Articles

144 thoughts on “Acupuncture for Depression

  1. Fifi says:

    Nice summation Medsvstherapy.

  2. squirrelelite says:

    You’re welcome, Alison.

    Steven’s post from today is also of interest on this subject:

    http://www.theness.com/neurologicablog/?p=1714#more-1714

  3. BillyJoe says:

    apteryx,

    “If a sizeable component of the “placebo effect” is due to the poor silly sheeple putting on a brave face to please the investigator, this implies that suppressing the activity of endogenous opioids makes people enormously less eager to please. I’d have to see some substantive evidence before I accepted that presumption.”

    There is nothing right about what you’ve said above from “poor silly sheeple” to your mathematics, but I’m having no luck explaining things to you so I’ll leave it to some other more competent person to have a go. :(

  4. pmoran says:

    Evidence concerning the potential for and the nature of placebo responses is vast, and derived from many sources.

    We can be sure that a lot of the benefits often attributed to placebo benefits are illusion and actually due to spontaneous happenings or patient reporting biases.

    OTOH, laboratory studies show very strong apparent placebo “activity”, and suggest the activation of endorphin activity and symptom-relieving cerebral pathways on MRI. If such data is being interpreted correctly, it offers very solid evidence that placebo reactions are genuine responses to treatment in every sense other than in being mediated by psychological rather than pharmacological or physiological influence.

    It would have been helpful if RCTs comparing patients receiving placebos with those on no treatment, or on waiting lists for treatment, gave conclusive results, but they are also difficult to interpret*. They actually provide “effect sizes” for placebo upon subjective symptoms such as pain that are not huge but too large to ignore. The usual clinical trial is not conducive to strong placebo influences so, again, the power of placebo remains elusive.

    Anecdotal evidence has certainly always suggested strong effects under the right conditions.

    So how powerful is the placebo? I don’t know for sure. I also suspect bias when anyone is dogmatic on the subject.

    * Hrobjartsson, A., Gotzsche, P. C. (2001). Is the Placebo Powerless?. N Engl J Med 345: 1276-1279

  5. Fifi says:

    Nicely said Dr Moran. Placebo effects don’t equal no effect. Another thing that’s worth considering – that I suspect a lot of the general public doesn’t really think about – is that most medications act as a push for a release (or dampening) of chemicals that already naturally exist in our system.
    One example of this is how drug addicts can get a phantom high from watching or going through the ritual of taking their particular drug of choice. Not just a craving but an actual sensation (not long lasting or likely to satisfy someone in withdrawal but it still happens). It’s a matter of having a Pavlovian or conditioned placebo response, a physiological response. This may or may not be different than one or some of the placebo effects people experience during clinical trials but it certainly points to our body’s capacity to be pushed into a physiological response via environmental stimulus.

    An interesting article that looks at the neurobiology of placebo response (for anyone interested in delving further into this topic).
    http://www.medscape.com/viewarticle/571129

  6. Fifi says:

    apteryx – The existence of placebo effects doesn’t actually make a good argument for the efficacy of acupuncture, nor does it prove that it “works” as anything other than an elaborate/dramatic (and often expensive) ritual that could be replaced by any other elaborate/dramatic (and inexpensive) ritual. And the fact that people have placebo responses doesn’t make them “sheeple”, it’s just an interesting facet of human nature and worth exploring for what it is (rather than dressing it up in false explanations/lies about how it “works” via meridians and chi and so on, or charging poor suckers lots of money and eating up years of their life to study lies about meridians and chi).

  7. BillyJoe says:

    BillyJoe: “There is nothing right about what you’ve said above from “poor silly sheeple” to your mathematics, but I’m having no luck explaining things to you so I’ll leave it to some other more competent person to have a go. :(”

    Thanks, Fifi, for having a go :)

  8. Chris Kresser says:

    Very interesting article. I was only able to skim the 107 comments, so I apologize if what I’m about to say has already been covered.

    Sample size is often based on an estimate of the expected size of the test and control treatment effects. If the sham intervention is assumed to be inert, calculations are based on general ideas about the size of the placebo effect. This is often pegged at 30% based on Beecher’s original work. However, his calculations have been found to be erroneous and most placebo researchers today believe the effect is higher. (http://bit.ly/bbFKFa) To demonstrate a difference between sham at 30% and active treatment at 70% requires a relatively small sample size. But if the sham intervention is closer to 50% effective, a larger sample size will be needed to demonstrate effectiveness. The majority of acupuncture trials have suffered from inadequate sample sizes. (http://bit.ly/9HJgOT)

    Trials that assume sham is inert are at greater risk of showing false-negatives. De Craen et al showed that if the effect of the noninert placebo is small, the will be biased if the treatment effects are also small. This bias increases if the effects of noninert placebo are larger. (http://bit.ly/cuhrTp) So, inadequate sample size is one factor that has compounded acupuncture research.

    It has long been known that surgery is associated with strong placebo effects. Various reasons have been offered ranging from the degree of physical discomfort during the procedure to the meaning attributed to shedding of blood. But the human body has protective and restorative mechanisms that are triggered by injury. These are not placebo effects – they are direct and predictable responses of the body to the trauma of surgical incision.

    Puncture of the skin by acupuncture needles, like surgical incisions, provoke a cascade of physiologic responses which not only repair the wound but also trigger a range of biologic effects that have systemic effects in the body. These responses are well-documented and include activation of the blood coagulation and immune complement systems. Recent work by Maier et al and Tracey et al have shed light on how the brain controls certain types of inflammation and therefore pain. Their work has given us clues on how acupuncture reduces inflammation and relieves pain. For example, they explain that inflammatory formation is transmitted through sensory nerves to the hypothalamus, where input signals are processed, resulting in anti-inflammatory output via the humoral system and the ANS via the HPA axis. (http://bit.ly/afuPqV)

    I’ve often heard the claim that “little is known about the analgesic effects of acupuncture.” On the contrary, according to Professor Bruce Pomeranz of the University of Toronoto, “We know more about acupuncture analgesia than about many chemical drugs in routine use. For example, we know very little about the mechanisms of most anesthetic gases, but still use them regularly.” (http://bit.ly/9L0NYV) I’d add aspirin to that list.

    The recently developed noninvasive sham needle of Streitberger and Kleinhenz is considered to be an inert placebo needle by its inventors. However, reports of its use suggest otherwise. One researcher in Germany cause bleeding in a patient using the Streitberger needle.

    Also, since the effects of acupuncture are largely nonspecific, involving the stimulation of innate healing mechanisms, the results of treatment will depend upon 1) the status of the patient’s immune system and overall potential for self-healing, and 2) the potential of the body to “self-heal” the condition in question. #1 depends on a multitude of factors difficult (nutritional status, medications, etc.) difficult to control for in a clinical trial.

    When placebo effects interact with or are inseparable from other effects, placebo-controlled trials become difficult if not impossible. This has prompted authors like Gotzsche to suggest that the concept of placebo should be discarded. He points out that the nonspecific effects of so-called “placebo” can’t be assumed to be the same in two arms of the same trial, or from one trial to another. Some effects, such as the increased skill of an investigator and “white coat syndrome” could be different in different groups. And unless the outcome was survival it would be hard to measure the effect in the untreated group without interference. (http://bit.ly/cph1cV)

    Finally, Moffet’s recent review in the Journal of Clinical Epidemiology demonstrates that acupuncture is distinguishable from placebo and affects outcomes. (http://bit.ly/9LoBJc)

    While I agree that 1) the effects of acupuncture are nonspecific, 2) there is no demonstrable difference between sham and active acupuncture, provided the sham stimulates the same nonspecific mechanisms as the active treatment, and 3) there is no support for traditional methods of point selection nor any indication that point selection makes a difference, I think it’s inaccurate to claim that acupuncture is indistinguishable from placebo.

  9. BillyJoe says:

    Chris Kessler

    “While I agree that 1) the effects of acupuncture are nonspecific, 2) there is no demonstrable difference between sham and active acupuncture, provided the sham stimulates the same nonspecific mechanisms as the active treatment, and 3) there is no support for traditional methods of point selection nor any indication that point selection makes a difference…”

    1) The effects of acupuncture has not been reliably demonstrated to be greater than zero.
    Note: acupuncture is defined as sticking needles through the skin to a depth of 1cm at specific acupuncture points.
    (If you are using a drastically different defintion, you are no longer describing acupuncture.)

    2) There is no demonstrable difference between sham and active acupuncture.
    Note1: It is now possible to do effectively double blind acupuncture trials using sham acupuncture needles.

    3) There is no support for traditional methods of point selection nor any indication that point selection makes a difference, or that you even need to stick the needles through the skin.

    “I think it’s inaccurate to claim that acupuncture is indistinguishable from placebo.”

    I’m going to assume this is a typo, because the part of your post that precedes it (ie the rest of your post) suggests the exact opposite: that the effect of acupuncture above placebo is zero – meaning that acupuncture does not work.

  10. Chris Kresser says:

    What of Moffet’s 2008 review of 78 trials indicating that acupuncture is distinguishable from placebo? (Link in previous post) To claim that the effect hasn’t reliably been demonstrated as greater than zero ignores the reviews that claim the opposite. Nor is it accurate to claim that studies reliably demonstrate the effect IS NOT greater than zero. There are numerous issues with the way acupuncture trials are performed that are still confounding results, as I described in my comment.

    It’s not correct that acupuncture is defined as needles piercing the skin at a 1 cm. depth. There are over 80 styles of acupuncture in China alone, not to mention styles in Japan, Korea and other countries. In Japan there is a style called “contact needling” that involves stimulation of the points with needles without breaking the skin. It cannot be ruled out that the so-called sham treatment in many studies is causing the same nonspecific effect that penetrative acupuncture is stimulating.

    According to Moffet:

    “The practice of making a priori assumptions about what are acceptable placebo effects in a trial needs to be re-examined, because this has been responsible for introducing faulty assumptions about the nature of placebo, and bias against the therapy being investigated. Often these a priori assumptions are based on ignoring published literature about treatment effects and biologic mechanisms that are known to arise in response to the proposed treatment methods.”

    If a non-invasive sham procedure is used in a trial, it has to be investigated for physiologic effects to establish that it is actually inert, and more complete procedures must be used to separately control for other nonspecific physiologic effects.

    Since the therapeutic effect of acupuncture is achieved by activating the body’s self-healing mechanisms, it’s essential to determine that these natural healing processes (that are in fact the goal of treatment) aren’t confused with placebo and are controlled for in sham trials. There are very few trials that have done this, and it remains to be seen whether it can even be done reliably.

  11. BillyJoe says:

    Chris Kessler,

    “What of Moffet’s 2008 review of 78 trials indicating that acupuncture is distinguishable from placebo? ”

    Of the 78 trials:

    22 trials compared acupuncture to usual care alone or other non-acupuncture controls – in other words they failed to control for placebo effects.
    That leaves us with 56 trials.

    40 trials compared acupuncture interventions mainly differentiated according to traditional acupuncture theories – still no placebo control.
    That leaves 16 trials.

    16 trials did use placebo controls; and 11 had statistically significant differences in outcomes.
    But, without being able to access the details – unless you are willing to pay for them – how do we know that the placebos where adequate to account for the total placebo effect associated with acupunture?

    I’m willing to bet – because I’ve seen this sort of thing many times before – that the 5 that showed no effect were the ONLY ones with an adequate placebo control.

  12. BillyJoe says:

    Chris Kresser

    “It’s not correct that acupuncture is defined as needles piercing the skin at a 1 cm. depth. There are over 80 styles of acupuncture in China alone, not to mention styles in Japan, Korea and other countries.”

    And what does this tell you about acupuncture?
    (hint: does it support your view of acupuncture or does it support my view that you can stick the needles in wherever you like and you don’t even have to stick them in.)

    “In Japan there is a style called “contact needling” that involves stimulation of the points with needles without breaking the skin.”

    And what do the clinical trials testing for the effectiveness of Indian Style Acupuncture show?

  13. BillyJoe says:

    Chris kessler,

    “If a non-invasive sham procedure is used in a trial, it has to be investigated for physiologic effects to establish that it is actually inert”

    Placebos are not inert.

    “Since the therapeutic effect of acupuncture is achieved by activating the body’s self-healing mechanisms:

    There is no point speculating about mechanism whereby acupunture achieves its effects if clinical trials have shpown no effect.

    “..it’s essential to determine that these natural healing processes aren’t confused with placebo”

    Whereas acupuncture has been shown to be ineffective, this is not true for placebos. Two placebo tablets are more effective than one placebo tablet, placebo capsules are more than placebo tablets, placebo injections are more effective than placebo capsules, etc etc.
    There’s lots of clinical trial support for placebo effects, and no support for an effect of acupuncture.

  14. apteryx says:

    How can you be willing to speak of placebos in conventional pharmaceutical dose forms being variably “effective” – as in, acknowledging that patients will gain more relief from a placebo capsule than a placebo tablet – but not grant the same status to acupuncture-as-placebo, even when the acupuncture gives yet more relief? If the sugar pill provides less relief than ibuprofen and the acupuncture provides more, how many people would say that the sugar pill is effective and the acupuncture ineffective? Your definitions of these words are starting to look not just different from those of many average people, but quite idiosyncratic.

  15. Chris Kresser says:

    “16 trials did use placebo controls; and 11 had statistically significant differences in outcomes.
    But, without being able to access the details – unless you are willing to pay for them – how do we know that the placebos where adequate to account for the total placebo effect associated with acupuncture?”

    And how do you know they WEREN’T adequately controlled?

    I hate to tell you, but “because Billy Joe has seen this kind of thing before” isn’t exactly a valid scientific reason for the rest of us to dismiss the the 11 out of 16 trials that showed a statistically significant difference in outcomes.

    And here’s the thing: as I’ve pointed out, there are few trials with a sham control that doesn’t activate the same physiologic mechanisms that are the goal of acupuncture treatment.

    Gotzsche’s definition of placebo is as follows: “The placebo effect is the difference in outcome between a placebo treated group and an untreated control group in an unbiased experiment.” It reflects the impossibility of defining the placebo effect in a single case, where biases of various types can’t be excluded (including regression to mean, investigator effects and changes in patient’s behavior).

    In studies with a parallel untreated group (i.e. no placebo or other intervention, as with waiting list controls) these additional nonspecific effects would produce the same response as in a “placebo group” – except for the true placebo effect. Then the effects observed in the untreated control group can be subtracted from those in the placebo group to get the true placebo effect. A three-arm study like this is the most unbiased way to evaluate placebo effects in the context of a clinical trial, but to my knowledge very few have done this with acupuncture.

    So-called sham acupuncture can’t be assumed to be adequate placebo control until 1) it is proven that sham doesn’t stimulate the same biological effects as “active” acupuncture, and 2) it is proven that sham acupuncture doesn’t improve outcomes more than an untreated control group.

    I’m not arguing that acupuncture is more effective than sham acupuncture. I’m arguing that there aren’t enough properly designed trials to support your claim that acupuncture isn’t more effective than placebo.

  16. BillyJoe says:

    apteryx,

    “How can you be willing to speak of placebos in conventional pharmaceutical dose forms being variably “effective” – as in, acknowledging that patients will gain more relief from a placebo capsule than a placebo tablet – but not grant the same status to acupuncture-as-placebo…”

    I don’t deny it.
    In fact I have acknowledged that the placebo effect associated with acupuncture is a very effective one. What you don’t seem to want to accept is that the acupuncture itself has no effect.
    Clinical Effect = Placebo Effect + Acupuncture Effect
    Acupuncture Effect = 0

    “…even when the acupuncture gives yet more relief?”

    The effect of the acupuncture itself is zero.
    It is the placebo effect associated with acupuncture that is more effective than many other forms of placebo.

    “Your definitions of these words are starting to look not just different from those of many average people, but quite idiosyncratic.”

    I’m trying my level best to be accurate.
    If your take home message from clinical trials of acupuncture is that acupuncture works, you are either wrong, or misleading, or lying. Acupuncture itself does not work. But the placebo efffect associated with acupuncture is pretty good as far as placebo effects go.
    But, in that case, and if you want to follow the evidence where it leads, why not use sham acupuncture and avoid the risk of inserting needles into the skin.

  17. BillyJoe says:

    Chris Kessler,

    “And how do you know they WEREN’T adequately controlled?”

    That’s a bit unfair.
    You direct me to some evidence, but I have to pay to find out what that evidence is. Well, it doesn’t work that way. If you have evidence that acupuncture is effective please provide it.

    “I hate to tell you, but “because Billy Joe has seen this kind of thing before” isn’t exactly a valid scientific reason for the rest of us to dismiss the the 11 out of 16 trials that showed a statistically significant difference in outcomes.”

    I don’t expect you to.
    It just my own reason to suspect what I will find if I looked at those trials ansd my reason for not paying to look at them.

    “And here’s the thing: as I’ve pointed out, there are few trials with a sham control that doesn’t activate the same physiologic mechanisms that are the goal of acupuncture treatment. ”

    What I am pointing out is that the physiological mechanisms activated by acupuncture are no different from the physiological mechanisms activated by the placebo effect.
    You mentioned 80 different types of acupuncture, but if we stick with what we might call Traditional Acupuncture”, these are the facts:

    1) You can stick the needles in anywhere you like.
    2) You don’t need to stickthe needles in.

    Where does that leave Traditional Acupuncture?
    And, if you want to consider the 80 other types of acupuncture, you will need to show that they are more effective than Traditional Acupuncture.

    I suspect that Moffet’s “40 trials comparing acupuncture interventions mainly differentiated according to traditional acupuncture theories” addresses this point and I suspect that it shows that [the placebo effect associated with] Traditional Acupuncture works better than the other types is was tested against, but I’m not going to pay to fine out

    “In studies with a parallel untreated group … A three-arm study like this is the most unbiased way to evaluate placebo effects in the context of a clinical trial, but to my knowledge very few have done this with acupuncture.”

    You haven’t looked very far. One example is a systematic review of acupuncture for the prophylaxis of migraine. You can google it if you like. And you won’t have to pay.

    “So-called sham acupuncture can’t be assumed to be adequate placebo control until 1) it is proven that sham doesn’t stimulate the same biological effects as “active” acupuncture, and 2) it is proven that sham acupuncture doesn’t improve outcomes more than an untreated control group.”

    The onus of proof is on acupuncture. You must show that acupuncture does not just “stimulate the same biological effects” as placebo.
    And, if you haven’t seen evidence of the effect of sham acupuncture above that of untreated controls you haven’t looked. I refer you again to that systematic review.

  18. Chris Kresser says:

    Billy Joe,

    What is your definition of placebo?

  19. BillyJoe says:

    Chris,

    There is a whole literature on placebos which anyone can access. A brief reply here is bound to be incomplete and would be a set up for a demolition job or endless rounds of discussion.

    Remember that it is up to proponents of acupuncture to prove an effect. They need to define what acupuncture is. You say there are 80 different types. Then pick one type and define what you think it is, what is does, and how it does it.

    Traditional Acupuncture proposes the existence of a number of entities that have never been shown to exist:
    - chi
    - meridians
    - acupuncture points

    You might say that that doesn’t matter if it can be shown to work. So how is it done? Well, there are specific points [which have nothing to distinguish them anatomically from any other points] on the skin [aligned along non-existent meridians] and if you select a specific group of these points [groupings which seem to have no logic behind them at all] and stick needles through the skin at these points to a depth of 1cm, you can help relieve all sorts of ailments [though AT BEST relief of some limited list symptoms such as pain, nausea, and depression only has been demonstrated].

    But the truth is that the use of placebos has demonstrated that:
    - it doesn’t matter where you stick the needles in.
    - it doesn’t matter if you don’t stick them in.

    So, I ask you again: where does that leave Traditional Acupuncture? And where are the studies that show any of the 80 other types of acupuncture are any better than Traditional Acupuncture.

    I think it’s informative that you have not been able to respond in any meaningful way to my criticisms, including my criticisms of the study you linked to in your support of acupuncture [at the very least I have reduced your list of 78 trials to 16 of which 11 apparently showed "some effect" - and you do know what they say about extraordinary claims, don't you?].

    BillyJoe

  20. BillyJoe,

    I think a few people have been asking, what if “stimulating the skin somewhat noxiously” has an effect on perception of pain? Could that be considered a treatment for pain in itself or must it always be considered a placebo?

  21. apteryx says:

    Clearly, the ancient Chinese were wrong about HOW acupuncture worked; there’s no such thing as meridians that carry qi. However, that does not mean that acupuncture, and/or every single other thing done by Chinese healers who believed in qi, does not work at all. It only means that when people do not have the necessary scientific infrastructure and equipment to explore the real microscopic causes of observed phenomena, they almost always invent explanations as a way of making sense of their world.

  22. Harriet Hall says:

    “what if “stimulating the skin somewhat noxiously” has an effect on perception of pain? Could that be considered a treatment for pain in itself or must it always be considered a placebo?”

    If that’s true, it’s an argument for stimulating the skin, not for all the rigamarole of acupuncture.

    Some people think one of the mechanisms for the placebo response is production of endorphins in the brain, but that happens after placebo pills without any noxious stimulation.

  23. Harriet Hall on acupuncture:
    “If that’s true, it’s an argument for stimulating the skin, not for all the rigamarole of acupuncture.”

    True. But there are some people on this thread trying to talk about whether stimulating the skin has an effect.

    When they posit that stimulating the skin could be effective in pain relief and therefore that someone who visited an acupuncturist could get pain relief from skin stimulation in addition to the ritual, they run into other people who say that “acupuncture doesn’t work.”

    I don’t think anyone on commenting on this thread has proposed the existence of chi that needs realigning.

    There are not-weird people who have proposed that skin stimulation might be beneficial for pain, and who have wondered whether acupuncture studies might support that hypothesis, but they often have trouble engaging knowledgeable people on this blog.

    I find this a little odd. How much does it matter if skin stimulation is called “skin stimulation” or “acupuncture”? Does changing the name of the intervention change the result of the study? Or do studies not support the idea that skin stimulation has any effect at all, no matter what it’s called? Or does it not matter, because skin stimulation is nonspecific and therefore can never be anything other than placebo?

    I don’t think these questions are that off the wall. I’m never going to see an acupuncturist so I personally don’t care, but I do think it’s odd when someone asks “Is there any evidence that skin stimulation relieves pain?” and are consistently responded to with “There’s no such thing as chi.”

  24. … A long time ago, I thought that acupuncture for pain relief was fairly well-supported, possibly through triggering of endorphins.

    Then I read that it wasn’t. Which is fine. I dropped my belief in acupuncture for pain relief.

    But when on this blog we look at the studies of acupuncture, while meridians are clearly not supported, and penetration of the skin is clearly not supported, it does appear that skin stimulation might be supported. Maybe it’s not, but it’s not clear to me that it’s not. And if it’s not, I wouldn’t mind a discussion of how we know that.

  25. Harriet Hall says:

    Most of us have had the experience of rubbing a painful area and thinking that it seems to help. There is some evidence that acupuncture relieves pain by causing endorphin release in the brain, but we don’t know what that means because it has not been adequately compared to other kinds of skin stimulation or counter-irritation. And we have some evidence that taking placebo pills can have a similar effect. What if we could definitely show that placebos actually produce endorphins. Would we have to stop calling them placebos? Would it be acceptable to prescribe them for patients?

    We know that only a minority of patients respond to placebos whereas a majority respond to morphine. I suspect we will find that rubbing the sore spot really doesn’t help much, just as acupuncture doesn’t really help much; so even if skin stimulation produces endorphins, the magnitude of clinical usefulness is likely to be low.

    My crystal ball says that studies will show that skin stimulation relieves pain, but not enough to matter. :-)

  26. BillyJoe says:

    Alison Cummins,

    “BillyJoe, I think a few people have been asking, what if “stimulating the skin somewhat noxiously” has an effect on perception of pain? Could that be considered a treatment for pain in itself or must it always be considered a placebo?”

    Well, it sure as hell aint acupuncture!

    A mother wincing in pain at her stubbed toe suddenly forgets all about her pain as she leaps to her feet to save her infant from falling off the change table.
    Is that a placebo effect?

    I dunno, but it sure as hell aint acupuncture.
    (Okay, that was little abstruse :D)

  27. BillyJoe says:

    apteryx,

    “Clearly, the ancient Chinese were wrong about HOW acupuncture worked; there’s no such thing as meridians that carry qi. However, that does not mean that acupuncture, and/or every single other thing done by Chinese healers who believed in qi, does not work at all.”

    Seems I’m not the only one being abstruse! :D

    ” It only means that when people do not have the necessary scientific infrastructure and equipment to explore the real microscopic causes of observed phenomena, they almost always invent explanations as a way of making sense of their world.”

    In other words, they were flat out wrong. ;)

  28. BillyJoe says:

    Alison Cummins,

    “…there are some people on this thread trying to talk about whether stimulating the skin has an effect.”

    Not really. They are trying to make a case for acupuncture out of it.

    “When they posit that stimulating the skin could be effective in pain relief and therefore that someone who visited an acupuncturist could get pain relief from skin stimulation in addition to the ritual, they run into other people who say that “acupuncture doesn’t work.””

    The two are not mutually exclusive.
    The ritual and the skin stimulation provides the total effect. The sticking in of needles at specified acupuncture points (ie acupuncture) adds nothing.

    “I do think it’s odd when someone asks “Is there any evidence that skin stimulation relieves pain?” and are consistently responded to with “There’s no such thing as chi.””

    I don’t think that is in any way an accurate characterisation of the discussion we’re having here.

  29. apteryx says:

    Dr. Hall: “We know that only a minority of patients respond to placebos whereas a majority respond to morphine. I suspect we will find that rubbing the sore spot really doesn’t help much, just as acupuncture doesn’t really help much; so even if skin stimulation produces endorphins, the magnitude of clinical usefulness is likely to be low.”

    I recognize you have a philosophical opinion about acupuncture that is just not going to change, but your argument is countered by studies that show acupuncture to be better than active medication for certain types of pain. If we really want to know what “the magnitude of clinical usefulness” for a given condition is, we have to depend on data rather than belief (i.e., what you “suspect”).

    BillyJoe: “In other words, they were flat out wrong.”

    Yes, indeed, about their explanation for WHY observed phenomena occurred. It is a natural human trait to try to find means of explaining what happens to you; you don’t want to think you live in a dangerous and chaotic universe without cause and effect. Ancient Greek and Asian physicians therefore came up with explanations for diseases, like imbalanced humors or qi, that later science has demonstrated to be false. That does not mean that those physicians weren’t able to recognize specific diseases, like diabetes, or that the patients they reported to have those diseases weren’t really ill at all. Likewise, they were able to observe that opium relieved pain; if they wrongly believed that it did so by rebalancing your humors or restoring your qi, that doesn’t change the fact that it did work.

  30. Chris Kresser says:

    Here’s the thing Billy Joe: we’re not actually that far off.

    I’m an acupuncturist (at least I will be in four months – I’m finishing school). But I don’t believe that qi = energy, or that energy is flowing around the body in imaginary lines called meridians.

    As a matter of fact, I just finished writing a series of posts on my blog explaining that the “energy meridian” model is a western creation. It’s not a historically accurate description of Chinese medicine. It’s instead the result of mistranslations of key Chinese characters by a French bank clerk named Soulie de Morant who lived in China in the early 1900s and was responsible for introducing Chinese medicine to the west. Read more at my blog: http://thehealthyskeptic.org/acupuncture.

    Further, I’m not making an argument for what is called TCA in the scientific literature. (Again, it’s a western idea that there is such a unified theory called TCA. Chinese medicine is in reality a collection of many different theories. “TCA” was created by Mao & co. during the communist revolution in an effort to standardize the medicine and make it easier to teach to the “barefoot doctors”.)

    I agree that the literature doesn’t support the idea that point selection matters, or that needling technique makes a difference, or that even breaking the skin makes a difference. However, I’ll also say that just because the literature doesn’t yet support a difference that doesn’t mean there isn’t one. As I pointed out earlier most studies on acupuncture are underpowered because sample sizes are based on the assumption of a 30% placebo effect, which is unsound.

    You say that stimulating the skin without breaking it is not acupuncture. Perhaps not as it’s defined on the west. But here we have another mistranslation. Jesuit monks living in China translated the character “zhenzi” as “acupuncture”. Acus (needle) punctura (puncture). A more correct translation of the term is needle therapy. Sometimes the Chinese punctured the skin and left the needles there. Sometimes they used them for bloodletting. (Before you dismiss bloodletting as hopelessly archaic, keep in mind that it’s still used in modern medicine today. In fact, a fairly recent study showed that leech therapy [bloodletting] was more effective than surgery for osteoarthritis of the knee.) In Japan, some styles don’t puncture the skin at all.

    My disagreement with you is on the nature, meaning and clinical significance of placebo.

    Last week a patient came in to the clinic with a history of severe acid reflux / GERD. He had an h. pylori infection with ulcer about twelve years ago, and since then he’s had intractable reflux. He’s been treated with all of the PPIs and H2 blockers, has seen numerous doctors and has tried several different special diets. Nothing helped. As a last resort he came to try acupuncture. This was a difficult decision for him because he works as a research scientist. He came in saying “I don’t believe in this stuff, but I don’t want to have surgery so I’m willing to try it.” Fair enough, I said. Let’s give it a shot.

    After four acupuncture treatments his symptoms have improved by 80%. For the first time in twelve years he is able to eat without having heartburn afterwards. And he’s been able to completely get off his medication.

    You may argue that this is “placebo”. You may be right. But I say, “so what”?

    What my patient is concerned with is getting rid of his problem. The western drugs, which arguably have a stronger placebo effect than acupuncture because of the widespread cultural belief in their effectiveness, weren’t able to help. Acupuncture has almost completely cured his problem in just two weeks. Do you think he cares whether it’s “placebo” or “active treatment”?

    Experiences like this are common in the clinic. That’s why more and more people are getting acupuncture, in spite of what the clinical research shows or doesn’t show. People want to feel better and address their health problems, and acupuncture helps them do that.

    I’m sorry this is so long but I still have a few points to make. My argument, as I already stated above, is that the concept of placebo has probably outlived its usefulness. Why? Because defining it in a consistent way that distinguishes it from specific treatment effects seems impossible. We should instead be focusing on the choice of outcome measure and the magnitude of the effect, rather than on interventions that are difficult to define.

    As an example, when compared with usual obstetric care, the presence of a support person during labor has dramatic effects on the use of analgesics, anesthesia, episiotomy, and cesarian section and the incidence of severe postpartum depression. (http://www.ncbi.nlm.nih.gov/pubmed/10796179)

    What do we call the additional effect of a support person, which are clearly measurable and are clearly producing real physiological changes? Do we call it placebo? Why? How is that distinguishable from the effects of the other methods of obstetric care?

    The answer is that it’s only distinguishable by the name and meaning we’ve applied to those effects. We’ve decided that they are secondary (and by implication less important) than the effects of the treatment being primarily studies. But I assure you that from the patient’s perspective (and one would hope the doctor treating her) those so-called “placebo” effects are no less significant in the outcome.

    Let me pose a hypothetical question for you. Say you could choose between two treatments for a particular condition. Both of these treatments have roughly the same outcome in clinical studies. However, one treatment carries significant side effects and risks, including irreversible physiological damage that in fact perpetuates and worsens the problem you sought help for. The other treatment is relatively free of side effects and risks, is well tolerated, and does nothing but improve your problem.

    Which would you choose?

    The scenario above is not hypothetical. It describes the choice a patient with depression has when deciding between an SSRI or a sugar pill. Sapirstein and Kirsch conducted a meta-analysis of 3,000 patients who received either antidepressants, psychotherapy, placebo or no treatment at all. They found that 27% of therapeutic responses were attributable to drug activities, 50% to psychological factors, and 23% to “non-specific” factors. In other words, 73% of the response to the drug was unrelated to its pharmacological activities – and antidepressants may be no better or more specific than placebos. (Read this article for more info: http://thehealthyskeptic.org/antidepressants-not-as-effective-as-research-suggests)

    If you’re going to be consistent with your terminology, you’ll have to refer to antidepressants (and some other commonly used drugs) as placebos. They are no less of a placebo than acupuncture is according to the research.

    Some doctors are in fact aware of this, and yet they go on prescribing them. They argue that the benefit of recovering from depression outweighs the ethical considerations of prescribing a placebo without the patient’s knowledge. Doctors are willing to do this in spite of the fact that these “placebos” (active drugs, in this case) have serious side effects and risks.

    If it were me, I’d take the placebo, thanks. I’ll get the same improvement without any of the risks.

    This is why I asked you what your definition of placebo is. The terminology is important here, because what you refer to as placebo is in fact a measurable treatment effect with a significant clinical outcome.

    Moreman and Jonas suggest what I believe is a more useful term, which is “meaning response”. They define the meaning response as “the physiologic or psychological effects of meaning in the origins or treatment of illness.” (http://www.annals.org/content/136/6/471.full) They provide many examples where the meaning ascribed to a treatment produces different outcomes.

    One of my favorites is a study in which 835 women who regularly used analgesics for headache were randomly assigned to one of four groups. One group received aspirin labeled with a widely advertised brand name. The other groups received the same aspirin in a plain package, placebo marked with the same widely advertised brand name, or unmarked placebo.

    In this study branded aspirin worked better than unbranded aspirin, which worked better than branded placebo, which worked better than unbranded placebo. Aspirin relieved headaches, but so does the knowledge that the pills you are taking are “good” ones. (http://bit.ly/9JxGe0)

    It is also known that placebo and acupuncture analgesia elicit the production of endogenous opiates. In both cases the analgesic effect can be blocked with injection of nalexone.

    So, to say that a treatment such as acupuncture “isn’t better than placebo” does not mean that it does nothing. Nor does it mean that acupuncture doesn’t improve clinical outcomes. Nor does it mean that acupuncture is not a better choice than another treatment with similar outcomes, when all factors are considered (side effects, risks, complications, etc.)

    In any event, while the research community goes on insisting that acupuncture is “just placebo” (an inane statement that reflects a lack of understanding of the nature and value of placebo), I will go on giving my so-called “placebo” treatment to patients and improving their health and quality of life.

    I leave you with a quote:

    “Most of us have been taught that western medicine is based on science, and science represents the unbiased, objective way to search for truth. However, subjective judgments lie behind all the facets of “scientific” research. (Not to mention that nowadays, most clinical research are funded by pharmaceutical companies)

    All data are theory-, method-, and measurement-dependent. That is, “facts” are determined by the theories and methods that generate their collection; indeed, theories and methods create the facts.

    This means that how the problem will be defined, which model(s) of inquiry will be considered to be relevant to the problem as defined, which model(s) of inquiry will be considered to be relevant to the problem as defined, where one shall look (and, by implication where on shall not look) for evidence–and even what one shall consider to be constitutive of evidence–are all determined by the paradigmatic “map” or world view to which the scientist is committed.”

    –John Ratcliffe, Notions of validity in qualitative research methodology

  31. BillyJoe says:

    “I agree that the literature doesn’t support the idea that point selection matters, or that needling technique makes a difference, or that even breaking the skin makes a difference.”

    So why did you become a acupuncturist?

    “However, I’ll also say that just because the literature doesn’t yet support a difference that doesn’t mean there isn’t one.”

    I don’t consider that a good reason to become an acupuncturist. Even it what you said is true, don’t you feel you need positive evidence of an effect rather than just no evidence of no effect before you decide to practise acupuncture?

    “In fact, a fairly recent study showed that leech therapy [bloodlettting] was more effective than surgery for osteoarthritis of the knee.”

    I’m guessing they compared leech therapy to arthroscopic lavage. Am I right? If so, it might interest you to know that Arthroscopic lavage has been shown to be no more effective than sham arthroscopic lavage. It may also interest you to know that the procedure has been largely abandoned for being ineffective (ie no better than placebo). Similarly for ligation of the internal mammary artery in the treatment of angina. It was found to be no better than sham surgery and has long since been abandoned.
    I think there’s a lesson there somewhere.

    “Last week a patient came in to the clinic with a history of severe acid reflux / GERD….After four acupuncture treatments his symptoms have improved by 80%. ”

    That is an anecdote.
    Here’s why anecdotes are underwhelming as evidence: I have to rely on your honesty in reporting the anecdote as described by the patient. You have to rely on the patients honesty in describing his history to you. Apart from honesty, there is the question of whether either or both of you have exaggerated the case or simply fooled yourselves in the retelling. I have to take it on trust that the patients GORD has been accurately diagnosed and that his present symptoms are indeed due to relux and do not represent psychosomatic symptoms for example. Maybe he has pancreatic cancer and your acupuncture has given him some temporary relief.

    “He came in saying “I don’t believe in this stuff…”

    Look that just makes me laugh, sorry. He doesn’t believe in this stuff, yet here he is fronting up for something he doesn’t believe in. I mean really. I don’t believe in the power of prayer, so I suppose I will just shuffle along to the parish priest when I need surgery instead! Goddamn how people can fool themseves – and others along with them!

    “Acupuncture has almost completely cured his problem in just two weeks. ”

    That would be a little difficult wouldn’t it, seeing as he walked into your clinic only a week ago? I only mention this little gaff to demonstrate the unreliability of testimony even when earnestly given. The more important point here is that you could have no idea – in only one or two weeks – that you have cured him.

    “Experiences like this are common in the clinic. That’s why more and more people are getting acupuncture, in spite of what the clinical research shows or doesn’t show.”

    People swear by homoeopathy as well, and for exactly the same reasons. Sorry, patient satisfaction surveys are no substitute for proper clinical trials.

    “My argument, as I already stated above, is that the concept of placebo has probably outlived its usefulness. Why? Because defining it in a consistent way that distinguishes it from specific treatment effects seems impossible. ”

    You know what the real problem is? Defining what acupuncture is. We started with the definition of traditional acupuncture and ended up with something that looks like sham acupuncture. No wonder you have trouble defining the placebo. You have taken it to define acupuncture!

    “http://thehealthyskeptic.org/antidepressants-not-as-effective-as-research-suggests”

    The phase 111 trials demonstrated an effect of antidepressants in patients with major depressive disorder. The follow up study demonstrated that these results are not applicable to patients with milder forms of depression. So, what is your point? It would seem to me that your headline should have read: antidpressants should be approved and precribed only for patients with major depressive disorder. If that had been you headline you would have had my admiration. But, it seems you’re too busy “challenging mainstream myths”.

    “If you’re going to be consistent with your terminology, you’ll have to refer to antidepressants (and some other commonly used drugs) as placebos. They are no less of a placebo than acupuncture is according to the research. ”

    Well, I just hope you can see your error now.
    If you ever decide to use acupuncture on a severely depressed patient, I hope you don’t end up with a dead patient on your hands.

    “Some doctors are in fact aware of this, and yet they go on prescribing them….If it were me, I’d take the placebo, thanks. I’ll get the same improvement without any of the risks.”

    Me? I’d just stop lying to people. And stop using useless treatments (whether its acupuncture or antidepressants in mild to moderate depression). Maybe, just maybe, patients might respond just as well to some empathy, care, and understanding. Is that such a novel idea?

    “what you refer to as placebo is in fact a measurable treatment effect with a significant clinical outcome”

    What you are saying here is that placebo works better than doing nothing. In other words, the patient who enters a clincal trial and gets paid attention to but gets the sugar pill does better than a patient sitting on the waiting list. Are you expecting me to disagree with that?

    “It is also known that placebo and acupuncture analgesia elicit the production of endogenous opiates. In both cases the analgesic effect can be blocked with injection of nalexone. ”

    Meaning what? That acupuncture works like placebo?
    Hmmm…I thought that was my argument!

    “So, to say that a treatment such as acupuncture “isn’t better than placebo” does not mean that it does nothing.”

    No. It means that acupuncture does not work better than placebo. It is really that simple.

    “Nor does it mean that acupuncture doesn’t improve clinical outcomes.”

    It doesn’t. The effect is due to placebo.

    “Nor does it mean that acupuncture is not a better choice than another treatment with similar outcomes, when all factors are considered”

    Agreed.
    Some placebos work better than others.

  32. apteryx says:

    Chris – as an educated member of the patient class, I hold pretty much the same opinions that you have expressed. Unlike BillyJoe (according to his previous remarks), I care far more for whether I am better off than why I am better off.

  33. BillyJoe says:

    We are just different then.
    I care for the truth first and foremost.
    Everything else comes second.

    And, in my opinion, any benefits based on the placebo effect of acupuncture can be achieved by methods that do not involve lying to patients.

    Also the areas in which the placebo effect of acupuncture can help are extremely limited. There is pain and not much else.

  34. apteryx says:

    As a scientist, I too am interested in truth. However, when people seem to be coming to different conclusions than I do because they have different philosophies about which truths matter or count and which don’t, I try not to reflexively respond by accusing them of “lying”, as if I were channeling Sen. James Inhofe.

  35. Chris Kresser says:

    You are still missing the point, Billy Joe. What you dismissively refer to as placebo can be the difference between pain and absence of pain, sickness and health, dysfunction and function.

    You say you care for the truth. But you make the mistake of assuming “truth” is found exclusively in randomized, “placebo”-controlled studies. Perhaps you should read this paper: “Why most published research findings are false” (http://www.plosmedicine.org/article/info:doi/10.1371/journal.pmed.0020124)

    Perhaps I should also remind you of the many times the so-called “truth”, as presented in the scientific literature, later turned out to be false. What are we to make of that, Billy Joe? For decades researchers told us that heart disease was caused by consumption of saturated fat and high LDL cholesterol. Today, the literature conclusively demonstrates that is not true. So what is “the truth”?

    While you cling to your myopic point of view, Americans are spending $34 billion on alternative healthcare. Why? Because 1) they’re dissatisfied with the level of care they receive from the medical establishment, and 2) they are getting better results from so-called “alternative” practitioners.

    You will likely argue that these millions of people are simply throwing their money away on placebos. But they don’t care what you think, because they’re feeling better and getting healthier.

    The arrogance and almost religious fervor of the “evidence-based” medicine movement never ceases to amaze me. Especially because there’s not much to arrogant about. The U.S. ranks just 34th in the world in life expectancy and 29th for infant mortality. Of 13 countries in a recent comparison, the United States ranks an average of 12th (second from bottom) for 16 available health indicators. (http://bit.ly/avlcYh)

    Even worse, a recent study by Dr. Barbara Starfield published in 2000 in the prestigious Journal of the American Medical Association demonstrated that iatrogenic incidents (events caused by medical intervention) are the 3rd leading cause of death in this country, causing more than 250,000 deaths per year. Only heart disease and cancer kill more people. (http://bit.ly/d0RZBW)

    Michael Jordan has a reason to be arrogant. Tiger Woods has a reason to be arrogant. The medical establishment does not.

    This will be my last post here because it’s abundantly clear that this is not a dialogue where information is being shared in an open-minded and exploratory spirit. I’m not even sure why I’ve gone this far. I’ve learned long ago that this type of debate is largely a waste of time.

    But I’ll leave you with a few quotes from a paper called “The placebo disavowed: or unveiling the bio-medical imagination” (http://bit.ly/dd3yHO) It’s a fascinating read for anyone interested in broadening their understanding of the significance of placebo in the context of healing.

    “What bioscience (especially in its contemporary corporate forms) wants and needs is to establish unilateral relations between material treatments and documented cures which it can then use both to demonstrate its practices and market its products…

    “The possibility that the imagination of a person who suffers might play a curative role within the event of illness is a priori excluded from scientific medicine because it introduces a non-reducible variable which de-centers bioscience’s self representation as the only legitimate agent of healing. Thus, within scientific medicine, the ‘placebo effect’ is bracketed not because it doesn’t produce healing effects, but *precisely* because it does.”

    “To designate an experience of healing as a ‘placebo effect’ is not only to set it apart from the ‘real’ domain of biochemical causality (while retaining the hope that some day a biochemical explanation may appear to account for this seeming deviation) but also to restrict the extent to which its existence can appear as a credible alternative to the deterministic claims of biomedicine. Moreover, the implicit causality retained within the placebo designation maintains the giver rather than the receiver of the placebo as the active agent in the healing process, despite some limited recognition that ‘placebo effects’ are a subclass of self-healing. Taken together, these interpretations defend advocates of bio-medicine against the recognition that they are *not the only agents of healing*, and moreover against the recognition that healing, whenever it occurs, occurs through the life process of the person who is ill…”

    And here are a few gems from Miller and Kaptchuk’s paper (http://bit.ly/9O0Hpw):

    “Healing resulting from the clinical encounter consists of a causal connection between clinician-patient interaction (or a particular component of the interaction) and improvement in the condition of the patient. That aspect of healing that is produced, activated or enhanced by the context of the clinical encounter, as distinct from the specific efficacy of treatment interventions, is contextual healing.”

    “The common description of the placebo effect as ‘non-specific’ is unsatisfactory. There is a valid contrast between interventions that have specific efficacy… and placebo interventions that do not. However, rigorous laboratory experiments have detected a variety of specific mechanisms underlying the reported effects connected with placebo interventions presented (deceptively) to research subjects as real treatments…. By virtue of causing a real change in a specified outcome, treatments that work only by means of the placebo effect *must work by some specific mechanism*. Just as placebo treatments with real effects are not absolutely inert, so they are not absolutely non-specific.”

    “Fixation on the efficacy of treatment intervention obscures the fact that the technological tools of medicine are always applied in some context, which itself may contribute significantly to therapeutic benefit.”

    “Instead of focusing exclusively on the therapeutic power of medical technology and thereby ignoring or dismissing context, we should see the context of the clinical encounter as a potential enhancer, and in some cases the primary vehicle, of therapeutic benefit.”

    “Attention to contextual healing signifies that there is more to medicine than diagnosing disease and administering proven effective treatments. This has long been recognized under the rubric of ‘the art of medicine’. However, biomedical science, animated by the search for specific therapeutic efficacy, has left the art of medicine shrouded in mystery.”

    Another perspective on “the truth”.

  36. BillyJoe says:

    apteryx,

    “I try not to reflexively respond by accusing them of “lying””

    Sorry, that was not meant as an accusation.
    I should have said “unintentionally lying”, or “unknowingly not telling the truth”, or “being a little inexact with the truth”.

    “people … coming to different conclusions … because they have different philosophies about which truths matter or count and which don’t”

    I think there are rare occasions where lying is acceptable. And I guess there IS a philosophy behind that. That philosophy being that you aim not to harm anyone. Sometimes lying to them is the only way to avoid that harm. A dying man doesn’t need to know that his favourite nephew just died in a car acident on his way to visit him.

    regards,
    BillyJoe

  37. rosemary says:

    Chris, “Perhaps you should read this paper: ‘Why most published research findings are false’ ”

    Assuming that is true, what is your point? That if objective studies usually give false conclusions, guesses and beliefs are more reliable than objective studies? If so, I think history proves you wrong.

    Chris, “You will likely argue that these millions of people are simply throwing their money away on placebos. But they don’t care what you think, because they’re feeling better and getting healthier.”

    I assume that you conclude that based on your personal experience. My personal experience leads me to a different conclusion. There are many people, I don’t have numbers, throwing their money away on snake oil. Many of them feel better after that for awhile for a variety of reasons only to eventually feel sick and unhealthy again. So they throw away more money on the same or a different kind of snake oil often continuing the procedure until the snake turns around and bites them. Most of the time they never tell their saviors, the healers or snake oil salesmen, that their “cures” were temporary and that they had to buy additional “cures”, and snake oil salesmen aren’t known for record keeping or follow up studies or even hearing unpleasant things so they usually never know.

    Chris, “The arrogance and almost religious fervor of the ‘evidence-based’ medicine movement never ceases to amaze me. Especially because there’s not much to arrogant about. The U.S. ranks just 34th in the world in life expectancy and 29th for infant mortality.”

    You have arrogantly and conveniently left out statistics comparing the health and longevity of those who live in areas where scientific medicine is available with those living in areas where it is not. Maybe you should visit Cambodia with Andrew or read some history.

    Chris, “This will be my last post here because it’s abundantly clear that this is not a dialogue where information is being shared in an open-minded and exploratory spirit.”

    Promises. Promises. You sound like one of the alts who writes to me out of the blue making all kinds of outrageous claims he can’t substantiate who is amazed when I respond, cannot provide me with the evidence I request, answers by calling me names and stating that scientific medicine, the kind based on independently verifiable evidence, is bad and therefore the alternative, the unscientific kind based on belief, has to be good. Then he adds that he is ending the conversation by which I understand that he has delivered his sermon and is leaving the room because he knows he hasn’t convinced anyone that he is correct and he knows that further discussion may just convince listeners how wrong his unverifiable beliefs are.

  38. BillyJoe says:

    Chris’ last post must have been in moderation when I looked last.

    He says he won’t be back so I guess there is no use responding. Just to say, that I thought all along that he must have an agenda below his calm exterior. His website says it all. Against all mainstream medicine and for all alternative medicine. Black and white with no shades in between.

    At least he showed his true colours in the end.
    And I expect we won’t see him back now that he has exposed himself so embarrassingly

    BJ

  39. BillyJoe says:

    …oh, and the extensive quotes – is that supposed to prove something like you can quote someone who agrees with you so you must be right?

    Anyway it’s a classic sign of someone who cannot actually think for themselves.

  40. daedalus2u says:

    There is real physiology behind the placebo effect.

    http://daedalus2u.blogspot.com/2007/04/placebo-and-nocebo-effects.html

    I would call the reduced need for pain treatment during birth with an attendant being due to the placebo effect. My definition of the placebo effect is any treatment that produces positive effects not mediated through either pharmacology or surgery (or some other physical intervention). Acupuncture is indistinguishable from a placebo.

    Having an attendant present does produce positive effects through a placebo effect because the individual feels safer because the individual actually is safer because there is someone there to help, or to summon more help if something goes wrong.

    Ignoring the pain of a stubbed toe when trying to catch your falling baby is placebo effect too. Pain is a signal. It is a signal that your body is exceeding physiological limits and damage is likely. Your body still lets you use body parts that are in pain, the pain is to let you know that damage is likely. When your baby is falling, the pain of a stubbed toe is inconsequential compared to the risk of injury to your baby, and so your body removes that distracting signal. The toe is not miraculously healed during that time, the pain signal is blocked to maximize function within the degraded physiological capacities of the already damaged toe, so as to better accomplish the needed task of catching the falling baby.

    I think it is ironic that acupuncturists lament millions wasting money on placebos, rather than spending money on acupuncture, a field they profit from and are actively promoting.

    It is also ironic to blame Harriet’s unwillingness to accept acupuncture as different than placebo because of her philosophy, without mentioning what her philosophy is, the philosophy of requiring data to support beliefs. Blaming her lack of belief in acupuncture as being due to her philosophy rather than to the lack of data supporting acupuncture is disingenuous. If acupuncture actually did work better than placebo, it would be possible to get good, substantial and reliable data that it is better. There is no such data. If there was data, Harriet’s philosophy would require her to believe it was effective.

  41. BillyJoe says:

    “There is real physiology behind the placebo effect.”

    That is news to Chris who seems to think that, because there is a physiological change when acupuncture is administered, it must be working. He wants to compare acupuncture to a placebo that has no physiological effect so that he can actually claim the placebo effect for acupuncture.

    And, yes, he has provided no evidence that acupuncture can heal, though he has made that claim.

    Anyway, there is little headway to be made with someone who rejects science when it disturbs his cosy little view of the world.

  42. AppealToAuthority says:

    Its pretty clear there is no strong evidence of acupuncture itself (specifically placed needles) being good for anything, and only glimmers of any possible mechanism by which it might work.

    But as often happens, in enthusiasm to dismiss it, people step over the mark. Eg:

    “Looking for a mechanism seems premature when there is no compelling evidence acupuncture works beyond placebo.”

    – I would have thought that positing a mechanism prior to finding or looking for evidence is part of establishing plausibility. We can’t have it both ways: if SBM requires plausibility, we have to accept people looking for possible mechanisms. Some medical research starts from a plausible mechanism and looks for effects; others start from an observed effect and looks for a mechanism. The problem lies with practices which are already established in supposedly therapeutic practice, which are being tested scientifically after the fact. Acupuncture is one such.

    I think Dr Novella’s previous article about placebos is instructive. “It is any and all measured effects other than a physiological response to the treatment itself.”

    For psychological ailments, this would have to be modified to “any and all curative effects other than a response to the treatment itself.” However, this is a problematic definition with depression, because:
    a) It is not always possible to separate responses to the treatment itself from other effects, even with a well-designed and controlled study;
    b) CAM proponents repeatedly redefine what _they_ mean by “the treatment”.

    However, that shouldn’t stop us trying. It is worth funding studies to provide evidence to help eliminate bad medicine of all kinds, particularly where there is already an entrenched non-scientific practice.

    But they must (among other things):
    * be well-designed
    * test for medical effectiveness, not just statistical significance
    * be thoroughly desk-tested at the experiment design stage, to ensure that the definitions of controls and placebo are pre-determined and agreed by factional interests.

    And then when they are reported, just don’t use the word ‘placebo’. It has too many meanings, and doesn’t help. “Acupuncture works better than doing nothing” or “Acupuncture works better than random poking” are worth testing and reporting. I’ve got a pretty good idea what the answers would be, but a lot of ‘health consumers’ (ie sick people) out there are much less clear, and more hopeful. We need quality information to talk to them with. I think outcome testing is the best information to use.

  43. BillyJoe says:

    AoA.

    “I would have thought that positing a mechanism prior to finding or looking for evidence is part of establishing plausibility.”

    Yes, but what about treatments that are already popular despite there being no plausible mechanism. It seems to me you have to procede to clinical trials despite a lack of plausible mechanism in these cases.

    “We can’t have it both ways: if SBM requires plausibility, we have to accept people looking for possible mechanisms.”

    Not if a wealth of clinical evidence already tells us that there is no benefit. Why search for a mechanism in that situation?

    “Some medical research starts from a plausible mechanism and looks for effects; others start from an observed effect and looks for a mechanism.”

    Nothing wrong with that in my book.

    “The problem lies with practices which are already established in supposedly therapeutic practice, which are being tested scientifically after the fact. Acupuncture is one such.”

    If a treatment is popular but unproven (because of a lack of clinical trials showing whether or not it works) then, despite lack of a plausible mechanism, we have no choice but to conduct those clinical trials.

    BJ

    BJ

  44. GeoffreyCoe says:

    Thanks to Steve for the review of the RCT.

    As a group pregnant women do not have medication options so it is rational that they will seek alternatives and have a right to have these treatments evaluated for their effectiveness.

    While I take the point about the lack of blinding in the massage group, blinding of patients is actually hard if not impossible to achieve in physical therapies. We have this problem in physical therapy clinical trials all time; unlike dishing out a tablet which is easy to conceal whether or not it has an active ingredient, there is usually considerable skill, and interpersonal contact involved in delivering the therapy. I understood the logic behind the massage was that you WOULD have a treatment that would produce some non specific effects just as would be the case in the two acupuncture groups and it was to see if there was any added benefit of depression-specific acupuncture

    The results were completely disappointing and the analysis and discussion were misleading in suggesting the improvement was useful. As is often the case with hypothesis testing. statistical significance was really quite clinically meaningless in this case. The effect size was unimpressive and with very little certainty that the effect size was clinically meaningful. And as Steve said a mean difference between the controls and the acupuncture for depression of 2.5 points on the Hamilton Rating Scale is virtually useless especially given the amount of treatment required. One would expect to get twice that amount of improvement just by going for a 10 minute walk.

Comments are closed.