A tonic for sceptics

August 29th, 2005 by Ben Goldacre in alternative medicine, bad science, homeopathy, placebo, statistics | 40 Comments »

Ben Goldacre
Monday August 29, 2005
Comment, The Guardian

Sceptics, and the placebo effect, are easily misunderstood. Since I’ve made a modest second career out of rubbishing alternative therapies (or rather the pseudoscience of the claims behind them), you might expect me to be pleased with a new analysis of 110 placebo-controlled randomised trials of homoeopathy, published in the Lancet, showing there is no evidence that homeopathic tablets perform any better than placebos. Obviously, it’s an important and useful finding. But it misses the mark.

Article continues
The placebo is arguably the most interesting phenomena in medicine, because it goes far beyond the effectiveness of little white sugar pills, into the cultural meaning of treatment. It has been shown, for example, that green sugar pills are a more effective treatment for anxiety than red sugar pills, because of the cultural meaning, we might parsimoniously assume, of the colours green and red.

Likewise, studies have found that salt-water injections can be a more effective placebo treatment than white sugar pills – not, I might add, because there is anything particularly useful about salt water injections, but because the ceremony of performing an injection is a far more invasive, authoritative and dramatic intervention.

It gets far stranger. A placebo operation in the 1950s was found to be as effective for the treatment of angina as the real operation it was being compared with. Reading the paper 50 years later, the most striking part is the discussion section, where they quietly drop the operation and nobody stands up to point out the incredibly strange discovery that a placebo operation works for anything, let alone angina.

Branding, of course, is the key to the efficacy of little white sugar pills. Marketing, after all, is nothing if not engineered cultural meaning. A four-way comparison among sugar pills and aspirin, in either unbranded aspirin boxes or packaging mocked up to look like the Disprin brand, showed that the brand-name packaging works, because of the huge wealth of cultural background material – the adverts, the word-of-mouth endorsement, the childhood experiences – that packaging plays on. The change in packaging had almost as big an impact on pain as whether the pills actually had any drug in them.

The implication for rationalists, who reach for generic, unbranded medications like aspirin and ibuprofen in preference to Disprin or Nurofen, is clear. It’s perfectly rational to believe that expensive Nurofen is more effective than cheap unbranded ibuprofen, even if they’ve both got the same active ingredient – but only, in a peculiar tautology, if that’s what you believe.

This, of course, is the key also to alternative medicine: homeopathy is what you might call a “complex intervention”, rich in cultural meaning and drawing on such attractive contemporary ideas as individualism, patient empowerment and personalised healthcare. But all alternative therapies also offer something very much missing from modern medicine, the idea of containment of symptoms.

Transparent modern medics often say: “I don’t know what the cause of your problem is. This might make it better, but it might not, and it might have these side effects.” They sometimes follow this with: “What do you think?”

Enter the alternative therapist, who understands your problems whatever they are, who is privately employed and has time to listen, who has an answer and who gives a complicated (often wilfully obscure but always authoritative) explanation of what is going on, maintaining the power imbalance in the therapeutic relationship with his or her exclusive access to arcane knowledge. If that’s not old-fashioned medical paternalism, I don’t know what is, and the paradox is clear: while modern medicine, without even pausing to discuss the question, has championed patient autonomy and informed consent – and thrown the placebo effect out of the window – the market has shown that the old paternalism, in a new guise, is still very popular.

Whether mainstream medics would want to go back to the old ways and embrace the placebo-maximising wiles of the alternative therapists is an easy question: no thanks. The didactic, paternalistic, authoritative, mystifying mantle has passed to the alternative therapist, and to wear it requires one thing most doctors are uncomfortable with, dishonesty.

So the fact stands, not even slightly mocking us, that in many cases homeopathy does seem to help, as a complex intervention, laden with branded cultural meaning, at least better than “doing nothing”. It is no better than placebo, because it is placebo, in all its rich glory.

But the homeopaths themselves can never admit this clear, compelling, evidence-based and parsimonious explanation: they need the memory of water, the power of arcane knowledge and all the rest. Homeopathy and the others can only work as long as their myths and meanings survive, and so medics, alternative therapists – and smarmy sceptics – will always, mercifully, be in business.

· Ben Goldacre is a doctor and writes the Bad Science column in the Guardian

If you like what I do, and you want me to do more, you can: buy my books Bad Science and Bad Pharma, give them to your friends, put them on your reading list, employ me to do a talk, or tweet this article to your friends. Thanks! ++++++++++++++++++++++++++++++++++++++++++

40 Responses

  1. MostlySunny said,

    August 29, 2005 at 11:44 am

    “It has been established beyond doubt and accepted by many researchers, that the placebo-controlled randomised controlled trial is not a fitting research tool with which to test homeopathy.”
    -Spokesperson from the Society of Homeopaths – quoted on the BBC webpage story about the Lancet reports

    um, what would they consider a “fitting research tool” – blind belief?


  2. amoebic vodka said,

    August 29, 2005 at 12:16 pm

    We suppose that just the fact that a GP might suggest homeopathy to a patient they have decided does not need a prescription would have quite an effect on the belief necessary for the placebo to work. We also wonder what level of skepticism is required before the placebo effect goes away. Sigh. Being a skeptic sucks.

  3. Administrator said,

    August 29, 2005 at 1:37 pm

    the society of homeopaths is great.

    this cosy little quote is from their code of practise:

    “Members must present a united front to the public and should not imply criticism of colleagues, either in writing or before clients or the general public.”

    they also put out an excellent leaflet series, including this one on travelling with homeopathy that includes excellent advice on how to protect your remedies at security barriers in airports: “You can protect them by using a lightweight lead-lined bag of the type sold for photographic films, or carrying them in your pocket.”


  4. Frank said,

    August 30, 2005 at 11:04 am

    But if belief plays a part in pain relief as important as the drug itself, what are the implications for trials of new drugs? Should non-sceptics be barred from trials in case they mask the results?

  5. Ben Goldacre said,

    August 30, 2005 at 11:06 am

    Doesn’t matter as long as trials are blinded…

  6. Guillaume Rischard said,

    August 31, 2005 at 3:20 pm

    The spokesperson wrote a funny letter to the Guardian too – I can’t find it at the moment. He says placebo-controlled randomised controlled trial are not fitting because homeopathy is more about the trust relation between the doctor and the patient, and the treatment of the patient as a whole taking all his symptoms into consideration.

    It sounds a lot like what you write about paternalism in your post, but let us, for one second, admit this for the sake of argument. How would this make a proper trial impossible? A homeopath would see patients, do his thing with them, and send them to get their sugar pills. The person who gives out the pills would randomly select half of the patients, and give them a placebo every time they are sent to him by the homeopath. The other half of the patients are given the homeopathy that was prescribed.

    We say the group that is being given the placebo should see the same improvement, on average, as the group given homeopathy. He says the group that is being given homeopathy should be better off. I say we challenge him in a duel with this – whoever turns out to be wrong will commit suicide by ingesting a large quantity of homeopathy.

  7. amoebic vodka said,

    August 31, 2005 at 9:48 pm

    Ingesting a large quantity of water a homeopathic remedy is dangerous:


    Don’t try this at home!

  8. amoebic vodka said,

    August 31, 2005 at 10:00 pm

    Sob. Who stole our html tags on the previous comment?

    More seriously, we thought that the appointment with a homeopath is supposed to be as vital as the remedy. In which case, your suggestion wouldn’t make them happy. Unless it showed a difference, maybe.

    Then we wondered, if seeing a homeopath is supposed to be so important, by their own logic over-the-counter homeopathic remedies are not very effective.

  9. Stuck as Skeptic said,

    September 8, 2005 at 4:26 pm

    I’m a skeptic and I’m stuck.

    Both a friend and I have eczema, a non-life threatening but chronic skin disorder.

    My friend believes in homeopathy. She went to a homeopath and got a nice little bottle of “medicine”. Her eczema has gone away most likely due to the placebo effect.

    I can’t go because i’m a skeptic. But when my body is covered with rashes, I seriously wish I wasn’t.

    There’s ethical dilemmas galore here. I have not told my friend that she got better because of the placebo effect. If i do i’d be putting her in the same boat of suffering that i’m in.

  10. Morag said,

    September 8, 2005 at 11:57 pm

    Look, how could “the placebo effect” cause eczema to get better? If it did, I want some!

    If there really was a significant, lasting improvement in the eczema after taking the homoeopathic remedy then there are only two rational interpretations. Either it was a total coincidence, or the remedy had a real, physical action. Labelling a “real” reaction “the placebo effect” is nonsense.

    My betting is that this eczema might not be as “gone away” as this tale suggests. Maybe it has improved a little, eczema is like that, it waxes and wanes, and if you take the homoeopathy when the condition is at its worst then there’s a tendency for an improvement to follow anyway. Or maybe it’s no different, but the whole homoeopathy ritual has convinced her to view her condition on a different light, and she’s perfectly convinced she’s better. Or maybe she’s really better and this was indeed a wild coincidence.

    Or maybe the homoeopathic remedy had a real effect (yeah, right!)

    It’s got to be one of these, and so far the last one doesn’t have much evidence in its favour. But if indeed there was a real effect, then there was a real effect and you can’t dismiss it just by shouting yah boo placebo.

    By the way, Guillaume, several trials have been done just as you outlined. Guess what?

  11. Joan said,

    September 12, 2005 at 10:54 am

    I have a daughter who’s now 5 years old. She was asmatic, the treatment is based on corthisona. We started homeopaty when she was 3, and from then the asma athacs have disapeared. I don’t think she realizes the difference of the kind of medicine she’s taking (she was happy with the inhalator, and is happy with the suggar pills) At night with the siroup for the cough after 4 hours she wakes up, because the cough comes again. Taking 3 of this small pills that “remember” its poison, she sleeps all night long. It could be a placebo for me….. but she was 3 years old. I was sceptic when we started the treatment, my wife also, but it was a chance.
    Some times the objective of the paper is fixed before starting it, so the objective of the publication is to find a way to demonstrate the “truth” needed, and the health is something also subjective. If I ask you if you fel better than yesterday? and you answer me a vage sentence I can interpret as yes or not depending on my interest
    Diifficult things to measure

  12. ej said,

    September 12, 2005 at 10:55 am

    hey, here’s an idea for the skeptic, with nothing to loose but a nasty rash. why not try the cream and prove that it doesn’t work! surely it would be worth a go, and at least would add a little real life weight to your ideas.

  13. Michael said,

    September 12, 2005 at 11:31 am

    Morag said, ‘Look, how could “the placebo effect” cause eczema to get better?’

    The rest of your comment indicates that this is a rhetorical question, which means the answer ist “that’s what the placebo effect DOES”. You seem to be mistaken about what “placebo effect” means. It’s the tendency for totally ineffective (arbitrary) remedies to have a considerable healing effect on a very wide range of illnesses when the patient believes that they should be effective.

    As for how and why this actually works, that’s the bummer – we don’t really know yet.

  14. BSM said,

    September 12, 2005 at 12:29 pm

    ” You seem to be mistaken about what “placebo effect” means.”

    I don’t think Morag is mistaking what the placebo effect means. What she is doubting is its real power.

    For conditions that have significant physical symptoms there is no real evidence that the placebo-effect is all that powerful. Its realm seems more to be subjectively reported and psychological symptoms. I have spent an considerable time looking at the claims of homeopathy and I started from a position of assuming that much of the response could be placebo effect, i.e. a real benefit derived form the therapeutic process, but not one which relates to the remedy itself. Sadly, what I have found is a depressing degree of lying and misrepresentation by the homeopathic practitioners of the real outcomes of their cases. The other process that produces dramatic improvements is spontaneous coincidental recovery. I’m afraid I am now wholly convinced that the stories of dramatic cures are either lies, mistakes or coincidences. That is why, when homemopathy is tested against controls, the casually reported miracle stories disappear and the results shuffle off to the boundaries of statistical noise.

    Homeopaths often tout both the dramatic anecdotes of cure and the sparse evidence from controlled trials of marginal benefits. But this is a contradiction. If it reliably produced dramatic cures then controlled trials would show hugely significant results. If it can only tickle a problem to produce some slight benefit, which is the most that the trial data are consistent with, then the miracle stores are just that, stories.

    The placebo effect is often invoked to explain homeopathy, but I think this is based on a misunderstanding of its likely power, but also there is an understandable desire not to be rude to homeopathy’s enthusiastic supporters by allowing that its effects might be real even if not related to the remedy. I used to share this generous view of homeopathy and homeopaths until about 3 years ago when I caught one out in a total misrepresentation of the outcome of a patient’s condition. It had never occurred to me that mistaken interpretation, at best, or outright dishonesty, at worst, might be at involved. Unfortunately I am now certain that there is a lot of both.

    Against that background, if I am led to think about a single case like the little asthmatic girl described in another post I think coincidence is most likely. It must never be forgotten that we tend only to hear from the miracle cures not from the non-cures. Coincidental recoveries may not be very frequent, but we only hear from those who have experienced them and this completely biases our assessment of their prevalence.

  15. Biju Chacko said,

    September 12, 2005 at 2:42 pm


    Re Eczema: If I recall correctly, there is some correlation between state of mind and intensity of eczema. The very act of visiting a homeopath may have caused enough stress relief to reduce the eczema.

    Little girl with asthma: I know of one case in India where the homeopath “cured” asthma with large doses of cortisone.

  16. PurpleStater said,

    September 12, 2005 at 6:11 pm

    Stuck Skeptic:

    My first, instinctive suggestion is that you should try the homeopathic medicine and see what happens. After all, you saw it help your friend, didn’t you? And you clearly believe that it works. (As others have pointed out, you could have written “her eczema cleared up by accident, independently of taking the medicine”, but you did not. Instead you said “it was the placebo effect”. This conclusion, independent of any other evidence, is as irrational as the placebo effect itself.)

    You believe in the placebo effect, don’t you? So why not try it?

    Don’t make the mistake of assuming that it won’t work because you are a skeptic. Skeptics are as irrational as everyone else. Unfortunately, some skeptics insist on pretending otherwise — impressed by their own ability to discern sense from nonsense, they overreach and begin to believe that they are actually immune to nonsense. They are wrong. Sometimes they are tragically wrong — unable to admit that he’s a sucker just like everyone else, an overreaching skeptic can be conned right out of his socks by a magician like Uri Geller and wind up wasting years making “scientific” studies of absolute rubbish. “How can it have been a trick,” he will say, “when I, a detached and rational skeptic, saw it with my own eyes?”

    Don’t be one of these people. And remember that you’re not trying to win a debate, or make a scientific breakthrough. You’re trying to use irrational means to convince an irrational person — yourself — to feel better.

    Now, I hate medical profiteers, especially the non-scientific ones, so I don’t actually recommend that you pay good money for a homeopathic remedy. I suggest you mix your own. That’s what I do — what kind of scientist would I be if I couldn’t mix up my own hyper-dilute placebo with confidence? Just be rational: don’t use anything poisonous or dangerous. Take something that even your medical doctor agrees is safe (edible herbs and spices are nice) and that you think might be effective. Dilute it a hell of a lot, and drink a bit. Don’t drink so much that you hurt yourself. And do the drinking with a bit of ceremony: I find that 3 doses spaced exactly four hours apart is quite effectively ceremonial.

    You can laugh while doing this if you want. Sometimes I find that doubling over with laughter at my own antics is good therapy.

    If I haven’t convinced you of the merits of self-administered homeopathy, there are lots of other alternative therapies under the sun. For example, try taking up bicycling, on the theory that it promotes healthy circulation, which will improve blood flow to your skin and clear up your eczema. Such a plan surely can’t hurt, and it might even be true, despite the fact that I just made it up.

    If you’re already a marathon bicyclist, try yoga.

    And — though I doubt I need to say this to a proper skeptic — you should always bring your health problems to a medical doctor if you can afford it. (I’m writing in the USA, where barbarians rule the earth and being able to afford a doctor is always an open question.)

  17. John Alan Wright said,

    September 12, 2005 at 7:53 pm

    The ‘science’ of homeopathy can be hotly debated, but my experience with it shows it NOT to be a placebo. My wife, who has 7 years of medical education, had fibromyalgia for several years. She started out with faith in her doctors, but when their therapies completely failed for several years, she began to lose her faith in them. A friend of hers told her about a chinese medic who practised homeopathy and she said she would try it. She had little faityh that it would work. In fact, she came home from her first treatment laughing and telling me how stupid it was. The physician, she said, asked her questions about her childhood and her symptoms and then prescribed little white pills which she took more to get to the end of the appointment than out of any belief that they worked. In fact, that night, her symptoms got worse. She concluded that the treatment had been completely useless.

    But the next day she felt better. WAY better. Sceptical, she thought of several other possibilities to explain her improvements, but she had to admit that it was at least very coincidental. She returned for appointments 2-3 times per week for a few months, and today her fibromyalgia is in complete remission.

    I was still sceptical (after all, I have a very scientific mind). I read Vitolkous’ studies and was unconvinced that homeopathic remedies could produce the effects in the body that they claimed they could produce, let alone that they could then cure the underlying condition.

    So when I got a terrible bout of prostatitis I sought out a urologist who prescribed several medications that turned my pee orange, dulled the pain, made me very sleepy and had no net impact on my condition.

    Then one day ,after about 3 months of fruitless efforts to control the symptoms through diet, exercise and medicine, my prostatis disappeared completely in about 3-4 days.

    Perplexed, though happy, I asked my wife if she noticed anything that I was doing differently. She told me she had treated me with a dose of a homepoathic remedy after consulting a homeopath. She slipped it into a glass of water and gave it to me to drink. She knew I would never take it if I knew what I was taking.

    So how does one explain this as a placebo?

  18. BSM said,

    September 12, 2005 at 9:23 pm

    “So how does one explain this as a placebo? ”

    I don’t.

    Ask yourself why we are not reading a hundred posts telling us about their remarkable non-recovery after taking a homeopathic remedy.

    Only those with a good story to tell are heard telling their stories. What we never hear from are the thousands of non-events when homeopathic therapy is not followed by a dramatic cure. Instances such as yours may seem amazing to the recipient of the coincidence but it isn’t really remarkable.

    We get to hear from the lottery winners, not the millions of non-winners. It’s not a mystery that the lottery is won. It just feels that way to the winner,

  19. The Dave said,

    September 13, 2005 at 12:47 am

    The real question for me is a little more complicated.

    During my time at the university, I spent more than a little time in placebo trials.

    So, now I am hopelessly addicted to placebo.

    Perhaps, then homeopathy would be an effective treatment for my addiction.

  20. Ben Goldacre said,

    September 13, 2005 at 12:58 am

    Do you think anyone’s ever tried homeopathic naloxone for treating opiate withdrawal? There might even be a black market in it.

  21. amoebic vodka said,

    September 13, 2005 at 10:39 am

    “During my time at the university, I spent more than a little time in placebo trials.

    So, now I am hopelessly addicted to placebo.”

    How did you know you were given the placebo?

    We’re hopelessly addicted to water too. We can’t go more than a day without it before experiencing unpleasant withdrawal symptoms.

  22. tom p said,

    September 15, 2005 at 6:29 pm

    John Alan Wright – I work in pharmacovigilance (looking at the safety of medicines), and I used to specialise in the herbal and chinese side of it.
    Most of the time miracle cures reported after the use of chinese medicines seems to be down to them putting in treatments that have been shown to be inappropriate for the age group (eg steroids for children with asthma) or unduly high doses of treatments that work for adults (again eg steroids with asthma) with little or no regard for the long-term safety of their patients.
    Even if they call themselves homeopaths, the term is often misused.

  23. Aloosia said,

    September 18, 2005 at 4:46 pm

    1. I have an *amazing* non-recovery on homeopathic medicine story – as a teenager, after having taken various mainstream treatments, I saw a very understanding lady homeopath every week for a few months about my acne (the curse of a middle-class youth). I stayed spotty until my hormones sorted themselves out. I did believe it would help at the time, my scepticism growing the older I got (and still going…), so it’s not like I didn’t give it my best shot.

    2. And on the subject of placebo being able to work on physical ailments: The mind seems to be extremely important in the signs, symptoms and recovery times of what would be generally classified as physical illnesses. The research group I belong to at the Swiss Federal Institute of Technology is loosly affililated with a group studying psychoneuroimmunology (www.ifvs.ethz.ch/research/index_EN), who state on their homepage:

    “Experimental results demonstrate that immune functions can be modulated by stress, classical conditioning, or psychological interventions. In addition, immune mediators can have pronounced effects on central nervous system and neuroendocrine functioning.”

    This is “good” science, and, as such, I see no reason that a homeopathic visit, chat about symptoms, family and personal history, emotional and physical problems and a nice sweety couldn’t produce a psychological state which would be helpful in clearing up eczma, which is classically understood to get worse under stressed states of mind.

    3. PurpleStater – I think your home-homeopathic-solutions are genius, and am just about to go off and dilute some herbs now. I am trying to design a little prayer-and-dance routine to go with it. And my commiserations about the barbarians.

    Happy homeopathy,


  24. Sasha said,

    September 20, 2005 at 11:20 am

    john alan wright, sounds like you covered all bases to explain your experience- except checking what’s actually in the remedy you and your wife took.

    I too have amazing tales on non-effectiveness of homeopathy. It didn’t work for my psoriasis, nor did it work for sleeping problems, or hangovers (the latter’s failure left me particularly dejected).

    However since I started watching CSI my psoriasis has cleared up steadily but comes back if I miss an episode. Strangely enough, this is only for the Las Vegas and Miami ones and not New York .Worthy of investigation.

  25. Hatter said,

    September 20, 2005 at 12:07 pm

    A placebo can only work if you believe the medication is genuine and will have an effect. You give me some cream and tell me it’ll cure a skin condition I’m going to ask is what is in it and for information on clinical trials.

    No-one has to believe in the placebo effect, it is demonstrable.

    Skeptics are human and therefore subject to being irrational. We are not intrinsically immune to nonsense, it requires constant vigilance. I find myself wanting to believe, but I remind myself to ask questions and think before accepting something. It is actually easy to practice, just watch advertising on TV; if you have children this can be fun for the whole family and you teach your children to instinctively rip advertising to pieces.

    “wind up wasting years making “scientific” studies of absolute rubbish. “How can it have been a trick,” he will say, “when I, a detached and rational skeptic, saw it with my own eyes?””

    To what skeptic are you referring?

    It would be pretty difficult for a conman like Geller to con a genuine skeptic. The first thing a skeptic would do is determine whether Geller is engaging in trickery. Only once you had determined that he was not would you devise theories and experiments to explain what he alleges he can do. Even I’d like to believe in psychic powers, but I want to see carefully controlled laboratory tests. Psychic powers, gods and the soul are all claimed to interact with and affect our physical world – if this is so then it must be possible to design equipment and laboratory tests that can detect this. There’s no such thing as metaphysics or the supernatural, it’s all physics and natural or it doesn’t exist.

    That statement from Spokesperson from the Society of Homeopaths reminds me of audio magazines attacking double-blind tests that found no difference between a wide range of amplifiers from just ordinary good quality ones through to those that cost as much as a house. Their conclusion was that double-blind testing isn’t appropriate because it didn’t find differences they know are there. As the continuing sales of pointless, overpriced cabling proves audiofools and their money are easily parted.

    I think you misunderstand what placebo effect means. It means believing so strongly that you have been given a real medicine that is going to help you that you experience the relief you expect.

    I’d expect children to be ideal candidates for the placebo effect. They trust their parents and will believe you have given them something that will help them feel better (sugar pills work really well on children 🙂 ). You could see if it still works once you put a negative spin on it – “This medicine probably won’t work because homeopathy isn’t real medicine, but we’re desperate. Mommy and daddy don’t expect it to work.”

    I don’t know what was in the cough syrup you were giving her at night, but it could well be that it is suppressing neurochemicals that rebound when it wears off or simply that she sleeps much more heavily for a short period then wakes rather than sleeping more lightly, but for longer without the syrup.

    Ashma is another ailment with a strong psychological component.

    amoebic vodka:
    Have you tried heroin? It is supposed to be a safe alternative to dangerous placebos 🙂 (pharmaceutical heroin isn’t particularly dangerous, it’s just physically addictive which is not really a big deal – tapering off pharmeceutical drugs with decreasing controlled doses is quite effective). Yes the government is very worried about the increased club use of placebos.

    Another failure, albeit with a skeptical patient – A friend has had chronic headaches for sometime. He went to various medical doctors, had scans and tests, to no avail. Eventually in desperation he went to a homeopath who spun him a ridiculous story about the likely causes and gave him something that was supposed to help. He was skeptical, but willing to take the medication for a month to give it a fair chance, afterall if it worked no belief would be required. It didn’t work.

  26. John Sutherland said,

    October 6, 2005 at 5:07 pm

    Am I missing something?

    How can you do a double-blind test to compare the efficacy of a homeopathic remedy and a placebo if you can’t tell them apart?

  27. Ben Goldacre said,

    October 6, 2005 at 5:47 pm

    hi john. well the thing is, you don’t know that there’s no difference in outcome between the two groups until after you’ve actually done the study.

    just to recap, this is what you do in a randomised placebo control trial. you take a bunch of patients. you split them into two groups randomly (dice, coin, sealed envelopes, computer random number generator etc). you give the people in one group placebo, and the people in the other group whatever your treatment is, let’s say homeopathic tablets. then you wait a while, and then you measure – in whatever way you decide – how each group is doing, after however long seems reasonable.

    blinding refers to whether the patients know what treatment they’re getting. “double blinded” means neither the patient nor the person giving the treatment know who’s getting placebo or real treatment.

    it’s a huge area, and incredibly important, and my fingers would drop off if i wrote a casual posting explaining it here. if you’re interested in reading about how research is done, the best book (until mine comes out next year of course) is “How To Read A Paper” by Trisha Greenhalgh. It’s published by BMA books and I can’t recommend it highly enough.

    How to Read a Paper: The Basics of Evidence-Based Medicine

    and in the glorious socialist utopia of the collaborative empirical project it’s also available online here:



  28. Matt S Trout said,

    October 19, 2005 at 3:55 am

    Hmm … two “placebo effect” successes, one for eczema and one for asthma. But how?

    As somebody commented, eczema can be state-of-mind-related. I had what *appeared* to be eczema between my fingers starting shortly after I took a job that turned out to be truly awful (systems coding, EA-like hours). Now I am no longer stressed, it’s gone again. Perhaps the placebo allowed the person in question to stop worrying about it sufficiently to break the vicious cycle that other Eczema sufferers I’ve known have got into over it?

    The asthma cure is maybe a bit harder, but given it’s primarily coughing that’s the symptom being relieved rather than full attacks, possibly it’s just the sugar pills causing the kid to fall asleep more relaxed and therefore less likely to start coughing .
    I’ve always preferred paganism to homeopathy, but I think that’s largely my approval for anything involving mead, placebo effect or not 🙂

  29. Morag said,

    November 14, 2005 at 11:12 pm

    Michael, and Hatter, I do not misunderstand what the “placebo effect” means. I am questioning whether what you describe actually exists. To be blunt. I do not believe that a strong belief on the part of a patient can cause them to “experience the relief they expect”. I think this is so much mystical claptrap. If you have evidence to the contrary (and remember, the plural of anecdote isn’t “evidence” either), then please present it.

    To underline what my friend BSM said, actual examples of apparent recovery following a placebo treatment may be seen as either coincidence, or imaginary. That is, either the recovery was going to happen anyway, or the patient is showing no objective improvement, but merely imagines that he or she is better.

    Consider the question of veterinary homoeopathy. If the “miracle cures” were real, due to the strength of belief on the part of the patient, there would be little or no effect on animals. In fact, we find that homoeopathy actually seems to work even better in animals than in people! Anecdotally, that is.

    The reason for this is simple. If the person doing the reporting is actually the patient who is suffering from the illness, while it is possible that they might be bamboozled enough by the claims of the magic sugar pills to imagine they feel better, they are also experiencing the symptoms first-hand, and these may be insistent enough to counteract the auto-suggestion. In contrast, where another observer is inserted between the patient and the physician, the sky’s the limit.

    Homoeopath: Oh look, isn’t he so much better.
    Owner (dubiously): Well, maybe….
    Homoeopath (confidently): Oh yes, he’s looking great! Look at that clear eye, the damp nose, and he’s wagging his tail! He knows we’re helping him.
    Owner (enthusiastically): Isn’t that wonderful! It’s a miracle!
    Dog: I feel terrible, you freaking morons!

    BSM is right. The idea that there is a real “healing” effect and not just a huge dollop of imaginative optimism combined with some coincidental recovery is simply a feature of the majority of the stories told being the dramatic ones.

    I can echo Aloosia’s experience. As a teenager with acne, I was taken to a homoeopath by my mother. The sugar pills I was given had no effect at all. My acne resolved at the time acne usually resolves. However, I’m not likely to be running around the Internet posting that enthusiastically on homoeopathic forums. (When I challenged a colleage about his espousal of homoeopathy, I was told that he had no doubt it worked because it had cleared up his acne when he was 29. His only regret was that he hadn’t tried it much sooner, because he had been left badly scarred. My enquiry as to why my anecdote was of less value than his, and my observation that he’d actually been quite unlucky to have had acne persist as long as age 29, went unanswered.)

    If belief and “mind over matter” can have any effect on actual physical disease (beyond what might be expected if anxiety can be lessened, in conditions where anxiety is actually contributing to the symptoms), then can we please have some evidence?

    Face it, your “placebo” is simply a description of what was going to happen anyway, with rose-coloured glasses.

  30. Scott said,

    November 21, 2005 at 9:04 pm

    “If belief and “mind over matter” can have any effect on actual physical disease (beyond what might be expected if anxiety can be lessened, in conditions where anxiety is actually contributing to the symptoms), then can we please have some evidence?”

    Of course, the number of physical diseases and/or symptoms where anxiety or other moods can have an impact is rather large. Outbreaks of eczema, for example, are frequently linked to stress, and there’s a whole field of psychoneuroimmunology investigating the impact of mood and cognition on the immune system, although I suspect that in the end the observable effects will turn out to be rather small. There’s actually somewhat stronger evidence for the nocebo effect than a strong placebo effect independent of the natural course of healing – a number of studies where people exhibited an inflammation response to innocuous stimuli they were expecting to be allergens.

    Anyway, to dismiss all of these possibilities as ‘mind over matter’ smacks of Cartesian dualism. Whenever you have a thought or emotion, there are real, physical changes going on in your brain, including changes in the balance of neurotransmitters. From there, even if the findings on the immune system turn out to be weak, it’s perfectly possible for the effects of these physical changes to be transmitted via the pituitary through the rest of the endocrine system, which can have significant implications for general bodily health.

  31. Morag said,

    November 21, 2005 at 9:59 pm

    OK, Scott. I asked for some evidence, and you have stated that such evidence exists. Would you mind actually presenting it?

  32. Duke said,

    February 5, 2006 at 11:49 am

    All fellow believers are invited to visit with me at:


    The Faithful Witness

  33. Xinxi said,

    January 8, 2007 at 2:07 pm

    Does anyone know a good source/site on the Placebo effect? Is there any clear definition of its range? Does the term only apply to cured illnesses?

    The experiments done by Dr. Herbert Benson seem to have a similar bent: Strange, irrational practices, believers and measurable effects in the end.

    My Psoriasis didn’t respond to Homoeopathy and just weakly to standard therapies (Cortison etc.). Strangely enough, the Communist version of TCM (“Traditional” Chinese Medicine) works. The only trouble is getting personalised packages full of weird herbs or even better: occult looking liquids abroad. By the way: Isn’t it time to bash Oriental medicine on this page? It is a bit hilarious that Chinese doctors are moving away from Acupuncture while the West seems enthusiastic about it.

  34. The Blog Planet - Medication Of Allergy Asthma said,

    June 15, 2009 at 1:34 am

    […] A tonic for sceptics – Bad Science […]

  35. The Powerful Placebo « Stuff And Nonsense said,

    November 13, 2009 at 10:33 pm

    […] further detail on the study, I found two posts on the Bad Science blog that also referred to it. Here, and here are the two posts. The findings are described as follows: A placebo operation in the […]

  36. wayscj said,

    November 21, 2009 at 7:33 am

    ed hardy ed hardy
    ed hardy clothing ed hardy clothing
    ed hardy shop ed hardy shop
    christian audigier christian audigier
    ed hardy cheap ed hardy cheap
    ed hardy outlet ed hardy outlet
    ed hardy sale ed hardy sale
    ed hardy store ed hardy store
    ed hardy mens ed hardy mens
    ed hardy womens ed hardy womens
    ed hardy kids ed hardy kids ed hardy kids

  37. iphone revolution said,

    December 30, 2009 at 8:35 am

    iphone wireless


    Apple iphone

  38. . o O ( Cure Your Eczema in 14 Days Review: Scam or Serious? ) said,

    January 6, 2010 at 10:13 am

    […] is a healthy, easy to make lunch menu? | Your Healthy AlternativeA tonic for sceptics – Bad ScienceLaundry detergents – Good & Bad « Natural E GREEN – Chemical free […]

  39. faculty of 1000 said,

    January 19, 2010 at 6:31 pm

    […] you might think that patient bias is reduced). This reminds me, parenthetically, of my favourite placebo story. Back in the Fifties a particular treatment for angina involved ligating certain arteries. […]

  40. Here we go again « faculty of 1000 said,

    January 19, 2010 at 8:14 pm

    […] you might think that patient bias is reduced). This reminds me, parenthetically, of my favourite placebo story. Back in the Fifties a particular treatment for angina involved ligating certain arteries. […]