Sep 16 2012
Can we finally just say that acupuncture is nothing more than an elaborate placebo?
You are currently browsing comments. If you would like to return to the full story, you can read the full entry here: “Can we finally just say that acupuncture is nothing more than an elaborate placebo?”.
33 Responses to “Can we finally just say that acupuncture is nothing more than an elaborate placebo?”
I am a little puzzled about the idea to express an advantage of a treatment in terms of SD.
When one is planning research, one needs to know the SD, and also the size of the conjectured effect in terms of this SD, otherwise one doesn’t know how many patients to include to draw any kind of meaningful conclusion.
But the SD refers to how much patients differ between themselves, and actually not even that: only the patients included in a particular investigation. But an individual patient doesn’t care much about how his or her discomfort or pain changes, compared to the range of discomfort seen in all patients of all doctors together. The pain of the individual patient goes up and down and possibly the patient will notice it when the decrease in pain is compared to the patient’s personal range.
On the personal level I think that 0.25 SD is unnoticeable. If your pain decreases by 0.25 SD on top of the normal fluctuations, you’ll still be worse off in about 40% of the cases if the degree of pain is normally distributed, whereas without this 0.25 SD you’ll be worse off in 50% of cases.
One can also express ’0.25 SD’ in the number of individual tests one must do to reject the null hypothesis with p=0.05, say. That number runs in the hundreds. Any effect that needs 17,922 experimental patients to establish, cannot be observed at all by a single patient, or even by a single doctor who carefully records what happens to all his or her patients.
Just imagine that you have to tell a patient: ‘It is impossible to know whether this unpleasant procedure will help you, but it seems that if you and 10,000 people like you are averaged, one can see some kind of difference.’
As a thought exercise: let’s examine this from an economic point of view. Medical care costs are high.
A key aspect that is missing here is cost control. If one placebo, acupuncture, is only mildly better than other placebos, why not look for a less expensive and time consuming placebo that is equally effective?
The next step, following the logic of this study, would be to compare acupuncture, M&Ms, and a massage chair in terms of pain relief. If M&Ms are as effective as the other treatments, then it would make sense in economic terms to replace all acupuncture “therapy” with M&Ms.
Thanks (to both of you) for looking into this one. I was curious as soon as i read the piece in Futurity. The piece (Wood et al) on bias toward beneficial effects suggests what you’re saying — whatever the main design, to include with it (if not as it) an analysis using only those studies that account for unconcealed allocation and are double-blinded. (I’ll read it in full later; thanks also for linking to it.) While the authors split their analysis into acupuncture–no treatment and acupuncture–sham acupuncture, it seems that they did not include a high-quality-only analysis.
Also:
Isn’t this description something of a p-value fallacy? That is, isn’t publication bias better described as the tendency for positive studies to be more likely to be published than negative studies? Pardon my pedantry.
I never pardon pedantry, as regular readers know.
But seriously, pedantry is notorious for derailing discussion threads into arguments about little points that don’t have an effect on the main message. If that little pedantic issue of wording bothers you so much, I’ll change it to exactly your words.
There, it’s done. Next topic.
Ha ha! I’m sorry to have sounded accusatory. I was actually slightly worried that i might have the wrong idea and didn’t want to just tell myself it was a typo and move on. Thanks again. Next topic.
they may help against that kind of pain, but they increase dentist costs!
Acupuncture is more than a placebo. If you used toothpicks, you wouldn’t get the entire effect that acupuncture has on the human body. For example: http://www.cbsnews.com/8301-504763_162-57503549-10391704/swiss-healer-accused-of-intentionally-infecting-16-people-with-hiv-using-acupuncture-needles/
Dr Gorski- were you aware that SESAP 14 has a question or 2 on acupuncture? It appears CAM woo has even crept into surgery! I believe it’s claimed to be effective for pain. It sort of embarrasses me as a surgeon and Fellow of the College. Your thoughts?
Cymbe,
Your link does not match your comment.
“Acupuncture is more than a placebo. If you used toothpicks, you wouldn’t get the entire effect that acupuncture has on the human body. ”
Acupuncture could just be a better placebo than toothpicks.
However, if you blinded patients as to whether needles or toothpicks were used….
BillyJoe, Cymbe was being sarcastic. With a placebo one can’t infect people that easily!
Yep, Cymbe’s link perfectly supports the comment – acupuncture had an effect which toothpicks would not. That effect was rather negative, but still it was an effect.
An excellent illustration of why, even if one were to provide a placebo treatment for whatever reason, that treatment should NOT be acupuncture, because it carries very real risks that can be avoided by choosing a different placebo.
In retrospect, after having written this post, it occurs to me that I don’t recall ever having seen a meta-analysis that expressed its results in that manner. That bothered me. It stuck out like a sore thumb. At the time I didn’t quite know why.
Better go easy on the birthday scotch
In other words, the meta-analyzer has been racking his brains for a way to make favorable sense of the data. If that is so, the question is: how many more methods he has tried in the seclusion of his office? Maybe it a case of ‘If you torture the data long enough, they will confess’ ascribed to Ronald Coase (Coase supposedly said ‘it will’ rather than ‘they will’).
http://en.wikipedia.org/wiki/Ronald_Coase
lawyer k. martin kuny – k. maritn kuny, pc…
Science-Based Medicine » Can we finally just say that acupuncture is nothing more than an elaborate placebo?…
[...] me as having the methodological rigor of a typical Andrew Wakefield or Mark Geier study—or an acupuncture study. Perhaps that’s why I wasn’t too surprised when one of my readers pointed out that one [...]
I had an incident as a child..5 or 6 years old, when my throat was closing because I slept in front of a fan on a hot summer night. My aunt and mom rushed me to her doctor, where he had me lay down, took off my shoe and sock, touched the bottom of my foot..for one slight instant, and my throat and congestion completely opened up and no more problem after! Is that a known problem..someone being congested by sleeping in front of a fan? To this day I cannot do it…I have to have air circulating me, but not pointed directly at me.
As I understand it..it’s call ‘acupressure’ now… and how it relates to acupuncture..it’s along that line suppose. I know I’ve heard it’s oriental healing originally. But, isn’t it possible that one nerve in the foot affects something in the bronchial area..being connected, and so perhaps there is something to it? Anyway, that’s all I know, and have to say on the subject
Acupressure is also known to cure DEATH!!!!
Seriously, as a person of Chinese ethnicity, I find the Western reading of “Traditional Chinese Medicine,” acupuncture, “Ayurvedic medicine” and acupressure completely offensive, Orientalist (see Edward Said’s definition of the word) and essentialist. When I say Western reading, I am not referring to Western medicine, but the interpretation Western practitioners of those CAMs have of their practices.
And just an FYI, RH, the Ming emperors actually banned acupuncture. And they are considered among the most intellectually inclined dynasties.
Exactly, Mr. Luong (I’m not adept at HTML to attempt the French letter in your first name)! It is offensive to claim “Western” versus “Eastern” on a planet that is close to being a sphere. I often tell those who claim that Western medicine is bad that means that the varicella vaccine and statins are good because they were developed in Japan. It is offensive to ignore the very real actual medical and science research that happens in Asia.
I didn’t know that Francois. But, it did help me that day….anyway..interesting. I don’t go into it myself..just wondered..thanks for that info
What helped you is something called the placebo effect, which makes you feel much better because you think something is being done to you. Which is also why snake oil have such a success in the free market.
How do you explain muscle twitches and the immediate relief of pain and increase in flexibility. Placebo? Your all double blind. Do you think all we do is just insert the needle and then pray. I guess the redness around the needle the light sweating and other skin changes are placebo. Also I like that you have to add in the elaborate placebo. At least it is better than just an M&M. Thanks for reminding us that you think we are either idiots or immoral a-holes. I love the pharmaceutical industry how they double blindly in random fashion seem to get away with all those side effects with a profit.
I especially like the story of the acupuncturist who spread aids. 1. he was a music teacher 2. the story stated they had no idea what tools he used to infect his students. Nice work love the research not biased at all. Science? more like old fashioned prejudice about another belief system. Seems like acupuncture is not much different than black people in the 50′s. Acupuncture should not have the same privileges as regular medicine.
kscrimgeour:
“Your all double blind. Do you think all we do is just insert the needle and then pray. I guess the redness around the needle the light sweating and other skin changes are placebo. Also I like that you have to add in the elaborate placebo. At least it is better than just an M&M. Thanks for reminding us that you think we are either idiots or immoral a-holes. I love the pharmaceutical industry how they double blindly in random fashion seem to get away with all those side effects with a profit…Science? more like old fashioned prejudice about another belief system. Seems like acupuncture is not much different than black people in the 50′s.”
You may recognise yourself in this article:
http://theness.com/neurologicablog/index.php/some-feedback-on-organic-farming/
Actually, the immediate pain relief can be a placebo effect – as soon as I swallow ibuprofen my headaches fade slightly, well before the medication has dissolved and reached any nociceptor. I was having breakfast with someone once who said they hadn’t had any coffee and their head hurt. I suggested it might be caffeine withdrawal. They took a sip of tea and immediately said their head felt better. So yeah, immediate pain relief can be placebo. There’s no question acupuncture has a physiological effect, the question is whether it is a new phenomenon (and for me, whether there is any benefit to training someone in anything besides sterility and avoiding organs). I certainly have no issue with “acupuncture” that doesn’t involve TCM, avoids elaborate (imaginary) diagnoses and uses thin needles that barely break the skin (or even better, toothpicks).
I was under the impression that acupuncturists also charged for their services, at sufficient levels to surpass their fixed costs, thus achieving what most people would call “a profit”. The only “double” I see in your comments is a double standard – drugs must be proven safe and effective before being used, meanwhile acupuncture proponents are asking for a free pass on criticism or an evidence base. That seems like hypocrisy.
You mean, like the Chinese government banning the practice of acupuncture in the 17th century?
“old fashioned prejudice about another belief system”?
What about new fashioned prejudice in favor of exempting CAM from the standards of science?
Remember, it’s not an either-or proposition.
Thanks for the replies. I can agree that avoiding organs and sterility is important and may be the most important. I question the theory when it comes to acupuncture and most agree the theory does not apply well to acupuncture but mainly for herbs. I have changed my practice to a more medical model and the results are better now I can agree.
There are a few techniques that involve meridian theory that I still see working quite well. These involve palpation as well. The best understanding of acupuncture by most medical doctors who practice is that it is actually stimulating the fascia which is found to contain a high concentration of nerve endings and is responsible for our ability to touch our nose with our eyes closed. This is bringing new light to how the early folk understanding came about.
I think the traditional theory might also be a bit much as well. And yes an number of emperors banned acupuncture but all where because the acupuncturist would not listen. The most historic was Hua tuo who treated an enemy soldier while working as the emperors main physician. All the acupuncturists who knew him were ordered to be beheaded. It was not because the emperor thought it didn’t work but more anyone who was an acupuncturist had one of Hua tuo’s books or so it seemed and would be deemed guilty by association.
There are a number of responses that occur during acupuncture and the client has a right to stop treatment at any time. This means if you do not get results quickly then they will and do stop coming. Some do get great results in fact many. Also realize most people are like yourselves they think it is bunk and do not want to try unless it is a last resort only. Usually after trying drugs surgery and physio to no avail. It saves money for many for the cost of acupuncture is far cheaper than seeing a specialist again who has already stated they will just have to continue with their pain meds and wait for it to get bad enough for surgury.
I want to share this article and I invite criticism.
http://well.blogs.nytimes.com/2012/09/11/acupuncture-provides-true-pain-relief-in-study/?smid=fb-share
Thanks Billy Joe for the reminder to tone down the rhetoric and tune up the logic fallacy theory. I don’t think I used any profanity and it was quite clear the article quoted as a proof of the dangers of acupuncture was perhaps closer to the article you referenced for me to read. It had NO SUBSTANTIATED facts what so ever about acupuncture. It was sort of an inside joke at an old boys club was my impression considering this is an evidence based forum.
I apologize for my perspectives on pharmaceuticals I work in detox so I see quite a few fresh faces every week addicted to pills they were given by their doctor and pharmacist. They took them as prescribed and now they need more resources to get off them. Up to a year for benzos this is very expensive for everyone. Nor T does this lead me to see antidepressants as very safe. Useful but safe that is a stretch but better than being depressed they are over prescribed and this is clear and evidence based.
Also I read a study that counted upwards of 300 000 deaths from properly prescribed medications in a ten year period I beleive it was 1995 to 2005. Considering you can count deaths from acupuncture on your hand my logic places acupuncture as a better first option considering the risk alone. Am I wrong?
Trying a harmless and very well documented as effective doesn’t it make sense to have it part of the system. If it avoids the risk of harm from medication and they are harmful and the safety is only statistical like driving. Acupucture you can be 100% sure with today’s standards no harm will come. The first premise of medicine is “Do No Harm.”
So Acupuncture can save the system money and to use it safely people need to see their MD and their acupuncturist. The result will be less over prescription less unnecessary surgeries. Considering a surgery at a basic cost is $10 000 or more that is about 200 acupuncture treatments. My guess is at least one of those 200 would benefit enough to realize they don not need a surgery even if it was placebo. This is guess work but with numbers so large if you hit 200 golf balls one of them will be on the fairway just by distribution logic.
So when you look at the big picture it only makes sense to save money by using a safe technique in conjunction with medical supervision and tests. The best thing for acupuncture is for it to be included as a CAM because it saves money and statistically it does not kill or harm at all.
“The best thing for acupuncture is for it to be included as a CAM because it saves money and statistically it does not kill or harm at all.”
Neither does doing nothing, when nothing needs to be done. How is it better than that?
The history of acupuncture does not justify its use, only its proven results. To date, those results support only the use of acupuncture to reduce pain and nausea for the short term, and no impact on specific conditions. I don’t really have a problem with these indications, provided it’s practiced with minimal risk, and no association with magical thinking.
Though I am familiar with the theory that acupuncture meridians are related to connective tissue planes or fascia, I am unaware of any proof for that theory. Further, wouldn’t a better approach be to abandon the historical acupuncture meridians and points, or at best test them to see if there are any specific associations? Rather than defending traditional practice, a science-based approach involves testing and discarding the useless parts – like diagnosis, qi, meridians and herbs.
The cost of acupuncture is a pure drain on the economy and individuals if it is ineffective. If acupuncture is a pure placebo, it’s questionable whether it’s worth reimbursing through insurance schemes and it’s definitely not worth spending a large amount of time and money on it. If acupuncture is as cheap as $5 for a brief treatment (which is all that is supported now considering it’s effects are short-term at best), that’s fine. But if it involves a lengthy consultation, a long treatment, and is accompanied by denigration of real (i.e. proven) medicine it is not worth the cost.
In addition, your invocation of a false dilemma (i.e. if drugs cause harm, acupuncture works) raises questions about your true understanding of medicine, science and the application of skepticism. No authors here believe drugs are without risks or that pharmaceutical firms are angels, and there are many posts on this subject. They merely apply the same standards for CAM – which generally lacks evidence of being effective beyond placebo. I question any treatment that is marketed with minimal evidence of efficacy, under the rubric “it’s safer than drugs so it must work”. You wouldn’t support a new drug that was supported by little more than Pfizer’s CEO saying “this is based on a thousand years of history and in my experience it works really well” – why should the same standard not apply to acupuncture.
I understand that as a (probable) acupuncturist, you want to believe and you have a strong economic motivation to find it effective (i.e. the conflict of interest you find so abhorent in Big Pharma). However painful it might be, from an ethical (and certainly a scientific) perspective, it is worth questioning the true efficacy of your profession rather than blindly and blandly accepting the rhetoric and inaccurate claims spoon-fed to the world about acupuncture. May I suggest you switch to a lower-cost, lower-risk model involving sympathy, guide tubes and toothpicks rather than elaborate prescientific diagnoses and skin-penetrating needles? You might be surprised that you can offer the same benefits with less risk. And if you are careful in recording the results, you might be able to publish and contribute to the scientific discourse rather than attacking it with double-standards.
WLU pretty well covered it, but I think this particular paragraph calls for a more detailed discussion:
Yes, your conclusion is wrong. This is because “considering the risk alone” is inappropriate. We must consider both the risk and the benefit, as the acceptable level of risk depends on how much benefit will be gained from taking the risk. As an extreme example, a 50% chance of death is entirely unacceptable if the benefit is minor pain relief, but acceptable if the benefit is avoiding a 100% chance of death.
Accordingly, we also must consider the benefits of medication (large) and the benefits of acupuncture (none beyond placebo convincingly demonstrated, and well-demonstrated to not be large).
Therefore, we see that the benefits of medication are larger than the risks, and therefore condone the use of pharmaceuticals. (This is of course a massive simplification, as each individual medication must be evaluated – and for each individual using it. But for our purposes here, it will suffice.) However, the risks of acupuncture (small but non-zero) are larger than the benefits (zero) and hence we cannot condone the use of acupuncture.