 Hi, welcome back. I'm Dr. Harriet Hall, and this is lecture 4 in a series of 10 lectures on science-based medicine. In the last lecture, I talked about chiropractic. In this lecture, I'll talk about another CAM that is widely accepted, acupuncture. When I say alternative medicine is medicine that hasn't been proven to work, someone always says, but what about acupuncture? It's been proven to work, hasn't it? Lots of people think so. After chiropractic, acupuncture is the next most widely accepted alternative medicine modality. It's not as popular as chiropractic. Only 6.5 percent of Americans have ever tried it. Most of the people who tried it apparently weren't very impressed. 22 percent of them only went once, and 70 percent of them went fewer than five times. A lot of mainstream doctors think there is good evidence to support it. Insurance pays for it, and the military is even teaching battlefield acupuncture to its doctors. Evidence-based medicine may accept it, but science-based medicine doesn't. The evidence is weak, inconsistent, flawed, and unconvincing to rigorous scientists. The whole body of evidence is compatible with exactly what you would expect to find if acupuncture didn't have any specific effects and was only a theatrical placebo system. The basis of acupuncture is the belief that an undetectable vitalistic energy called chi flows through the body. According to acupuncture theory, chi follows meridians and can be accessed at acupoints on those meridians. Illness is caused by obstruction to the flow of chi. Needling at the correct combination of acupoints is somehow supposed to improve health by somehow restoring the flow of chi. They're very vague as to how that might actually work. Scientists have never been able to verify the existence of chi, meridians, or acupoints. They are imaginary. In fact, almost everything you've heard about acupuncture is wrong. Here are some of the things you may have heard about acupuncture. You've probably heard that it's an ancient Chinese treatment method and that it involves sticking needles in acupoints. You may have been told that it's widely used in China. You may have heard that it works to relieve pain and nausea and that it works for other conditions like infertility. You may have heard that it can replace anesthesia for surgery. You may have heard that it's harmless, that it has no side effects, and you may have heard that it has been proven to work by scientific research. Every one of these statements is false. I'll explain. An ancient Chinese method? Every word of that is wrong. It's not so ancient. It may not even be Chinese and it's not a method, but many different methods. Acupuncture is said to be 3,000 years old, but the earliest Chinese medical texts don't mention it. The earliest reference to needling is from 90 BC and it refers to bloodletting and to lancing abscesses with large needles or lances made of stone and bone. Early diagram showed the location of blood vessels, not meridians. The first accounts of Chinese medicine reached Europe in the 13th century. They didn't mention acupuncture. Before the 20th century, acupuncture needles were commonly inserted at points of pain. A Frenchman, Sully de Morin, was the first to use the term meridian and to associate it with energy and that was in the 1930s. Ear acupuncture was invented by another Frenchman, Nojie, in 1957. That's what the Air Force is teaching for battlefield acupuncture. Acupuncture may not have originated in China. Historians now think the idea is originated in ancient Greece and traveled to China via the trade routes. And there's not just one acupuncture. There are many acupunctures. If anyone makes a claim for acupuncture, ask them which acupuncture. There are lots of them. There's the kind that just sticks needles in acupoints, but there's also electroacupuncture where acupoints are stimulated electrically, either with or without needles. Then there's acupressure, a kind of massage. In one system of acupuncture, gold beads are implanted under the skin. In electrodermal acupuncture, a biofeedback machine measures electrical conductance at acupoints and treats by electrically stimulating those points. And there's moxibustion, where burning mowward is placed on the skin at acupoints. Sometimes a permanent scar is deliberately produced. Sometimes the burning material is removed before the skin is damaged. And there's cupping, where a bulb with heated air is applied to an acupuncture point. And as the air cools, it creates a vacuum, which sucks up a glob of flesh, leaving an ugly red lesion on the skin. Sometimes they nick the skin so the cup fills up with blood. And there's auricular, or ear acupointure. A French doctor named Nogier invented this system all by himself in 1957. He imagined that the outer ear looked something like a fetus curled up in the mother's womb. He assigned parts of the body to corresponding parts on the ear, and he stuck acupuncture needles in those spots and convinced himself that it worked. There are no connections between the ear and those other parts of the human body. Nogier had a great imagination, but he was no scientist. This is the kind of acupuncture that is being taught to our military doctors for use on the battlefield. As a retired Air Force Colonel, I find that very disturbing. There are other systems of acupuncture that only deal with the limited part of the body. Scalp acupuncture, foot acupuncture, hand acupuncture, even tongue acupuncture. Here are some more variations, both with skin penetration and without. With skin penetration, they can use needles alone, needles with manipulation, needles with electrical stimulation, tiny gold beads implanted under the skin, or injection of homeopathic remedies at acupoints. That last one really takes the cake, injecting non-existent medicine into non-existent points. Without skin penetration, acupuncture points can be stimulated with light, sound, pressure, heat, electricity, electromagnetic frequencies, vacuum, color, or even just by waving hands over acupoints. No, I didn't make that up. Here's a non-needle option, SOG color puncture. The website explains that it uses colored light, infrared frequencies, ultraviolet frequencies, brainwave frequencies, sound and crystal treatments, as indicated by Curly Energy Emission Analysis to gently unlock and release emotional trauma and energetic blocks, which often underlie our illnesses. It assists the flow of spirit information in the body, making sure it does. Here's the silliest one ever, Tong Wren. I think of it as acupuncture meets voodoo. If you ever want to have a good laugh, look at videos of Tong Wren. I put a link to one of those videos in the course guide. You will see a whole room full of people rhythmically tapping a certain meridian on little acupuncture figurines with a little metal hammer. There is so much silliness surrounding acupuncture that a doctor created a successful hoax by saying he'd found acupuncture points on the buttocks that worked better than traditional acupuncture points. He was actually invited to present his research at an integrative medicine conference. It was no sillier than a lot of the other ideas they believed in, so they thought he'd made a real discovery. He had to decline the invitation and explain to them that it was a hoax. He published the story in the British Medical Journal. How many meridians are there? Take your pick. Different ancient texts list 9, 11 or 12. Or is it 14 or 20? Eight extra meridians got added somewhere down the line. And of course, some kinds of acupuncture disregard meridians entirely and only use points on the tongue, scalp, ear, or hand. How many acupoints are there? Well, again, take your pick. There were originally 365 because acupuncture was originally related to astrology. They chose a point to symbolically represent every day of the year. Now there are over 2,000 acupoints. In Korean acupuncture, there are 300, but they're only on the hand. In auricular acupuncture, there are only points on the ear. There were originally 30. Now there are 120. So what's left? Is there any spot left on the body that is not an acupoint in some system? Well, yes, there is one place that has no acupoints in any system. The male genitalia. What do you suppose that is? Now think about it. How do you think they made the discoveries and determined where the meridians and points were? How do you think they figured out which acupoints corresponded to which organs? How do you think they determined which set of acupoints were indicated for treating which disease? What would it take to verify all those correlations? Do you think they use statistics, control groups, and careful record keeping? I don't. Acupoints and meridians are imaginary. No anatomist has ever found one. Acupuncture is widely used in China. Turns out that's a myth too. The Chinese government tried to ban acupuncture more than once. The most recent attempt was by the nationalist government in World War II. Chairman Mao cynically restored it to respectability during his barefoot doctor campaign. He didn't believe it worked and he didn't use it himself. But he founded a convenient expedient to pacify the masses who couldn't afford the kind of Western medical care that he demanded for himself. Today, Western medicine is the treatment of choice in China. Traditional Chinese medicine and acupuncture are used mainly by the elderly and the poor. In Taiwan, only 6% of the population has ever used acupuncture. It seems it's more popular in California than in China. Some people believe surgery can be done with nothing but acupuncture for anesthesia. That's a myth too. New York Times journalist James Reston developed a appendicitis while covering Nixon's trip to China in 1971. The myth says that he had an appendectomy under acupuncture anesthesia. In reality, he had conventional anesthesia for his surgery. He did get acupuncture for post-operative pain, but he got regular pain medication right along with it. This picture allegedly showed a patient undergoing open-heart surgery under acupuncture anesthesia. There's something terribly wrong with this picture. The incision doesn't look like it's in the right place. And when the chest cavity is opened like this, the atmospheric pressure collapses the lungs and the patient can't breathe without assistance. Why isn't she intubated and on a respirator? Acupuncture is used for surgery in China, but patients also get local anesthetics, sedatives, and narcotics. I can tell you from personal experience that those are very effective all by themselves. The patients are selected by very strict criteria, and they estimate that acupuncture is only useful in about 10% of cases. And there are a lot of psychological factors involved. There's strong suggestions similar to hypnosis. Patients are motivated to please authority figures. They may have no choice because they can't afford real anesthesia. They may submit under threat of exile from job, home, and family. My friend Kimball Atwood is an anesthesiologist who studied the claims for acupuncture anesthesia. He asks, rather than being an important Chinese achievement, was acupuncture anesthesia more a form of torture perpetrated by a totalitarian government on its own citizens with the forced complicity of physicians? People claim that acupuncture works on animals, so it can't be just placebo. Yes, it can. As I explained in lecture two, pets can't tell you how much it hurts, so their response has to be interpreted by owners observing their behavior. And the placebo effect may be working on the owners. If they expect the animal to improve, that will influence what they observe, and the pet may respond differently because the owners are treating the pet differently or giving it more attention. And conditioning may be at work like Pavlov's dogs. Veterinary acupuncture textbooks show acupoints and meridians for a variety of animals. The diagrams for animals are simply transposed from those for humans without any attempt to study whether there are differences. Here's horse puncture. Note all the gallbladder points. News flash, horses don't have gallbladders. And then there's cow puncture, cat puncture, dog puncture, pig puncture, goat puncture, camel puncture, and my favorite chicken puncture. I suspect foul play. When I was in medical school, my professor of anesthesiology thought the gait theory of pain might explain acupuncture. But the research didn't pan out, and that theory has been pretty much rejected. Could the needles act as counter irritants, like hitting your thumb with a hammer so you don't notice your headache so much? Studies have shown that acupuncture releases endorphins, natural pain killers in the brain. But that doesn't really tell us anything. Endorphins are also released when people take placebo pills. And if you throw a stick for a dog to fetch, it releases endorphins in the dog's brain. The medical letter concluded acupuncture has been shown to increase endogenous endorphins, and functional MRI shows effects on central pain pathways. But it's not clear that the same effects couldn't be achieved by needling or electrical stimulation at non-acupoints. Acupuncture researchers have tried to verify that acupoints exist. They have looked at things like neurovascular bundles, trigger points, connective tissue fascial planes, reduced electrical impedance, enhanced migration of nuclear tracers. But the studies are flawed and inconclusive, and they contradict each other. And no agreement has been reached. Here's a couple of examples from those flawed studies. They've tried to measure electrical skin resistance, but it's inaccurate and unreliable. The measurements vary with factors like the shape and surface area of the electrode, dryness of the skin, local variations in skin thickness, scanning speed of the device, pressure placed on the electrode, the inclination of the electrode, the electrode gel used, room temperature and humidity. There have been at least three attempts to validate the concept of meridians using radio tracer injections. Early investigators concluded that an injected radio tracer was cleared by non-limphatic and non-venous pathways and suggested that this demonstrated the existence of acupuncture meridians. But a subsequent attempt to replicate this finding was unsuccessful, and the authors concluded that the radio tracer was removed along normal venous pathways, after all. In a 2000 article in a Japanese physiology journal, traditional acupuncture, moxibustion and a hemostat on the tail were compared for their ability to block pain and mice. The hemostat one, poor mice, so that would seem to support the counter irritant theory. But surely we don't want to relieve one pain by just by causing another. This is Ray Hyman, one of the founders of the modern skeptical movement. His categorical directive is, before we try to explain something, we should make sure it actually happened. Applying that to acupuncture, before we try to explain how acupuncture works, we should establish that it does work. So, does acupuncture work? Maybe all those chi and meridian explanations are nonsense, but acupuncture might still work anyway. Here's a list of some of the things acupuncture has been claimed to work for. You won't be able to read them all now, but the list is in your course guide. What's important to know is that acupuncture has been tested for all of these conditions, and it has only been shown effective for these two, pain and post-operative nausea and vomiting. And even for those two conditions, it may only work through placebo effects. Acupuncturists can cite any number of studies showing that it works, but you can find a study to support almost any claim, and there are plenty of other studies showing that acupuncture doesn't work. And remember, when a popular but ineffective treatment is studied, false positive results are common. And acupuncture studies are inherently flawed. Blinding is difficult. People notice when you stick needles in them, and double blinding is next to impossible. The best studies use sham needles with a sheath, where the needle just touches the skin and then retracts into the sheath like a stage dagger, so the patient can't tell if the skin was penetrated. But even that isn't foolproof, most patients can still tell the difference between touch and penetration. Here's the Cactus study. It was a pragmatic trial that compared usual care alone to usual care plus acupuncture. It was a poorly designed study, and it didn't use any placebo acupuncture controls. It concluded the addition of 12 sessions of five element acupuncture to usual care resulted in improved health status and well-being that was sustained for 12 months. Now when you add anything to the usual care, you're practically guaranteed to get some response, so a study without adequate controls is meaningless. Here are some of the graphs from that study. I know it's too small to read, but they measure four different outcomes like pain and the number of doctor visits. And the blue lines represent the acupuncture group, and the pink lines represent the control group that only got usual care. You can see at a glance that there really isn't much difference between the groups. Sometimes the pink line is on top, sometimes the blue line, and they stay pretty much together. David Cahoon has analyzed this study, and he concluded this paper, though designed to be susceptible to almost every form of bias, shows staggeringly small effects. It is the best evidence I've ever seen that not only are needles ineffective, but that placebo effects, if they're there at all, are trivial in size and have no useful benefit to the patient. He also said, this paper was published with conclusions that appear to contradict directly what the data show. The best designed most rigorous acupuncture studies with good sham controls have taught us these lessons. It doesn't matter where you put the needles. It doesn't matter whether you use needles or pretend to use needles. In one study, just touching the skin with a toothpick worked just as well. All that matters is whether the patient believes he got real acupuncture. If they got acupuncture but thought they didn't, it didn't work. If they didn't get acupuncture but thought they did, it worked just as well as if they got real acupuncture. A recent acupuncture study didn't use any placebo controls, and the authors explained why. We did not add a sham acupuncture arm to this study because superficial acupuncture, needling of the skin, sham acupuncture, needling of non-acupuncture points, and placebo acupuncture, needling with blunt tips that does not penetrate the skin, also provide an active therapeutic effect. Even light touch and caressing may activate the C tactile afferent nerve fibers that alleviate unpleasantness and reestablish patient's sense of well being. Therefore, neither minimal superficial sham acupuncture nor placebo needles may be regarded as placebos because they are not inert. I still can't believe I read this in a major medical journal. The author didn't even realize that he just completely destroyed any possible rationale he might have had for using acupuncture in the first place. You might as well just use light massage instead of an elaborate system of choosing acupoints and inserting needles. A more logical interpretation is that if real acupuncture doesn't work any better than sham acupuncture, real acupuncture doesn't work. There's a double standard here. If it were a drug that didn't work any better than a sugar pill, would they argue that the sugar pill is also an effective drug? The positive results of acupuncture studies are questionable. Reported benefits are likely due to the surrounding ritual, the beliefs of patient and practitioner, the non-specific effects of treatment like attention from a charismatic provider, long appointments, TLC, relaxation, and hands-on treatment. Remember, never trust one study in isolation. We have to look at all the research. Studies disagree. Some say acupuncture works, some say it doesn't. Next step, systematic review. Unfortunately, the systematic reviews disagree. Some of them conclude that acupuncture works, and some of them conclude that it doesn't. So the next step is to do a systematic review of systematic reviews. Edzard Ernst and two Chinese colleagues did exactly that. They systematically reviewed the last 10 years worth of systematic reviews of acupuncture for pain. 57 systematic reviews met the criteria they set for inclusion in their analysis. They found a mixture of negative, positive, and inclusive results. There were only four conditions for which more than one systematic review reached the same conclusion. In three of those cases, they agreed with each other that acupuncture didn't work. There was only one case where they agreed with each other that it did work, and that was for neck pain. In their discussion, they explained how inconsistencies, biases, conflicting conclusions, and recent high-quality studies throw doubt on even the most positive reviews. A small excess of positive trials is predicted by poor study design, publication bias, and noise in the data. Here's the double standard again. What if a systematic review of systematic reviews for a pain pill gave the same disappointing results as Erzt's review of 57 reviews? What if three reviews showed that the pain pill didn't work, and only one review showed that it relieved neck pain but didn't relieve pain anywhere else in the body? I don't think many doctors would want to prescribe that pain pill, and I doubt if it would get FDA approval. Another myth is the idea that acupuncture is perfectly safe. Well, maybe it is most of the time, but Erzt found 95 published cases of serious adverse effects, including infection, pneumothorax, and five deaths. You may remember seeing this in the news back in 2011. The ex-president of Korea was hospitalized for a cough, and he ended up getting major surgery to remove a 6.5 centimeter acupuncture needle from his lung. Acupuncture websites list a number of contraindications to acupuncture. You shouldn't get acupuncture if you have metal allergies, bleeding disorders, or on anticoagulant drugs, or have a skin infection. A popular pregnancy book warns against stimulating acupuncture points on the ankle because it can produce uterine contractions and bring on premature labor. I'm not so sure about that one. Some acupuncturists have been convinced by the evidence that it's no better than placebo, but they argue that if a placebo makes a patient feel better, we should use it. That sort of sounds reasonable, but there are reasons not to use placebos. I covered these in lecture two. One of the biggest reasons is that doctors are not supposed to lie to patients. If a patient finds out that a doctor is fooled him with a sugar pill, he may never trust the doctor again. Here's another big reason not to use acupuncture as a placebo. If patients think they feel better, but they aren't actually getting better, that can be a real problem. In this study of patients with asthma, the bars from left to right represent the reduction in symptoms with four treatments. The bar on the left is for an albuterol inhaler, a proven pharmacological treatment. The two middle bars are for placebos, a placebo inhaler and a sham acupuncture treatment. And the bar on the right is for a control group that got no intervention at all. The patients reported feeling almost as much better with the placebos as with the real drug. They felt better, but were they really better? No, they weren't. This graph shows what they found when they measured the patient's lung function with pulmonary function tests. The only group where lung function improved was the albuterol inhaler group. Lung function in the placebos and no intervention control group didn't improve. For asthma, this could be deadly. If patients think they're getting better but their lung function is unchanged or actually getting worse, they might not go to the emergency room in time. Patients with asthma who don't realize the severity of an attack are more likely to die. I mentioned the book Tricker Treatment in lecture two. Here's what it said about acupuncture. While there is tentative evidence that acupuncture might be effective for some forms of pain relief and nausea, it fails to deliver any medical benefit in any other situations and its underlying concepts are meaningless. Note the qualifications in red. When the medical letter reviewed acupuncture in 2006, they concluded acupuncture alone has not been shown in rigorous duplicated studies to benefit any defined medical condition. The Center for Inquiry issued a position paper on acupuncture in 2010. They said it is becoming increasingly clear that the placebo effect accounts for much of the optimistic research on acupuncture published between the 1970s and 1990s. With the development of sham needles during the past decade, needles which retract like stage knives, the researchers have been able to more accurately assess the therapeutic effects of acupuncture. The result has been a complete unraveling of nearly all acupuncture claims. The bulk of recent research strongly tends towards the hypothesis that acupuncture's positive effects are mainly due to a built-in expectation. Here's the bottom line. There is no credible evidence that acupuncture works for anything other than pain and postoperative nausea and vomiting, and it may not really work for those. Any improvement is likely due to the non-specific treatment effects, and it has nothing to do with acupuncture points or needles. Acupuncture is no more than a theatrical placebo. That quotation is from David Cahoon in Steven Novella's article in the journal Anesthesia and Analgesia. When a treatment has been extensively studied for decades and the evidence continues to be inconsistent, it becomes more and more likely that the treatment isn't truly effective. Proponents of acupuncture always say more studies are needed. Really? After 3,000 trials, isn't it about time to stop? Research money is scarce, and it would be better spent on other areas that are more likely to be productive. When I was growing up, whenever I told my dad I'd changed my mind about something, he'd say, if I had a mind like that, I'd change it too. I heard from an acupuncturist who changed his mind. He read an article I wrote in Skeptic Magazine, and he emailed me to say, your article about acupuncture made me angry. I thought that you had to be wrong, because of course I had heard about so much research about the evidence for acupuncture. You got me to actually look at the literature instead of just knowing that it exists. In trying to prove you wrong, I proved myself wrong. One textbook made a claim that acupoint such and such is good for stimulating the pituitary to produce anti-diuretic hormone for patients with polyuria. I thought, wow, that's really interesting. I wonder where the reference is. I looked for a reference, and I was struck by the fact that there were no references at all anywhere in the book for any of the claims. I have gradually let go of the belief that acupuncture has any basis for treating anything, and he said he was going to quit acupuncture and find a more science based way to make a living. I wish I could take the credit, but I didn't convince him. All I did was give him an opportunity to get angry and to convince himself. Not many people could do that, but at least there's, now there's one more convert to science. So acupuncture is clearly not science-based medicine. In the next lecture, I'll talk about homeopathy, which is not only not science-based, but is monumentally silly.