Placebo isn't a dirty word

When a fake treatment works as well as, or better than, the real thing, that's usually when medical researchers go back to the ol' drawing board. To which Harvard's Ted J. Kaptchuk asks, "Why to the who what now?" Kaptchuk is pushing placebo, not as a cure-all, but as a way to sooth pain when "real medicine" doesn't work.

Though recurring tummy aches from irritable bowel syndrome are among patients' most common complaints, drugmakers have had trouble coming up with a safe and effective treatment.

[Kaptchuk's] magic cure: fake acupuncture delivered with lots of warm talk from a sympathetic acupuncturist—but no needles. In a trial of 262 patients with severe IBS, 62% of those who received the fake treatment got better, according to results published in the British Medical Journal [2008]. By comparison, only 28% of a control group of patients put on a waiting list saw their symptoms improve markedly. A third group who got the fake acupuncture, but without any warm talk, showed in-between results: 44% improved.

Forbes: The Nothing Cure

This link, btw, came from Wired's Steve Silberman, who wrote his own amazing story about the power of placebo last year.


  1. Sigh. What could go possibly wrong with a plan where doctors lie to patients to make them feel better?

  2. Interpersonal psychotherapy was developed as a placebo treatment for therapy studies. It turned out to be efficacious in its own right, with rates of improvement around the same as cognitive behavioural therapy. Designing placebos and controls for therapy treatments is hard, as someone listening to a person for an hour can have quite a profound effect, which seems to be the action that the fake acupuncture had on people in the above study.

  3. @2: Exactly.

    There are massive ethical issues with doctors giving something to a patient, knowing that there is no active ingredient, and lying to make them feel better. I wouldn’t want my doctor doing that to me!

    Also, the placebo effect isn’t nearly as mysterious and magical as a lot of people make it out to be:

    Drug companies should rightly be trying to minimize it in drug trials. They aren’t trying to measure the combined effect of what is mostly statistical and reporting errors and conformation biases; they are trying to measure how well their drug works!

    Disappointed in this link. I guess that’s what you get for getting your science from Forbes magazine.

  4. I think scientists call this “woo.” If it reduces pain, great. Just don’t even think about charging me $100 per bottle.

    1. I think scientists call this “woo.” If it reduces pain, great. Just don’t even think about charging me $100 per bottle.

      Clinical research suggests that expensive placebo treatments work better than cheep ones. Seriously.

  5. When the lack of real medicine cannot cause more serious problems then I only see some ethical issues. Namely the problem of deception. If you have to lie to the patient you have broken core rules of ethics.

    And of course the problem of further harm applies mostly to pseudo-medical stuff like crystal healing and homeopathy where some practitioners claim the abilities to cure almost anything, but have failed to show these miraculous skills. They still achieve placebo effect in many instances, and sometimes a little above that (personal attention from a “doctor” IS a form of treatment). That just isn’t very good compared to scientific and tested medicine.

  6. This treatment actually consisted of two mechanisms: the placebo and social support.
    While both elements seemed to result in an improvement, using the placebo still seems questionable.
    The placebo is, regardless of the intent, a lie. Using placebos as a therapy could endanger the trust in their doctor or doctors in general. This trust is crucial, however, to the function of a doctor.

    Lost trust might result in less compliance concerning medication. Why take some pills, if they don’t work anyway?

    It might be a better idea, to stick with social support only.

  7. We started marketing a high grade placebo which we branded as Obecalp.

    We received an enormous amount of terrific international press about our products!

    One of the best defenses for utilization of placebo must be the following peer reviewed paper called: “The ethics of the placebo in clinical practice” peer reviewed and published by the BMJ’s Journal of Medical Ethics and paid for the the US Government.

    Click here for link:

    We are developing another complete line of 8 placebos to be sold soon.

    We fully believe you should discuss any and all medical options with your medical care giver.

    (I wonder if any provisions were made for placebo in the new health care reform).

    Efficacy Brands LLC

    1. This disappoints me, as I’ve always planned to retire on the sales of my new pain-fighting drug, Placebex.

  8. As a cure for an actual physical illness that is quantifiable, I just say no to placebos. In the case of pain management; however, why not give it a shot? The only negative effect would be the pain that you feel on a daily basis anyway.

  9. 1) Before anyone rushes out to spend money on ‘alternate’ remedies, check out

    2) The placebo effect is quite well known, even if not completely understood. Sometimes controlling pain (which is where most of the placebo effects are found), it’s just a matter of complaining less – both to others and to yourself.

    3) Harvard? Hmm, wasn’t that Bush’s Alma Matter?

    /just sayin…

    1. would be a bit more believable if they listed the millions of people killed by drugs like hormone replacement therapy, Vioxx, etc. It’s clearly a shill site. Yes, home birth can be dangerous, but it’s not nearly as dangerous as an unneeded c-section. For each horror story for a homeopath, I can give you an equivalent one of an MD whose handwriting was so bad that someone got the wrong medicine, who didn’t check for allergies, who did check for drug interactions, etc. I’ll worry about the chiropractor when you start worrying about Vioxx — which per the FDA caused between 88,000 and 139,000 heart attacks, 30 to 40 percent of which were probably fatal, in the five years the drug was on the market.

      1. Thalia, I can actually give you more examples of problems within the real medical community than from the homeopathic community.

        Partly because there is NO internal policing in homeopathy. And because they do not get sued if people in their care die. There is no system for keeping records and they do not share any findings with each other.
        I once read a homeopathic “journal” (opinion piece magazine, actually) where one of the fanciest of the HP’s threw about wildly inaccurate ideas about quantum effects in HP without understanding quantum physics and of course without any evidence.

        Do doctors screw up?
        We all screw up, and the higher the pressure the more we screw up.

        Does homeopathic treatment work?
        That has NEVER been proven. Not for any one of their treatments. They were very proud once when it seemed like a allergy treatment was getting about 30% cure ratio. “Ahh, see, we do science” they said. When it was then shown that the science was faulty (low internal validity, multiple uncontrolled confoundings) they changed it to “science just doesn’t understand what we do”.
        My sister in law is in that crap so I’ve had some opportunities to examine the journals, books and science. It doesn’t work.

        Pharmaceuticals are dangerous, no question, but side effects must always be balanced against what is being cured. Anything with an effect can have side effects, even a proper massage can affect blood clots.
        Chiropractors can kill you. This was discovered not due to their internal policing, but because regular doctors noticed something strange (young people dying of blood clots in brain) and decided to investigate.
        Turns out that if the crick your neck (which is quite common) that often tears small veins in your neck, causing blood clotting. A few days later those blood clots can become unseated due to neck movement and depending on luck the clot either breaks down and gets cleaned out OR gets lodged in your brain and you die. This was discovered by Canada’s public health system.

        Everything with an effect has side effects. It’s then just a case of documenting those side effects and judging whether that treatment should go onto the market or in many cases stay on the market, research doesn’t stop just because it has reached market. In real science, anyway.

        1. The risk of dying from a cervical artery dissection is orders of magnitude less than the risk of dying from acetaminophen or NSAIDS. The incidence of adverse events from a cervical adjustment is well researched, and is considered to be less than one in five million. There is a similar risk of dying from a cervical artery dissection after seeing an MD, so causation between adjusting and the dissection is not strong. I do not believe there is a surgical technique with a risk as low as one in five million. When it happens, it’s never a news story. Increase in chiropractic utilization in the general population also did not result in an increase in dissection.

          This is a excellent review of current literature:
          Spine: 15 February 2008 – Volume 33 – Issue 4S

          And if you’re not in the field, “Spine” is largely a medical journal, not a chiropractic one.

          Our malpractice insurance, which is a decent estimate of risk of adverse outcomes, is an order of magnitude or two lower than that of an MD.

          I will NEVER tell a patient about the risk of death from surgery. When asked directly about surgical risk, I tell the patient that that’s part of a surgeon’s day job to explain it to you. Why? Because there may be a time when that’s their best option. I would feel shitty if I unneccesarily increased their fear, or dissuaded someone from unneccesarily avoiding something that might end their pain. When appropriate, I refer patients to well-regarded neurosurgeons/orthopaedic surgeons, who are conservative in who they choose to operate on. I am also aware of patients who’ve been told by well-meaning friends and doctors that they may die if they come see me. Not helpful. And I haven’t killed anyone yet.

          I’m also aware that there’s several surgical techniques that have an element of placebo. When a patient does seek surgery (or a cortisone injection/epidural injection or other pain management technique), I will tell them of other patients that I’ve known who’ve had good results with that technique, in the hopes that the “placebo” effect is strengthened.

          There are people who’ve gone to multiple MD’s and PT’s who’ve then come to me and gotten relief from pain. If it was solely belief, I feel they would have gotten placebo relief from the doctors they had faith in to go to first. If you decide, evidence be damned, that the important take-home message about chiropractic is that we kill people, then people listening to you may unneccesarily stay in pain. Some may use techniques with a higher risk. Keep researching and keep your mind open.

          1. Dear Snig
            I said nothing about chiropracty not working. Only that there are risks which MOST chiropractors do not tell their patients about.
            In fact, I have gone to chiropractors myself, as I am pretty certain of some of the effects of their work. One chiropractor then did the neck adjustment thing. I asked about risk and the answer was that there was NO risk. This person was the head of the Icelandic chiropractors association (trained in USA like so many). The best known chiropractor in Iceland also maintains a “no risk” attitude and claims to fix problems (unlike traditional medicine which they claim works only with symptoms).
            But the fact is that most users of chiropractors go regularly to be adjusted, after an initial treatment which may be quite regular.
            That’s not fixing or healing, that’s symptom management.

            And the fact remains that the unusually high number of deaths which the Canadians caught went completely unnoticed by the chiropractic practitioners. Very limited information was forthcoming aboutthe diagnosis of their patients, only what adjustments had been performed. All medical information had to come from coroners and regular medical professionals.

            I have two main gripes with chiropractors. Misrepresentation/misunderstanding of what they actually do (energy lines ring a bell?) and misrepresentation of the risk.
            Then of course the profession has some utter quacks as well who adjust young children and such (with unhardened skeletal systems).

            As for the danger of pharmaceuticals, yup, even aspirin (which is a chemical from the bark of the willow tree) can be deadly. Over the counter medicines are seriously underestimated by most people. Paracetamol (panodil and countless others) is one of the biggest chemical killers in both UK and USA. This is well known within the medical community, but trying to get people to understand the danger is harder than it sounds! I’ve been trying for years now.

            You sound reasonable, which would make me quite interested in chatting with you.

          2. You sound reasonable too.

            It was definitely wrong for the chiro to adjust your neck without getting permission. I don’t do that.

            I appreciate that you are mentioning that there is a risk to medication. While many MD’s are responsible, others don’t see chronic use of medication as a real problem.

            I’m annoyed that while you mentioned “medications are dangerous”, you introduce chiro with “chiropractors kill people”. As do surgeons and dentists, but that’s not really their defining feature.

            I really only work with people in pain. My goal is to have patients symptomatically improve so that they can fire me. While there are “wellness” practices like you talk about, there’s plenty of chiros who just want to fix pain.

            Energy lines don’t actually ring a bell. I tell people (short form), that one of the reasons some people have neck or back pain is that their joints are not moving well, and by adjusting the area we are doing a vigourous stretch that may restore mobility to the region. I also do physical therapy, so work on stretching and do focused massage. I teach people stretching. If a patient hasn’t improved in 2-4 weeks (depending on diagnosis) I figure something is not working, and reassess my diagnosis and frequently refer out to orthos or neuros. I don’t bash medicine or MD’s.

            I generally tell people that the risk of a cervical adjustment is about the same as taking NSAIDs or tylenol. It’s actually a little less risky than that. This is more warning than is generally given for the general populace when they buy an OTC. There is a theory that CVA’s occur in individuals, they have neck pain, and are essentially in the process of stroking out. Whether they see an MD or a DC, unless this relatively rare syndrome that looks pretty much like very common garden variety neck pain is accurately diagnosed, they may well die. Obviously, everything has some risk. I mention the less than 1 in 5,000,000 estimate, but the doctor you mention may not be too far off in stating there’s no significant risk. If you’re not getting your neck adjusted (the majority of my patients) the risk is pretty minimal.

            I don’t believe their was a Canadian Chiropractic massacre. I’m aware of lawsuits and a study around it, but don’t believe it’s quite like you said. Here’s a shorter review on CVA than the one I mentioned earlier:

  10. I don’t see what’s new here: we’ve had research documenting the efficacy of placebo for decades, which is why we have double-blind clinical trials before claiming any treatment is effective in its own right (i.e., more effective than the placebo control).

    When I trained as an RN, I learned to use both. “This is morphine, and it’ll make you feel better in the next 5 minutes.” The expectation of relief enhances the effect of the medicine. Both effects are real, and they add up.

    Of course if I ever lied to a patient, saying I was giving them a treatment when I wasn’t, I would hope I would have had my license yanked. Which is what should happen to Kaptchuk. In the U.S., research involving human subjects is usually vetted by a committee charged with making sure it passes ethical standards (usually called “Committee D”). You would think if a researcher said “I want to study the efficacy of lying outright to patients,” the committee would have turned thumbs down. Apparently Harvard’s Committee D was asleep when Kaptchuk’s study was considered.

  11. At some point you have to wonder how much the anxiety over having things like IBS contributes to its continued existence.

  12. Worth noting, this article is talking about placebo as a cheaper alternative to pain treatment when conventional medication doesn’t work. Acknowledging the benefits of placebo for pain relief and talking about how it could be used ethically is not the same thing as magic crystal healing woo.

    My doctor, for instance, (the first doctor I’ve ever had who was really dedicated to evidence-based practice, btw…he rocks my socks!) has told me that, based on placebo research, he chooses to spend time chatting personally with patients and slowing down the pace of an office visit. Evidence shows things like that make people feel better, so why not do them?

    1. Yes it is exactly the same as magical crystal healing woo. The placebo has no physiological effect, it is mostly patient and experimenter bias, you think you are going to get better so you tend to subconsciously put less emphasis on the bad periods and more on the good. That’s why the placebo effect works so strongly on things like nausea and pain, things that require the patient to self report and rate against previous experiences.

      Combining what we know about the placebo effect with legitimate treatments will improve the way a patient feels about the treatment and that is an excellent thing and something that we should encourage but giving a patient a fake treatment, lying and misleading them in the hope that they will overlook real symptoms is the very worst idea I have heard real doctors express.

  13. One real harm from placebos is cost. To work, people have to think a placebo is medicine, so it has to be priced comparable to similar treatments. There are some real issues, though, about charging someone for something fake.

    Before most people learned what it is, snake oil truly could work as a placebo. Even so the people who sold it are not remembered fondly for the service.

  14. @Paul R: just as a matter of fact, President George W. Bush went to Harvard Business, which is very different than Harvard Medical. One is a money making machine for the university, the other is a legendary name in research. One’s alumnus should not affect the other’s reputation.

    While it is certainly unethical to use lying as a form of medical treatment, it nevertheless is a phenomenon worth a further look. The Forbes article, in typical layman science reporting, has jumped to some rather extreme conclusions based upon limited research. But that should not be why such research is dismissed; new treatments are always coming from the most unexpected places.

    However, my major problem is that traditionally any results generated by placebos is treated as noise in the experiment. It’s a good control, in that it accounts for the role of the action of being treated as a separate variable from actually being treated. Therefore looking for meaning in placebos strikes me as a sort of data mining. Kaptchuk and other researcher may find trends, but that doesn’t necessarily mean they carry all that predictive power, much less treatment.

    1. Duly noted, grphiw, duly noted.

      Though I’m not sure that most doctors would agree that ‘lying as treatment’ is unethical. (For the record, I never want to be lied to by my physician…)

      In Canadian law, it’s well established that your doctor doesn’t have to tell you the truth.

  15. There is a real and distasteful paternalism in lying to your patients to make them happy and leave your office.

    More worrying still is the opportunity cost for the patients. The opportunity to have their condition and treatment options thoroughly evaluated. The opportunity to engage with an intervention that actually does something to mitigate their situation.

    We know in the field of Tinnitus and pain management that directive counseling which demystifies the condition and empowers the patient with effective cognitive and practical techniques can significantly reduce perceived suffering. No lying required.

    There is an additional danger with refractory conditions that the placebo touted stops working and the integrity of the caregiver is damaged.

    If there is an indirect effect that is understood by the caregiver to be a placebo sure don’t knock it, but it is unethical and irresponsible to engineer such a situation.

  16. Yeah, Maggie, I totally agree with that.
    And a doctor who takes time for you is almost magical, getting better results in many cases (shown in experiments).

    The only big issue is the ethical one, because the placebo (cheap as it is) stops working when patients KNOW that it is a placebo. And as Anon#15 points out, price is a factor. This has also been shown, that our cognition, our understanding of the world seems to impart more efficacy on pricier treatments even when they contain the same pills as the cheaper meds.

    In an experimental environment the placebo effect works ok, that does not transfer the effect to the real world, and doctors are bound to tell the truth (or lose their license, and all patient trust) which makes this impossible in any ethical way.

    I just mentioned the crystal woo because too many people (including a homeopath I know) try to tout the placebo effect as a PRO for what they’re doing! Incredible, don’t you think?

  17. First rule of being a doctor – primum non nocere – first do no harm. Lying to people is not nice, and other posters list the harms well above.

    There is no doubt that a solid, warm therapeutic relationship is powerful and helpful for patients. As an oncologist it’s what I rely on as my last-line, when the drugs don’t work anymore, and I mention this frequently – if the chemo doesn’t help control the cancer, we’ll focus on looking after you and relieving your symptoms.

    And to be controversial I think any benefit seen with scientifically unsound ‘therapies’ like homeopathy and naturopathy is simply due to the relationship between practitioner and patient. And if people feel better, awesome, love it (just don’t try to tell me that a 30x dilution of Tabasco sauce cures cancer).

  18. Reminds me of a M*A*S*H episode: Season Six, Episode Twenty-Four, “Major Topper.” The unit runs out of morphine and decide to hand out a placebo.

  19. Placebo isn’t a dirty word: it’s mostly a statistical effect.

    In other words, most of the ‘effect’ observed is a product of the research methodology: suggestibility, eagerness to please bollixing up the self-report process. There’s just no there, there.

  20. Maybe I’m the odd one out but I’m much more worried about my doctor not telling me about the harmful side-effects of the actual drugs that are being prescribed. At least the placebo cannot harm the patient.

    No one advocates using a placebo when there is an actual medicine that works. But given the number of serious disorders that have no medical solution, I don’t see the downside of placebos.

    To deal with the consent issue, it can be simply resolved by addressing it up-front, at the first visit. “I will provide you the best treatment that I can. Sometimes, that might be a placebo. Do you accept that option.” Most of us would say yes. Opting out is the patient’s choice.

  21. If I have a problem, and a placebo is likely to help, then give the the damned placebo.
    That said – I’d generally like more normal treatment as well.

  22. It seems fairly clear that the key here is the social interaction and support.

    So, rather than saying “we should/shouldn’t be using placebos”, how about we do a study on a healthy, well integrated community where its members care for and look after each other?

    Homoeopathy’s effectiveness appears to based on the same effect. Homoeopaths take the time to talk to and listen to their patients too.

    You may find that the need for any kind of treatment at all is reduced when people feel safe and secure and loved, rather than like livestock to be milked of their money.

  23. Lying to your patients is unethical. But if you can get past that teensy problem you have to deal with this: The placebo treatment might cure that headache or whatever minor thing is going on. And I don’t doubt pain-relief would come of it. But what do you say when that patient asks for the acupuncture to treat cancer? It can and has happened. Do you tell the patient “oh yeah, that worked then for bullshit claim A but it would never work with bullshit claim B”?

  24. Folks, the key here is he’s talking about IBS which isn’t a disease, it’s a syndrome. It’s no wonder that 62% of people who received the treatment got better…they relaxed. True relaxation “cures” IBS.

    So there is nothing fake about it even though they were told it was accupuncture and the needles never went in.

  25. That’s not fixing or healing, that’s symptom management.

    Like all classes of drugs except antibiotics?

    Also not true. I used to have frequent, severe neck and sacral pain. After about a year of chiropractic, it was occasional, mild pain, even without further treatment. Since misalignments can stem from trauma, a single adjustment can fix a problem. And with ergonomic teaching, gait training, etc., even problems stemming from chronic causes can be permanently fixed.

  26. Yeah, Antinous, exactly like many pharmaceuticals. But the difference is that many alternative therapies claim to do something else, which is untrue.
    You would be the only one I’ve heard of who doesn’t need an tune-up regularly. Even if your muscles have finally been relaxed to the point of you moving properly (and you think about yourself better than you did previously, I would wager) you still probably need some treatment once in a while.
    This is excellent, of course, and I can say for chiropractors that I had severe back pain due to an injury many years previously which was then aggravated by bad working conditions. I kept the pain down using increasingly strong painkillers but decided that enough was enough. A chiropractor helped me get mobile and after that I was able to manage it by the right type of sports (rock climbing in my case). No more drugs. YAY!

    But it isn’t fixed. The problem still remains. Now I just have much better tools to manage it.

    Saying that cervical adjustment has very low risk, or roughly comparable to tylenol may not be such a good idea. Tylenol (paracetamol) is the number one pharmaceutical killer in the world, precisely because its danger is so hugely underestimated (liver toxicity).

    But you seem to do normal treatments which is excellent. We need to try to minimize the drug use, especially if there are safe alternatives which actually have an effect.
    Then it’s time to get rid of the kooks and make sure that going to alternative practitioners is safe and certified as having an effect (chiropractors wouldn’t be scared of having to prove an effect, some other alternative therapies wouldn’t want to have to do that)

    You can google the energy lines/chiropractic stuff but here’s an example (, I walked out of one office where the chiropractor started talking about miraculous wonders with a very high woowoo factor. The problem is that some alternative therapists just take it on faith that all alternative therapies work, and that “traditional western medicine” poisons the body.

    The way I see it is, if you work within your specialized field and become good at it then you can work WITH all the other medical professionals. Which you seem to do.

    I’m sorry that this text reads as if written by a drunkard, but I am insanely tired and am going to rest a little now :)

Comments are closed.