i looked at the definition of placebo. Something i should have done 2 days ago. The word “deception” that was used to help define placebo has changed my understanding. I have to agree totally that a placebo based on deception is not a path I think we should follow. Is it possible to have placebo effect without the deception? I suppose not if we have to deceive to achieve it.

      • Neil, every treatment has a placebo component, including surgery, prescription medications, bodywork, etc. Research has shown that “warm, empathic communication” is key to creating a higher placebo response (which translates to the patient feeling heard, calm, and confident in the good intentions of the practitioner).

        Here is another study on IBS with only sham acupuncture, but different styles of communication:

        Placebo effects, as far as I know, only change subjective symptoms, such as pain, nausea, and anxiety. If those symptoms are caused by an underlying disease process which is treatable (such as infection, tumor, broken bone), giving a placebo treatment may make the patient report dramatic improvement in their pain, nausea, or anxiety, but may leave the underlying problem to get worse. However, if a practitioner who can actually diagnose and treat the problem does so with care, confidence, and compassion, the problem will be fixed *and* the patient will report quick resolve of their pain, etc. A brusque, rude practitioner who appears to not listen and dispenses treatment in a cold, rushed style may also fix the underlying problem, but the patient may not have as quick of relief from their pain or anxiety.

        • Ex-Acupuncturist, if research has shown “warm, empathetic communication” is key to creating a higher placebo response can we still call it placebo? Surely its best practice and evidence based, not placebo. The fact that we can’t quantify the exact reason for the level of placebo, doesn’t remove the consistency of the effect?

          I don’t mean to be awkward but I’m new to this type of discussion and would be grateful of your opinion.

          • If we take a theoretical patient with high blood pressure accompanied by headache and feelings of stress/anxiety who chooses to try acupuncture (hoping it will let them reduce or stop their medications), one would hope that the acupuncture points which are claimed to lower blood pressure or treat the TCM patterns (usually of Liver Yang Rising or Excessive Liver Heat where high blood pressure is part of the picture. Other patterns such as Kidney Yin Deficiency, Dampness in the Spleen, etc., are often said to be contributing as well) would work better than random needles put in by drunken monkeys (with good bedside manner). Not only would we hope that the ‘real’ acupuncture would do better than Drunken Acupuncture by Monkey Nurses (DAMN), but that it would have a lasting effect (i.e. unblock the meridians on a long-term basis, or to get more wooish, energetically motivate the person to change their lifestyle in healthy ways).

            ‘Real’ acupuncture gets divided into 2 types: individualized treatment and point prescriptions. Personally, I feel that a point called ‘lower blood pressure point’ out to do just that to earn the name, and that if 10 textbooks say Liver-3 and Large Intestine-4 with Spleen 6 and the Ear Lower Blood Pressure Point should treat most hypertension patients effectively, we should see a difference using that prescription on a big enough group compared to DAMN.

            The SHARP trials (Stop Hypertension with the Acupuncture Research Program) compared individualized acupuncture, standard point prescription acupuncture and DAMN sham acupuncture (well, I doubt they used monkeys…). Results? No difference between the groups.


            Free full text:

            Here’s the neatest chart from the study:

            That was with 12 acupuncture treatments–about $720 out of pocket on average in the western world, and probably 12+ hours of time over a couple months. By the end of the study, the blood pressure numbers were really close to where they were to begin with, though they all dipped for some time (though never to healthy/safe levels).

            So here is where we can say with confidence that ‘traditional acupuncture and standardized acupuncture work no better than DAMN placebo acupuncture for hypertension.’ While one can find smaller studies in pro-CAM journals which make it look like acupuncture can help hypertension, I haven’t seen any larger studies with equal or better methodology showing different results.

            This is where one can say that ‘acupuncture is just a placebo when it comes to treating high blood pressure,’ and that it would be a dangerous waste of time and money for a patient to go to an acupuncturist hoping they could reduce or eliminate their anti-hypertensive medications through acupuncture.

            Therefore, acupuncturists who advertise to treat hypertension with acupuncture are at best ignorant of the evidence (I see them as kind of like brainwashed cult members at this time–I used to be one). Some may say that they also give herbs or counsel lifestyle changes which may help hypertension… Giving both acupuncture and herbs was the standard practice of my first Chinese teachers. I thought that was a poor practice as then one wouldn’t know which one helped the patient’s problem. However, now I see that giving both makes more money and gives a stronger placebo effect… I was taught the traditional way!

            The theory of TCM says that if the underlying imbalance is corrected the symptoms will go away. But I haven’t seen research which backs up that hypothesis. If someone has hypertension (let’s say it’s genetic, not due to obesity), herbs which lower blood pressure may work (similarly to prescription drugs, through diuresis & vasodilation), but they don’t create a permanent ‘cure’ after a short course of therapy.

            Many MDs will tell a person with hypertension that, if obesity or lack of physical activity is a contributing factor, exercise and dietary restriction may reduce or eliminate the need for medications. Some patients will apparently not hear that advice coming from their doctor, and then later report that “all the MD wants to do” is give them pills. They may go to an acupuncturist or chiropractor, who may agree that MDs are pill pushers, but then suggest that they come get an ineffective treatment every week for months or years (at $30-$75/week) and not take the pills in order to be ‘more natural.’ One can only hope that the CAM practitioner also uses post-hypnotic suggestions to encourage the patient to exercise more and skip the deep-fried lard.

            Anyhow, it’s late and I’m digressing. To go to an acupuncturist to treat something like hypertension, hot flashes, or apparently even low back pain because you think that ‘real’ acupuncture is better than DAMN acupuncture shows that one is not informed of the best evidence. Since most of us can now access the best evidence from our cell phone, tablet, or laptop no matter where we are, there is little excuse for ignorance. An acupuncturist, chiropractor, or homeopath/naturopath now needs to hope their patients don’t do a good job searching the internet or reading Wikipedia about their treatments. Either that, or they need to become rabidly anti-science and convince their patients that their is a a vast conspiracy to cover up good evidence on Wikipedia, etc..

  • Does sham acupuncture work, or is it only effective when the patient thinks it’s real acupuncture?

    • Researcher Fabrizio Benedetti has written highly regarded books about placebo effects (

      He proposes ‘placebo balanced 4 arm trials’ which make a lot of sense to me, and would be great for acupuncture research. Group 1 gets the real treatment, is told it is real. Group 2 gets the real treatment, is told it is placebo. Group 3 gets the sham treatment, is told it is real. Group 4 gets the sham treatment, is told it’s placebo.

      Given that many acupuncture points (and groups of points) are taught as having specific effects (lower blood pressure, induce perspiration, increase endurance, relieve headache… or in TCM terms, clear liver heat, disperse wind-heat, supplement the spleen Qi, soothe the liver Qi…), this type of research could be done. What Dr. Ernst points out is that some acupuncture researchers seem to structure their studies in a way which doesn’t produce solid answers or new information, and is virtually guaranteed to produce a “positive” outcome. That’s what the A+B>A means: usual care plus more attention with positive, compassionate communication and coached relaxation will almost always lead to patients who report more improvement in subjective symptoms. It’s a waste of time and resources to do studies which don’t provide actual new insights, unless one’s goal is simply to promote a pet modality which actually has little substance to it.

      • How are clinical trials done now? One group gets the treatment, another gets the placebo, and both are told that it’s treatment, or both are told that it’s randomly assigned? The latter is true, while the former involves deception, as does the 4 arm trial. Surely any agreement the subjects sign can’t lie to them.

  • I think the placebo effect of acupuncture would be greater on me if it was administered by a monkey. Id be that impressed that the monkey could organise the correct procedures to satisfy the health and safety executive that Id forget about what took me there in the first place.

    Thanks for the posts Ex-Acupuncturist

  • Thanks for the thanks, Neil. A famous acupuncturist, Miriam Lee, is eternalized for the 10 point combination known in the field as ML-10. She ran a busy practice and used the same 10 points for almost every patient. Interestingly, my first Chinese teachers of acupuncture told me, sheepishly, that one only needed to know about 10 points to run an acupuncture clinic, but 365 points to pass the tests. I didn’t ‘get’ it at the time.

    In her book, _Insights of a Senior Acupuncturist_, Miriam Lee talks about the Traditional Chinese Medical theory of ‘why’ the 10 points work so well on such a variety of patients. She goes into the ‘energetics’ of the stomach, spleen, liver, lung, and how these points address deficiencies or excesses in each meridian as associated with most diseases and modern living… Never does she suppose that they could work as well as other point combinations because virtually all acupuncture is a theatrical placebo, with the possible addition of small non-specific effects from poking needles in people (endorphins, adenosine, dopamine…).

    One of the big ‘movements’ in the acupuncture world right now is Community Acupuncture. This is a low-cost approach with many patients in one big treatment area in recliners (generally, that’s the approach) who do simpler treatments such as the Miriam Lee 10 and the ear 5 point needle protocol (5-NP) which came from the acupuncture drug detox world (but unfortunately doesn’t work better for drug detox than any other combination of points, which is not very well at all in any case). I mostly did ’boutique’ acupuncture at $45-$60 a session with just 2 treatments rooms (but only 10-20 patients per week). However, if I were convinced that acupuncture was more than a placebo and actually helped to fix medical problems rather than just temporarily improve subjective perception of pain, nausea, etc., I’d get into Community Acupuncture. But if acupuncture is only as good as DAMN and sham, it doesn’t seem ethical or actually helpful to target lower income people by giving cheaper treatments to more people… Part of me still wishes that it weren’t that way (after all, I still owe tens of thousands of dollars of student loans for a Master’s degree in acupuncture) and that I could believe that it actually helps people, but I’ve chosen to look for the best scientific evidence about it, rather than the most hopeful anecdotes from dedicated patients and practitioners. And so I’ve moved on towards creating a new life where I can actually help people and make money without deception. I deceived myself, and unwittingly my patients, for too many years.


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.

Subscribe via email

Enter your email address to receive notifications of new blog posts by email.

Recent Comments

Note that comments can be edited for up to five minutes after they are first submitted but you must tick the box: “Save my name, email, and website in this browser for the next time I comment.”

The most recent comments from all posts can be seen here.