Articles

Benedetti on Placebos

There has been an ongoing debate about placebos on SBM, both in the articles and in the comments. What does it mean that a treatment has been shown to be “no better than placebo?”  If our goal is for patients to feel better and they feel better with placebos, why not prescribe them? Do placebos actually do anything useful? What can science tell us about why a patient might report diminished pain after taking an inert sugar pill? The subject is complex and prone to misconceptions. A recent podcast interview offers a breakthrough in understanding.

On her Brain Science Podcast Dr. Ginger Campbell interviewed Dr. Fabrizio Benedetti, a physician and clinical neurophysiologist who is one of the world’s leading researchers on the neurobiology of placebos. A transcript of the interview [PDF] is available on her website for those who prefer reading to listening. The information Dr. Benedetti presents and the expanded remarks by Dr. Campbell after the interview go a long way towards explaining the placebo phenomenon and its consequences for clinical medicine. Dr. Campbell also includes a handy list of references. I’ll try to provide a summary of the main points, but I recommend reading or listening to the original.

A common misconception is that the response to placebos is a purely subjective psychological response involving only the cortical level of the brain; but evidence is accumulating that real, measurable, objective subcortical neurophysiologic phenomena are involved. One of the first hints was a 1978 study showing that the placebo response to pain could be blocked by naloxone, a narcotic antagonist drug, indicating that the placebo must have actually caused an increase in endogenous opioids.

Placebos are used in clinical trials methodology, where a placebo group is compared to an active treatment group. Typically, about a third of subjects in the placebo group show improvement, but this doesn’t mean all of them have responded to the placebo. There are many other factors that can cause improvement after administration of any treatment, whether effective drug or placebo, including spontaneous remission, regression to the mean, and various kinds of bias on the part of patients and experimenters.

Clinical trials are done to look for a difference between the outcomes with placebo and the outcomes with the drug. Researchers in those trials are not interested in trying to understand why there is an improvement in the placebo group. That requires a different kind of study. Dr. Benedetti is using “placebo balanced design” to tease out the influence of verbal suggestions — expectations — on the action of drugs. Subjects are divided into four groups. The first group of subjects receives the active treatment and is told it is the active treatment (the truth).  The second group receives the active treatment and is told it is placebo (a lie). The third group receives placebo and is told it is the active treatment (a lie). The fourth group receives placebo and is told it is placebo (the truth). This design allows researchers to separate the specific effects of the active treatment itself from the nonspecific effects of suggestion and expectation.

In studies of surreptitious vs. open IV morphine administration, patients experienced less pain relief from the morphine when they were unaware that they were getting it. In one brain imaging study, the metabolic response to methylphenidate decreased by 50% when subjects were told they were getting a placebo.

The placebo response is not limited to pain. In Parkinson’s disease, placebos have been shown to cause a 200% increase in dopamine release in the brain, with an alteration in the activity of neurons and a corresponding clinical improvement in motor function. Dr. Benedetti has even documented a physiologic placebo response at the level of single neurons. In deep brain stimulation studies, the stimulator doesn’t work as well if patients are told it is off when it is actually on.

There is not just one “placebo effect,” but many different placebo effects that work by different mechanisms, including (1) anxiety reduction, (2) activation of the reward mechanism (with dopamine release in the nucleus accumbens), and (3) learning.

Placebo responses can be divided into two types: conscious and unconscious. Conscious responses involve suggestion and expectation. Unconscious responses occur with classical Pavlovian conditioning.  If you give morphine 3 days in a row and on the 4th day you replace it with a placebo, almost 100% of patients will respond to the placebo as if they had received morphine. When a pill is administered, there is a completely unconscious association between two stimuli: the pharmacologic effect and the psychosocial context, which may include such things as the color and shape of the pill. Unconscious conditioning has been demonstrated in both animals and humans.

Conscious expectations are important for conscious physiological functions like pain and motor performance. Unconscious conditioning is more important for unconscious physiological functions like hormone secretion and activation of different immune mediators. Dr. Benedetti gives this example: if you give a patient a placebo and tell him it will increase his growth hormone levels, nothing happens. But if you give a drug that really does raise GH levels for two days and on the third day replace it with a placebo, the levels will rise with the placebo.

Clinical Implications

Dr. Benedetti gives placebos to subjects in a research setting; he doesn’t recommend giving them to patients in the doctor’s office. Placebo studies are important for understanding how the brain works, but Dr. Benedetti is not sure that they will lead to any clinical application. He says

the take-home message for clinicians, for physicians, for all health professionals is that their words, behaviors, attitudes are very important, and move a lot of molecules in the patient’s brain. So, what they say, what they do in routine clinical practice is very, very important, because the brain of the patient changes sometimes… there is a reduction in anxiety; but we know that there is a real change…in the patient’s brain which is due to… the ‘ritual of the therapeutic act.’

Experimentally, morphine requirements can be reduced by starting with morphine, substituting a placebo for later doses and periodically reinforcing the conditioning with morphine. This sounds like a good thing, but before it could become a useful option in clinical practice, it would have to overcome a number of ethical and practical hurdles. The unconscious placebo responses require conditioning and the conscious ones are problematic too. Even if some patients might get a degree of pain relief from just being handed a sugar pill with a strong suggestion, the effects can’t compete with effective pain treatments: the response is generally smaller, less reliable, less predictable, and not sustainable over the long term. I can’t imagine that ever becoming standard medical practice, and not just because of the ethical issues.

The doctor-patient relationship is essential to medicine, and placebo effects are an inherent part of that interaction. Communication is vital. Doctors must tell patients what a treatment is supposed to do. When pain medication is given without the patient’s knowledge, it doesn’t work as well. A clear understanding of the diagnosis is important: a positive diagnosis of something menacing like cancer can make the patient anxious and more aware of symptoms, while a negative diagnosis can reassure, relieve anxiety, and divert attention elsewhere. Lying to patients would undermine a trusting doctor-patient relationship and ultimately even interfere with the ability to evoke placebo responses. Prescribing placebos is uniformly rejected by medical ethicists: instead, we can put our increasing knowledge of placebo neurophysiology to good use without lying to patients.

 

 

Posted in: Basic Science, Neuroscience/Mental Health

Leave a Comment (134) ↓

134 thoughts on “Benedetti on Placebos

  1. nybgrus says:

    Dr. Hall – I have a question. I haven’t listened to Benedetti’s podcast yet but I will tomorrow.

    My take on the placebo effect – at least as it has been described in the usual common and broad way – is that it is nothing more than study artifact and reporting bias/error. This is because I tend to view things like what Benedetti describes on a neurophysiological level as an active intervention. CBT, for example, is a non-pharmacological way to improve psychiatric states. To me this is evidence of neural pathway restructuring which includes new synaptic formations, increased NT release, and Hebbian learning. But CBT is also an active therapy. These responses from the placebo administration indicate to me that the expectation/interaction is an active psychological intervention, restructuring neural pathways in the brain. Sometimes this can be synergistic with the medication being given, sometimes antagonistic, and there is always the “nocebo” effect as well.

    Since all neural pathway changes must start at the single neuron level, it makes sense to me to find that as well.

    My question is – am I totally off base or being overly pedantic on how I define placebo? It seems useful and important to me to separate the active psychological interventions (and their limitations) noted in such placebo studies from the actual placebo itself. Perhaps I am being overly pedantic and hypersensitive because of the way CAM and IM has begun to adopt placebo effects as meaningful and useful medical interventions. I guess I’m just stuck on the traditional definition of placebo being inactive, and thus any “effect” from it must be artifact and bias/error. Including what we can now reasonably deem as active interventions under the placebo is not something I am comfortable with – but I am open to insight.

  2. nybgrus says:

    sorry – I was writing quite quickly before sitting down for an officially proctored NBME practice exam and I don’t think my last statement was quite clear enough.

    What I meant to say is: I think that the part of the “placebo effect” that actually effects the sorts of objective changes Benedetti notes is separable and definable as an active invervention in the same way CBT and other forms of counseling can be considered active interventions. Perhaps we could use some convenient eponymous term to describe that separate active intervention. Perhaps even find reasonable ways to control for it and thus get an even better resolution on effect size with RCTs. Or perhaps I am being too pedantic or overzealous.

    I am very curious as to what you and the commentariat think, though.

  3. Gregory Goldmacher says:

    I recall in Ben Goldacre’s book Bad Science there is a whole chapter on the placebo effect, and I think he cites examples of measurable physiological phenomena like gastric ulcers decreasing in size with placebo treatment. This suggests that there is something other than just changes in perception of pain occurring with placebos. However, I may be misremembering what I read, and the book is at home while I’m in my office and can’t look it up now. Can anyone confirm or correct my recollection?

  4. nybgrus says:

    I actually haven’t read Goldacre’s book (though I have heard him speak and overall I think he is pretty good) but I really should.

    I’m also not familiar with studies on ulcers as you described. But I’d venture a guess that the ulcers were able to recede because stress increases sympathetic tone, which induces cortisol response, which decreases immune response, which decreases prostaglandin production, which makes ulcers worse.

    That yet-to-be (to my knowledge) epnoymous part of placebo to which I refer would be acting to reduce anxiety and thus stress and increasing prostaglandin protection of the gastric mucosa so the ulcer may heal.

    I could be wrong on that, but the mechanism would seem relatively straightforward. But am I wrong to tease that out and call it an active intervention? It’s that to me, the placebo wouldn’t be healing the ulcer. The active psychological treatment of reassurance and comfort from the doctor would be effecting those neural changes and thus the downstream effect. Am I being overzealous or overly pedantic? And that truly is a genuine question.

  5. WilliamLawrenceUtridge says:

    That’s really weird. From what I learned in undergrad psych classes, at least one of those effects is exactly the opposite of what I’d expect. I was taught that when a drug is anticipated, the body overcompensates in the opposite direction to maintain homeostasis. Heroin increases pleasure, decreases respiration and pain, and prevents orgasm. An addict sees heroin and their subjective response, the withdrawal symptoms, are the opposite of the drug effects – subjective misery, hyperventilation, increased sensitivity to pain and spontaneous orgasms.

    In this case, you give someone a drug, and I would expect the endogenous levels of said drug (or its analogue) would drop. Instead it increases! I’m confused, anyone have any ideas?

    Fascinating!

  6. Gregory Goldmacher says:

    @ nybgrus:

    The physiological mechanism you describe sounds perfectly reasonable and consistent with my understanding of the physiology behind ulcers (caveat – I am a radiologist, so my knowledge of GI physiology is limited).

    I guess whether this is a placebo effect of not depends on where exactly we place the boundaries of what we call a placebo effect. To my mind, broadly, the placebo effect combines the psychological elements of expectation, reassurance, comfort, etc. It’s everything that is based on the interaction with the person providing treatment, rather than on the physiological changes induced specifically by the treatment itself (drug binding to specific receptor, etc. etc.).

    Your definition of the placebo effect may be somewhat more constrained. I do not know if there is a technical definition that draws clear boundaries… but this is certainly the forum where I’d expect someone will let us know.

  7. Harriet Hall says:

    @nybgrus,
    ” It seems useful and important to me to separate the active psychological interventions (and their limitations) noted in such placebo studies from the actual placebo itself.”

    Yes, and that’s the whole point of Benedetti’s studies. The placebo itself does nothing, but the ritual of the therapeutic act does. There are nonspecific psychological and physiological responses that occur when a substance that has no specific effect is given. The psychological and physiologic responses have been traditionally lumped under the term placebo. It’s a useful term, widely accepted; and trying to change the terminology wouldn’t work any better than trying to eliminate the term CAM. Rather than rejecting it, I would propose being careful when we use it, qualifying and clarifying what we mean.

    In RCTs, the response of the placebo group does include study artifact and reporting bias/error, but it also includes physiologic responses. It is not important to separate these out to tell whether the active treatment works. It is important to separate these out to understand what is happening to those in the placebo group, and that is what Benedetti’s approach is intended to elucidate.

  8. Harriet Hall says:

    @Gregory Goldmacher,

    “I think he cites examples of measurable physiological phenomena like gastric ulcers decreasing in size with placebo treatment. This suggests that there is something other than just changes in perception of pain occurring with placebos”

    Did you read what I wrote? I gave examples that clearly demonstrate that there is something other than just changes in perception of pain. In fact, that is the main point of my post.

  9. Harriet Hall, thanks so much for the overview and link. Got a chance to listen to the interview and it was really interesting, even to a laypersons like me.

    My son is slated for surgery in a couple months. It occurred to me listening to the interview that often he has been given pain and nausea medication, through the IV, without specifically telling him what he is getting. Maybe we can increase the benefit of those medications by communicating more clearly what the medicine should do. “This will make your tummy feel good. This will take away the pain.”

    @nybrgus, you gotta listen to the interview. HH does a great job summarizing, but there is still more details about the subgroups of the placebo responses he’s found that offer a lot of clarity.

    Good stuff.

  10. Gregory Goldmacher says:

    Hi Harriett,

    I did read what you wrote, and I was essentially offering further evidence in support of that idea while responding to the earlier comment. However, since I am not an expert in this and since I could not clearly recall the citation, I was being cautious.

    This raises some very interesting questions about the limits of placebo responses, or the extent to which they can reach beyond what can be controlled cortically. The data you cited about morphine and methylphenidate are nice examples about the effects of expectation on states like pain, concentration, etc. that seem amenable to conscious or semi-conscious control. This seems to me to be less surprising than findings such as placebo effects on Parkinson’s. Reaching outside the CNS entirely, such as in the GI ulcer example, is even more fascinating.

    ~G

  11. WilliamLawrenceUtridge says:

    MIM, depending on how old your son is, he might benefit more if it came from you than a doctor or nurse. Or both! Come to think of it, a study on adult versus child placebo reactions would be pretty interesting…

  12. ceekay says:

    @nybgrus

    Your point is a reasonable one– re-naming active responses that accompany placebo administration according to the putative specific mechanism:

    e.g., expectation, therapeutic ritual etc

    This would help us remember that placebo is nothing more than an inactive comparator….

    A caveat here is that this re-naming would not prevent CAM practitioners from appropriating said medicine…

    Thus, instead of asserting the superiority of their treatments in eliciting placebo effects, they could assert that their “therapeutic ritual” elicited better responses than the ritual used by conventional, evidence-based physicians.

  13. wales says:

    According to HH “A common misconception is that the response to placebos is a purely subjective psychological response involving only the cortical level of the brain; but evidence is accumulating that real, measurable, objective subcortical neurophysiologic phenomena are involved. “ Based on F. Benedetti’s interview.

    According to SN “Placebo effects are largely an illusion of various well-known psychological factors and errors in perception, memory, and cognition – confirmation bias, regression to the mean, post-hoc fallacy, optimism bias, risk justification, suggestibility, expectation bias, and failure to account for multiple variables.”

    After reading the Benedetti interview it appears that SN’s comments are dismissive of some potentially important information, or perhaps he hasn’t read the interview.

    Either HH or SN care to weigh in, or are these two opinions irreconcilable?

  14. cloudskimmer says:

    I have a question about how long these studies are conducted. In watching someone close to me try unsuccessfully to solve a chronic pain problem, everything “worked” for about three months, and then the pain gradually returned. At this point, nothing works. Over the long term, are placebo interventions invariably going to diminish in effect because they aren’t doing anything real to address the problem? And how much does it depend on experience and perception? It’s hard to conduct studies lasting for months, so what I have observed would generally not be found in most studies.

  15. Harriet Hall says:

    @wales,
    SN’s comments were qualified by the word “largely” and refer to placebo-controlled studies. So no conflict.

    @cloudskimmer,
    I don’t think there are any long-term studies, but there have been many observations like yours of waning placebo effects over time. As I pointed out “the effects can’t compete with effective pain treatments: the response is generally smaller, less reliable, less predictable, and not sustainable over the long term.” And I wouldn’t be surprised if repeated disappointments could condition the patient to expect less from placebos in the long run and so be less responsive to them.

  16. wales says:

    Ah, I see, lots of wiggle room.

  17. WLU, Yes, hard to tell which spokesperson would elicit the best response, possibly the volunteer who delivers the little toys to keep kids occupied would impress my son the most. :)

    And it is also good to remember that kids are not just small adults, that their developmental difference or some physiologic (is that the right word?) difference might mean differences in placebo responses.

  18. I thought this was an excellent analysis of the placebo proper as well as the relationship between placebo effects and non-specific treatment effects in clinical trials.

    However, the question it raised for me, which didn’t seem to be directly addressed, is whether physiological placebo responses, due to conscious or unconscious causes, should be viewed as “real” treatment effects. The CAM folks seem to be developing the argument that even if their interventions don’t outperform placebo in clinical trials, the placebo effect meaningfully alters the disease process and so should be viewed as a valid form of treatment. The information in this interview about neurochemical and immune variables altered by unconcious placebo effects (e.g. classical conditioning) would seem to support this line of argument, and I expect CAM proponents are likely to cite this work as evidence that CA “works.”

    Any thoughts or responses?

  19. BillyJoe says:

    Harriet,

    “A common misconception is that the response to placebos is a purely subjective psychological response involving only the cortical level of the brain; but evidence is accumulating that real, measurable, objective subcortical neurophysiologic phenomena are involved“

    Apologies if this is a silly question.
    (I haven’t had time to view the podcast yet)

    Doesn’t the cortical psychological response produce the subcortical neurophysiologic response. Otherwise what is the (non-cortical) mechanism that produces the subcortical neurophysiologic response?

  20. BillyJoe says:

    …in other words, don’t those ulcers heal as a result of the downstream effects of the cortical psychological response. If not what actually is healing these ulcers? It’s not magic after all.

  21. Harriet Hall says:

    @Brennen McKenzie,

    Physiologic responses to placebo are certainly “real” but they are nonspecific. Leading patients to think the placebo treatment has a specific effect constitutes lying, and lying to patients is not only unethical, it is ultimately counterproductive.

    Since different placebos can produce the same nonspecific response, there is no basis for choosing one CAM placebo treatment over another, and no basis for preferring any of them to a treatment that also has specific effects.

    The CAM folks’ arguments ultimately undercut the whole rationale for what they are offering. If acupuncturists really believed acupuncture was justified for the placebo response, they could offer sham acupuncture without skin penetration, since it has been shown to work as well and couldn’t cause any of the risks associated with needle insertion. And they could give up all the rigmarole about acupoints, since treatment of non-acupoints works just as well.

  22. Harriet Hall says:

    @BillyJoe,

    “Doesn’t the cortical psychological response produce the subcortical neurophysiologic response.”

    Yes, but only for the conscious kind of placebo effects, and in essence, the placebo is affecting the cortex and the cortex is affecting the subcortex. In the other kind of placebo effects, the unconscious ones, the cortex is not involved at all. The misconception I was referring to was the earlier view that the placebo effect was “all in the mind” with no neurophysiologic correlates at all.

  23. pmoran says:

    Harriet: Since different placebos can produce the same nonspecific response, there is no basis for choosing one CAM placebo treatment over another, and no basis for preferring any of them to a treatment that also has specific effects.
    .
    The CAM folks’ arguments ultimately undercut the whole rationale for what they are offering. If acupuncturists really believed acupuncture was justified for the placebo response, they could offer sham acupuncture without skin penetration, since it has been shown to work as well and couldn’t cause any of the risks associated with needle insertion. And they could give up all the rigmarole about acupoints, since treatment of non-acupoints works just as well.

    Thank you for airing this material, Harriet — long overdue.

    Surpise, surprise :-) , I disagree with most of what you say in that quoted above.

    I think you yourself have referred to studies suggesting that pill color produces different placebo effects. There is also little doubt that procedural placebos perform better than pills.

    In actual CAM practice patients mostly use treatments because of the testimonials or recommendations of other people and that provides their “basis for choosing one CAM placebo treatment over another” as well as predictably influencing outcomes.

    There might be a basis for preferring a placebo over a treatment that has “specific effects” if in widespread use it satisfies some of the medical need but with fewer serious side effects. (e.g. glucosamine vs NSAIDs for osteoarthritis).

    Sham acupuncture may only work as well as real acupuncture when the patients believe that the skin is being penetrated.

    We have to look at everything differently when trying to understand the psychogenic aspects of medical interactions. It is not easy to shake off the mind-sets imposed by simpler therapeutic concepts.

    While I have, via slightly different pathways of reasoning, tentatively arrived at the same conclusion as you, i.e. that the mainstream should not be embracing CAM in any systematic way, I do think all the uncertainty about placebo influences should have some influence upon how we react to CAM and CAM practitioners.

    We have no clear mandate regarding some of their lesser claims.

  24. “Physiologic responses to placebo are certainly “real” but they are nonspecific….Since different placebos can produce the same nonspecific response, there is no basis for choosing one CAM placebo treatment over another, and no basis for preferring any of them to a treatment that also has specific effects.”

    Excellent and very clear, thank you. It seems like the examples of Parkinson’s and pain muddy the issue, because the non-specific placebo effects happen, in these cases, to have physiological effects that directly address the pathophysiology of the disease. But of course we have to remember that a) the exact same non-specific effects will occur in placebo “treatment” of other diseases for which the unconcious effects will NOT be relevant, and b) of course a therapy with specific effects as well as non-specific effects, when available, is always preferable to a pure placebo.

  25. nybgrus says:

    I will listen to the podcast after this comment and see if I get anything new out of it.

    So it seems that I can sum up the responses as follows (including Dr. Novella’s piece):

    The placebo effect is a large and nebulous descriptor that lumps in all the variables we can’t reasonably control for in experimental design so we can just account for the chunk wholesale. For clarity we could, and should, delineate that the majority of the effect is logical fallacy, bias, artifact, etc and that the active pyschological part can have physiological outcomes in limited circumstances. It can be further stated that the effects are highly variable, typically quite short lived, and demonstrate diminishing returns on repeated use.

    Nobody seems to think that my clear distinction and labeling of the psychological aspects as an active intervention separate from the placebo itself is incorrect, just that trying to reformulate the label and definition of “placebo effect” to be distinct from said active psychological intervention is less useful than carefully and consistently applying the above caveat.

    I’m honestly not sure how much I like that. Words hold great power – which is why we see quackery constantly rebranding itself into CAM and IM. So when I read Dr. Hall writing:

    The misconception I was referring to was the earlier view that the placebo effect was “all in the mind” with no neurophysiologic correlates at all.

    I cringe a bit, because to me it is the active psychological intervention, completely independent from the proposed therapeutic modality in question, that effects those neurophysiologic correlates. I think that, of course, Dr. Hall recognizes what she means when she says this, but most people would not. And the same way we discuss the bait and switch of CAM, I can see this causing problems. It isn’t a bait and switch in this case, but an unintentional confabulation which can lead to videos such as this one that talk about the “power of placebo” and how “placebo effects can be so strong that people want them banned from sports.”

    The problem I see is that even amongst med students and physicians, most people want a sound bite. Saying “placebo effect is just as good as anti-depressants” is a sound bite. It makes people think that the placebo pill is just as good as the SSRI. And most people don’t know or ignore the active psychological part of the placebo. Taking the time to explain the different aspects and applications of the “placebo effect” and why that matters means you’ve already outlasted the average attention span by at least five fold – and that’s if you’re quick and concise. And ultimately you’re left with eyes that glaze over and a response of, “Well it made me feel better so that means it worked.”

    Sorry, I’m rambling a bit because I find this to be interesting, nuanced, and important. And frustrating.

  26. nybgrus says:

    @michele: I absolutely agree. When I worked in the ER and assisted in reducing fractures in children (or anyone for that matter) I would always take extra time to explain what the medicine was, what it would do, and how it would work (in appropriate terms of course). I would also be quite up front and say, “This is going to hurt. My goal isn’t to hurt you and the medicine will really help take the edge off, but this is a broken bone and no matter how hard we try it will always hurt some. But I am pretty good at what I do, so I can do it quickly, and that means most of the time it doesn’t hurt as much as you might think it would. But I want you to be prepared that it will hurt at least a little.” And I would say that at least 90% afterwards looked at me and said, “Wow! That really wasn’t bad at all!”

    I think that giving people an expectation of effect is vital but I also think giving them a realistic expectation of consequences and outcomes is supremely helpful. It empowers them because they aren’t dealing with an unknown black box. Sometimes, when the patient is particularly freaked out, I would take extra time and explain things in more detail and ask them about any specific fears they may have. Much more often than not, that resolves the issue. The child life specialists we had at my hospital would always check to see if I was on duty if they had a kiddo that needed sutures or reductions, and I took that as both a very big compliment and a sign that what I was doing was likely effective.

  27. nybgrus says:

    @pmoran:

    For the life of me I still can’t figure out what the hell your stance actually is. You seem to consistently be against CAM, but at the same time are so wishy-washy with how you choose to interpret and apply evidence that you come across as quite paralyzed in your ability to actually say anything concrete.

    In actual CAM practice patients mostly use treatments because of the testimonials or recommendations of other people and that provides their “basis for choosing one CAM placebo treatment over another” as well as predictably influencing outcomes.

    That is not addressing what Dr. Hall said. She wasn’t commenting on why individual patients pick a specific CAM. She was commenting on a rational assessment of how to pick one CAM over another.

    You are essentially saying “People pick Mazdas over Toyotas because a friend said so and that is justification for distinguishing and picking Mazdas.” Dr. Hall is saying, “There is no rational justification beyond personal preference for picking Mazdas over Toyotas since there is no material difference between the two.”

    There might be a basis for preferring a placebo over a treatment that has “specific effects” if in widespread use it satisfies some of the medical need but with fewer serious side effects. (e.g. glucosamine vs NSAIDs for osteoarthritis).

    And now you are back to advocating the use of placebos in the management of patients. Dr. Hall’s point is “Why pick glucosamine?” After all, glucosamine doesn’t actually do anything for OA. So why not tell them to go pray? Or go to a day spa? Or get a massage? Or watch a movie and have a nice dinner with a loved one? Or drink water upside down from the far side of a glass? The only reason for recommending glucosamine over any of those others is because there is a big cargo cult of CAMsters that claimed specific effects for it in OA. But as a physician, you should know better, and recommending it instead of an NSAID is prescribing a placebo which is unethical – period.

    Personally, I would feel rather comfortable telling a patient that a massage or a relaxing dinner would help with their symptoms – because that is true. I would say to keep the NSAIDs around PRN but that said lifestyle changes would likely reduce that need. But I would not tell them the massage or dinner would help with the OA itself – since it wouldn’t. The same way glucosamine would not help with the OA. Telling them it will is a lie. Leaving it open, knowing that they will find information saying it will is a lie by omission.

    Sham acupuncture may only work as well as real acupuncture when the patients believe that the skin is being penetrated.

    I’m lost as to the point of this. Or how it in any way refutes what Dr. Hall has been saying.

    We have no clear mandate regarding some of their lesser claims.

    I read this as, “We can be wishy washy about things with little clinical significance and turn a blind eye when convenient.”

  28. pmoran says:

    Nybgrus:For clarity we could, and should, delineate that the majority of the effect is logical fallacy, bias, artifact, etc and that the active pyschological part can have physiological outcomes in limited circumstances. It can be further stated that the effects are highly variable, typically quite short lived, and demonstrate diminishing returns on repeated use.

    We are actually limited in what we can say about the potential of non-specific psychogenic influences, because most of what we believe about them is obliquely derived from studies that are not designed to provoke them.

    You allow that they are “highly variable” which allows for the possibility that a charismatic “healer” will have more spectacular results with suitably primed patients than can be easily captured within prospective clinical studies.

    And that is actually what CAM requires us to be able to explain.
    It is that kind of story that sustains CAM in the minds of both users and practitioners and presumably contributes to it being a remarkably resilient multi-billion dollar industry.

    I would also prefer that this were not so, but that “science-based” tag places certain constraints upon what we can claim to know for certain.

    “Limited circumstances” is also unclear. There are subjective elements to most illnesses. I would like to know, for example, whether acupuncture could reduce opiate requirements in patients that are open to that treatment. That would make me think again about about some matters.

  29. pmoran says:

    Sham acupuncture may only work as well as real acupuncture when the patients believe that the skin is being penetrated.

    I’m lost as to the point of this. Or how it in any way refutes what Dr. Hall has been saying.

    The point is that part of the placebo potential of acupuncture will be conveyed by the perception that it is doing something invasive. So poking the skin with toothpicks (“they could offer sham acupuncture without skin penetration”) would almost certainly not have the same effect when it became obvious to the patient.

  30. nybgrus says:

    You allow that they are “highly variable” which allows for the possibility that a charismatic “healer” will have more spectacular results with suitably primed patients than can be easily captured within prospective clinical studies.

    I should then act the further caveat that they are highly variable but the maximum effect size is very small. And yes, of course, a doctor who is charismatic, personable, and likeable will elicit a stronger psychogenic placebo response. And 2+2=4. You are not demonstrating anything profound in that statement, nor anything I have not taken well into account (in case you hadn’t read my other posts on the topic).

    We are actually limited in what we can say about the potential of non-specific psychogenic influences, because most of what we believe about them is obliquely derived from studies that are not designed to provoke them.

    I don’t go on belief. I go on data and evidence and integrate that with scientific understanding and logic. I also like to toss in a dash of ethics while I am at it. And the data demonstrate pretty consistently that the potential of psychogenic influences of placebo is quite limited to situations where a plausible and physical mechanism can be used to explain the outcome – much like my quick breakdown of placebo effects on ulcers. The data also demonstrate that the effect in most cases where it does manifest is short lived and the effect size is consistently less than most other types of active intervention.

    There is a convergence of data and concordance among data sets. You seem highly unwilling to accept that it all points to a conclusion since (I postulate) that would take away your ability to turn a blind eye to placebo use in clinical management. And there is indeed data from studies designed to test the placebo effect – or did you not read the article to which you are commenting?

    And that is actually what CAM requires us to be able to explain.

    It’s been explained. Not sure why you can’t see that.

    I would also prefer that this were not so, but that “science-based” tag places certain constraints upon what we can claim to know for certain.

    And here is that paralysis coming in again. Science has not demonstrated unequivocally and with absolutely zero margin of doubt that evolution is true. The creationists would say we need to see one species give birth to another, else they won’t be satisfied of the “truthfullness” of science in regards to evolutionary theory. But when data converge on a common answer, we find that type of argumentation vacuous. And tedious when it becomes repeated ad nauseum as you try and defend this nebulous right to sometimes, maybe, in some ways, kinda sorta, utilize placebo to manage patients but most certainly CAM is bad… but it does harness the placebo response well and we have to address that and maybe there is some clinical utility to it because, well, we just don’t know every detail yet.

    I would like to know, for example, whether acupuncture could reduce opiate requirements in patients that are open to that treatment. That would make me think again about about some matters.

    For someone so informed on this topic, you seem to forget things rather easily. Of course, acupuncture could (and I think would) reduce opiate requirements. So would simply telling the patient that the medicine is “really strong” and explaining how it will help the pain. So tying this back to the whole point of these posts on placebo – why would you choose acupuncture as your method of opiate reduction versus anything else that can and does harness that (small) subset of the placebo effect that engenders psychogenic neurophysiological changes? Sure you can claim that each is different and patients respond differently and some are more amenable to Reiki while some to acupuncture and still others to a coffee enema or other invasive pseudo-procedure. And if we left it there, I’d concede you were right and why not investigate these modalities to quantify the effect size of the psychogenic placebo effects and utilize them to improve patient management. But then that bugger of ethics comes into play. And lying to patients about the mechanisms involved is unethical. So out goes all those wonderful modalities that rely on the concept that the modality itself actually treats the pathology. And when talking about acupuncture the very real (albeit small) risks involved completely and utterly obviate the discussion. Already unethical, adding potential for risk (no matter how small) is simply untenable. In that regard I’m more comfortable with Reiki than acupuncture. But Reiki still doesn’t pass the smell test.

    The point is that part of the placebo potential of acupuncture will be conveyed by the perception that it is doing something invasive

    No, that is not part of the placebo potential of acupuncture. That is just part of the placebo response for everything. The more invasive something is the more profound expectancy is. That has nothing to do with acupuncture and once again misses the point. Perhaps we should do acupuncture with railroad spikes since that is even more invasive?

  31. Harriet Hall says:

    @pmoran,

    “the possibility that a charismatic “healer” will have more spectacular results with suitably primed patients than can be easily captured within prospective clinical studies.”

    Now you are really stretching.

    Yes, that’s possible. But so what? It is a truism that a charismatic practitioner is better at eliciting a placebo response. It is a truism that a tiny minority of patients might respond to any treatment in ways that are not captured by the usual clinical studies. But what does that mean for clinical practice? Should we prescribe the antibiotic that worked best in trials, or should we try the less effective one because it just might work uniquely well for an individual? Should we bring back bloodletting and Perkins’ tractors because they might get spectacular results in certain patients? Perkins’ tractors might reduce morphine requirements as much as acupuncture: then what? What about charlatans, obviously fraudulent quacks? Some of them have been very charismatic and have had spectacular results with suitably primed patients.

    We have to have some standards.

  32. ConspicuousCarl says:

    Harriet Hall says:
    The misconception I was referring to was the earlier view that the placebo effect was “all in the mind” with no neurophysiologic correlates at all.

    All of the results presented sound like important things to know, but the framing doesn’t fit into my head well.

    If I were to use the phrase “all in your head” (though I usually don’t), I wouldn’t mean to exclude some of functions presented here. I consider those things to be part of our neurology (or whatever the correct word would be). I thought it was well-accepted that many of the experimental results were the kind of things to be expected from psychological cues. Engaging in solution-oriented behavior, or just a familiar process, produces a dopamine reward. Anxiety enhances pain, therefore relieving anxiety will reduce pain. And I thought the whole purpose of anxiety was to produce stress responses, so we shouldn’t be surprised to see related chemistry come and go.

  33. ConspicuousCarl says:

    nybgrus said:
    My take on the placebo effect – at least as it has been described in the usual common and broad way – is that it is nothing more than study artifact and reporting bias/error. This is because I tend to view things like what Benedetti describes on a neurophysiological level as an active intervention.

    I think it is a matter of whether or not the mechanism or effect is part of the claim and intent.

    If you jab someone in a control group with a needle in the course of administering a placebo injection, and that act of stabbing with a needle produces a huge improvement via a mechanism which you did not expect (mechanisms at the level you/he describe) and were not testing as the central claim of the experiment, then it is a placebo effect.

    If you learn from further study that all people love to be stabbed with needles and get a huge dopamine rush from it, and you then administer stabbings to treat patients suffering from boredom, it is an active intervention.

    If you learn all of that but can’t avoid stabbing patients with needles to administer drugs during a trial, and you are not testing a claim related to the joy of being stabbed with a needle, then it is still a placebo effect in that specific case.

  34. ConspicuousCarl says:

    Um, in the previous post, I wanted “(mechanisms at the level you/he describe)” to mean including such things, not limited to such things.

  35. Wait a second…isn’t it frequently stated on this site by many of the bloggers that placebo effect IS subjective and has no measurable physiological mechanism? Or was that just specifically in reference to the asthma study where the placebo had no effect on lung function? Are you saying that placebo DOES have a measurable physiological impact when it comes to things like pain, but perhaps not with more complex processes like inflammation?

  36. err, not inflammation…chronic inflammatory disease…wish there was an edit button!

  37. nybgrus says:

    NYK: Both are true, which is why I have been endavouring to dissect out and clarify the issue.

    Most of what the placebo effect is is indeed subejctive and has no measurable obective physiological outcome. However there is a psychological component to the placebo effect – what I am calling the “psychogenic portion of the placebo effect” or “psychogenic placebo effect” to see how well that works – that is an active psychological intervention (like CBT or psychological counseling). Since the brain functions at the level of neurotransmitters and neural interactions, such psychogenic placebo effects can have that active effect on the brain and thus anything reasonably downstream from there. So in the case of the ulcer for example, it isn’t the placebo effect at all – it is the psychogenic placebo effect decreasing anxiety which decreases stress states which decreases cortisol and thus allows more prostaglandin synthesis to protect the gastic mucosa. But, from what I know, if you were already as relaxed as possible and the ulcer was from, say H. pylori, all the placebo effect, psychogenic or otherwise would not help since it could not ramp up prostaglandin synthesis – ony remove that inhibitory factor that actually is under direct neural control.

    Some diseases are more amenable to that psychogenic placebo effect, most are not. Even then, the effects are ephemeral.

    Sorry for the poorly written and incomplete response, but I have to get running. Hopefully that cleared up a little bit for you. I need to be at the morgue for an autopsy and they get mad if you are late. I’ll be listening to Benedetti’s podcast on the way over and come back to expound further when I return.

    Ciao for now!

  38. ConspicuousCarl says:

    nobodyyouknow on 27 Sep 2011 at 10:29 pm

    Wait a second…isn’t it frequently stated on this site by many of the bloggers that placebo effect IS subjective and has no measurable physiological mechanism?

    Not completely. I think the common position is that CAM treatments which work no better than placebos in trials tend to only be popular for subjective symptoms, which is not the same as saying that all of the effects are subjective.

    The key to that difference is that not all measurable changes in a patient’s biology are going to produce useful changes in whatever condition is supposedly being addressed. As I said above, the fact that changes in a person’s mental state can cause changes in chemical activity is not new.

    Since the usual topic here is insanely fake medicine with no valid scientific relationship to the diseases they are supposed to cure, it is usually the case that any physical changes which might occur from a treatment are not going to produce the desired physical results. The experiments mentioned above are good for proving the concept of psychological manipulation being able to cause physical changes, but they are not representative of the claims normally debunked in other postings on this website.

  39. nybgrus says:

    I think Carl said it quite well and I fully agree.

    I’ve now listened to the full podcast and found it interesting and quite edifying. Benedetti is basically saying everything I have been, and all the new tidbits I learned made perfect logical sense to me.

    The main difference is that he and I seem to have our nomenclature transposed. We both think that there is artifact, bias, and reporting error involved but he says that is not the placebo effect whereas that is exactly what I have been calling the placebo effect. What I call the active or “psychogenic” placebo effect is what he refers to as just “placebo effects” in toto. Just goes to show how easy it is for even trained professionals to get confused on such a complex and multifaceted issue. Honestly, I’m not sure how I feel about the nomenclature issue anymore except that I will now have to take an extra few seconds and be abundantly clear in my future discourses on the topic.

    I think that the definition Carl gives above is reasonable but fluid enough to sow confusion and difficulty. I have been trying to pin down some immutable terms so that frame of reference can be mitigated when describing the neurophysiological mechanisms that Benedetti talks about vs the artifact and noise I refer to as “placebo effect.” I think that whether you intend it or not, the mechanism is still the same and can be described as such. How potently the mechanism is activated would certainly hinge on intent. But my real goal here (besides getting a deeper understanding myself) has been to come up with a convenient way to discuss the topic such that CAMsters cannot lay claim to efficacy when they clearly do not deserve it.

    And to tack on a last little bit for NYK – what we are saying is that there are tangible and real effects from the psychogenic placebo effect which can affect any disease process…. as long as there is a link between the CNS and the disease process in some way. So a chronic inflammatory condition theoretically could be affected by said psychogenic placebo effect via decrease in cortisol levels from stress reduction. This just made me think of an interesting postulate (and if anyone knows any studies on it please send them my way!). I would hypothesize that in cases of chronic inflammation a psychogenic placebo would either have zero effect or a negative effect. Conversely a psychogenic nocebo would either have no effect or a positive one.

    I say this because the issue with chronic inflammation is overactivity of the immune system. Cortisol suppresses the immune system. If you use a psychogenic placebo to reduce anxiety and stress and thus reduce cortisol you would worsen inflammation be removing the damping effect of cortisol. A nocebo would increase stress and cortisol and thus dampen the immune response and improve symptoms (and the disease itself, actually). I say there is a chance that nothing will happen in either case because you may be either maximally anxiety free or stressed, repsectively, or the effect size may be so small as to make no clinical difference.

  40. Conspicuous Carl “If I were to use the phrase “all in your head” (though I usually don’t), I wouldn’t mean to exclude some of functions presented here.”

    Sorry to nitpick, but, I think you are using “all in you head” incorrectly. The emphasis of the phrase is on ALL and “head” is a substitution for “mind” not “brain”. It is most often used for or by a patient that reports symptoms for which no physiological cause can be found. The question is whether they are imagining or exaggerating their symptoms. If there is a physiological response, such as placebo induced dopamine release, that is a response that happens outside the mind(thought process) and therefore not ALL in the head.

    In HH’s usage, I believe the emphasis was on subjective reporting, (symptom is the same but the patient thinks it is different due to distraction, ect.) is an appropriate use because subjectivity is considered in the realm of the mind.

  41. nobodyyouknow “# nobodyyouknowon 27 Sep 2011 at 10:29 pm
    Wait a second…isn’t it frequently stated on this site by many of the bloggers that placebo effect IS subjective and has no measurable physiological mechanism? Or was that just specifically in reference to the asthma study where the placebo had no effect on lung function? Are you saying that placebo DOES have a measurable physiological impact when it comes to things like pain, but perhaps not with more complex processes like inflammation?”

    Benedetto specifically mentions that at least one of the placebo effects can have impact on immunology, immune systems. Sadly, cause I would have liked to hear it, they didn’t go into it. I’m guessing that topic would be in one of his books.

    I think the key is that different types of placebo stimulations do not create the same kinds of placebo response. See HH quote

    “There is not just one “placebo effect,” but many different placebo effects that work by different mechanisms, including (1) anxiety reduction, (2) activation of the reward mechanism (with dopamine release in the nucleus accumbens), and (3) learning.

    Placebo responses can be divided into two types: conscious and unconscious. Conscious responses involve suggestion and expectation. Unconscious responses occur with classical Pavlovian conditioning.” etc.

    Or listen to article. I’m curious what kind of stimulation and response he found for the immune system.

  42. The problem with using the term “placebo effect” to refer only to the biological response to reduced anxiety etc. is that this is not what is measured in the placebo arm of a clinical trial. So – the operational definition we live with every day is dominantly artifact and subjective reporting, with a generally small and highly variable neuro-endrocrine component that is significant mostly for pain and those symptoms that respond to anxiety and stress reduction.

  43. @nobodyyouknow, a little search on placebo and Benedetti resulted in this summary of immune system placebo effects

    “Placebo responses in both the immune and endocrine system can be evoked by pharmacological conditioning. In fact, after repeated administrations of a drug, if the drug is replaced with a placebo, immune or hormonal responses can be evoked that are similar to those obtained by the previously administered drug. This occurs with immunosuppressive placebo responses, that can be induced by repeated administrations of cyclosporine A (unconditioned stimulus) associated to a flavoured drink (conditioned stimulus). Likewise, if a placebo is given after repeated administrations of sumatriptan, a serotonin agonist of the 5-HT1B/1D receptors that stimulates growth hormone (GH) and inhibits cortisol secretion, a placebo GH increase and a placebo cortisol decrease can be found. These studies support a conditioning mechanism in both immunosuppressive and hormonal placebo responses.”

    http://www.scitopics.com/Placebo_and_Nocebo.html

    It should be noted that Benedetti is not suggesting that placebo be used this way clinically, his interest is focused on the using the placebo effect to tease out psychological/physiological effects.

    I will also note that placebo effect is limited in the sense that, for me, it don’t pay no bills. Gotta get to work.

  44. ceekay says:

    @nybgrus

    Stress negatively impacts many immune system processes including surgical wound healing (where I think your hypothesis would be most relevant). I think this is reasonably well established (See Sivamani 2009 PlosMed for details).

    And, not surprisingly, relaxation therapies appear to positively impact wound healing.

    But, interestingly, I did find one study in diabetic ulcers suggesting depression (which is associated with high levels of cortisol and stress) is associated with positive outcome.

    Very curious…..

  45. PharmScep says:

    Bendetti specifically stated that this whole discussion of the placebo effect is interesting from a brain research point of view – from the point of view of the neurophysiologist and other researchers. He also said that the clinical usefulness of this information covered is uncertain.

    It appears to me that while there is a few here and there “effects” from “placebo”…there simply is no strong data that come anything close to making me believe that placebo’s “effects” are clinically relevant.

  46. pmoran says:

    STeve: The problem with using the term “placebo effect” to refer only to the biological response to reduced anxiety etc. is that this is not what is measured in the placebo arm of a clinical trial. So – the operational definition we live with every day is dominantly artifact and subjective reporting, with a generally small and highly variable neuro-endrocrine component that is significant mostly for pain and those symptoms that respond to anxiety and stress reduction.

    I have been trying to avoid these distracting conceptual difficultiess by referring to the “psychogenic influences within medical interactions”. That is basically what we mean.

    Your statement above is a little ambiguous, as “generally small” does not mean the same as “significant”. Also, even oft-quoted Hrobjartsson publications allow that some studies comparing placebo programs to no treatment (not yet discussed here in detail) permit the existence of clinically worthwhile responses. .The results of such studies can be predicted to be very dependent upon the extent to which the researchers try to provoke or suppress placebo responses — usually the latter in mainstream studies.

    I allow that it remains difficult to distinguish placebo responses from biased reporting in these comparisons but they should reduce other artifact such as spontaneous improvements in symptoms.

    However, is the size of any effect the critical matter now that we know from the neurophysiology that there is almost certainly something “real” going on? Can small subjective benefits over a lot of people (or large ones for a few) be simply disregarded ? Mainstream medicine has not yet reached perfection, not in the effectiveness and safety of its remedies for all conditions nor in its ability to deliver them effectively to all and sundry so there is always a considerable amount of unmet medical need out there. .

    I personally find it difficult not to take this into account in my dealings with CAM users and its more honest practitioners and also within the politics and rhetoric of medical scepticism.

    We have considerable rights in choosing what we personally may offer our patients within the confines of our offices, but we have very limited rights over the bodies and minds of our patients once they step outside. .

    1. Harriet Hall says:

      @pmoran,

      If you want to accept some CAM treatments because of subjective benefits, to be logically consistent wouldn’t you have to accept the subjective benefits produced in the victims of every quack and snake oil salesmen? Some of them are very good at those placebo effect-generating interactions. Should we go back to using Perkins’ tractors? George Washington used them and got subjective placebo benefits.

  47. nybgrus says:

    @Dr. Novella – I agree fully. That is why I have been trying to leave “placebo effect” as the error and bias and call the “real” stuff “psychogenic placebo effects” (I agree with Benedetti that the plural is important).

    @michele: Thanks for that. I was wondering how he found the psychogenic placebo effects in immune response. That makes sense to me as well – Pavlovian conditioning with Hebbian learning. Though I’m still a little unclear as to how cyclosporin effects would work, but off the top of my head all I know is that it is a calcineurin inhibitor so I’ll have to look that up further when I get a chance.

    @ceekay: Yes, I am aware of those (well, the diabetic ulcer one not so much… but I can guess at a mechanism – DM ulcers are caused by peripheral vascular disease – the microvasculatre gets occluded. Increasing stress would increase catecholamines and could potentially get more peripheral blood flow going and thus improve wound healing). But I was referring to chronic inflammation, not wounds, which I am postulating would have the reverse effects as those we find in wound healing.

    @pmoran: You are arguing into the wind again.

    However, is the size of any effect the critical matter now that we know from the neurophysiology that there is almost certainly something “real” going on? Can small subjective benefits over a lot of people (or large ones for a few) be simply disregarded ?

    No. And we aren’t ignoring them. If we were, we wouldn’t be having these discussions and doing all these studies to find out exactly how the psychogenic placebo effects work and their limitations. We want to better educate ourselves as physicians to make use of every little edge we can get in patient care – hence my posts to Michele about how I have handled patients in the past (even before knowing all these details) and how I will continue to do so in the future. So once again you set up a straw man and…..

    I personally find it difficult not to take this into account in my dealings with CAM users and its more honest practitioners and also within the politics and rhetoric of medical scepticism.

    …then set it ablaze. More and more I am convinced you really are a CAM apologist hiding in physician’s clothing. Benedetti himself, Campbell in the interview, most people here, and every reputable study has all said the same thing – the psychogenic placebo effects are real, can be quite potent, but cannot be used in isolation in clinical practice. CAM is psychogenic placebo effects in isolation.

    Your argument is one of laziness. You are implying that we can’t harness the small and variabel effect size of the psychogenic placebo effects in our own medical practice and therefore should pass the buck over to the sCAMsters because they can do it oh so well.

    And if one straw man wasn’t enough, you go for a second just for good measure:

    We have considerable rights in choosing what we personally may offer our patients within the confines of our offices, but we have very limited rights over the bodies and minds of our patients once they step outside. .

    I’ve lost count of how many times it has been said to you (by me specifically and pretty much everyone else here including Dr. Hall) that we have no desire to forbid patients from employing whatever CAM they wish. We do have a desire to keep our profession from employing CAM by demonstrating how worthless it is. And in the same way that you would be OK with a consumer protection group policing automobile manufacturers by demonstrating safety and cost effectiveness (or lack thereof) we also strive to do the same for CAM in the general public.

    So your arguments are completely off base. And lazy and unethical. You are, in a nutshell, arguing that the prescription of a placebo in the form of CAM be utilized since you believe medical doctors cannot sufficiently employ the psychogenic placebo effects. The first is unethical. The second is lazy.

  48. daedalus2u says:

    I have been busy with other things, so I am joining the discussion late. I read the podcast transcript and found nothing to disagree with and nothing that is inconsistent with my interpretation of placebo effects.

    I think he is using the term “placebo effects” more broadly to include both positive and negative health outcomes and not just positive or health improving.

    What I especially liked was his demonstration that placebos can exert effects independent of expectations, that for some placebo effects expectation plays no role.

    In reading the section about the therapeutic ritual, I remembered reading a post by a parent of a premature infant who was in the NICU for months. Apparently when the various IVs and tubes and such needed to be changed, the first thing the staff would do was unroll multiple lengths of tape to have them ready to hold the various needles and tubes in place. Fifteen years later, the now teenager would have an anxiety attack whenever there was a crinkling that sounded like tape being taken off a roll. Eventually the mechanism was understood and deliberate desensitization was successful to some extent. I think I made the suggestion that for infants who are going to be in the NICU for a long time, that using a deliberate sound that is unique to that infant (a bell of a certain tone perhaps), would be beneficial in that the other infants wouldn’t have an anxiety attack in the NICU, and the unique sound could be used later to do desensitization therapy. It would probably be helpful, but there likely isn’t any way that research to demonstrate it is helpful could be done.

    I think that he is right that it goes all the way back to single cell avoidance of noxious stimuli. Of course, any prior warning of something noxious coming would condition organisms to respond to it ASAP.

    That is the whole point of having a nervous system and why various organs have neural control, to be able to respond via neurogenic signals from a central control agent instead of waiting for local control. Some things can’t be controlled only at a local level. Blood flow for example. The heart has limited pumping capacity. Blood flow is directed to different tissue compartments by dilating vessels leading to tissue compartments controlled to get more blood and by constricting vessels leading to tissue compartments controlled to get less blood.

    Blood flow is controlled by vasodilatation, which is regulated by nitric oxide. Neurogenic NO for dilatation and neurogenic superoxide for constriction is how central control is exerted. There are non-neurogenic signals via hormones put into the blood too.

    Any resource that the body has in limited supply (that would be all resources) pretty much has to be allocated centrally. The better that central allocation is, the more resources can be diverted to reproduction and away from survival. That was the whole point of the first 3+ billion years of evolution, so it is hard to change that now due to the idiosyncratic conscious desires of particular individuals.

  49. pmoran says:

    “However, is the size of any effect the critical matter now that we know from the neurophysiology that there is almost certainly something “real” going on? Can small subjective benefits over a lot of people (or large ones for a few) be simply disregarded ?”

    Nybgrus: “No. And we aren’t ignoring them. If we were, we wouldn’t be having these discussions and doing all these studies to find out exactly how the psychogenic placebo effects work and their limitations. We want to better educate ourselves as physicians to make use of every little edge we can get in patient care – hence my posts to Michele about how I have handled patients in the past (even before knowing all these details) and how I will continue to do so in the future. So once again you set up a straw man and…..

    Oh, dear! At the risk of starting a flame war — what utter humbug!!!

    You had already made up your mind that placebo effects don’t have to taken seriously in any context, and without looking deeply into the relevant evidence at that (since Benedetti’s work is apparently new to you and I am sure you have never looked at and thought deeply about Hrobjartsson-style studies).

    You were content to mouth prevailiing sceptical dogma that slotted into personal biases, and you still are. You cannot tolerate differing opinion to yours and you respond to it with bluster and ill-mannered and ill-judged ad hominem.

    May I quote what you said in your first response to this thread? —

    I guess I’m just stuck on the traditional definition of placebo being inactive, and thus any “effect” from it must be artifact and bias/error. Including what we can now reasonably deem as active interventions under the placebo is not something I am comfortable with – but I am open to insight.”

    I suggest that your mind is as open to insight as is mud and as muddled, since in the above response you wanted it both ways – maintaining that the placebo does nothing but now that we know it probably does “do” something we will call it something else — an “active intervention” in your selection of words. How the hell does that clarify anything?

    A simpler and more obvious viewpoint is to accept that the placebo “does” nothing in itself except to the extent that certain characteristics of it can add to all other aspects of the therapeutic evnironment in eliciting non-specific patient responses.

  50. nybgrus says:

    I have never, ever said that placebo effects were not to be taken seriously in all contexts. I said they were not to be taken seriously in the context of CAM insofar as CAM is only placebo effects and nothing more. If you want to go back and look to some of my very earliest posts ever on this forum, you will find that I consistently agree with the import of the placebo effects in the context of clinical care using real medicine. I have never changed that stance.

    I have challenged your continued assertions that it is OK to prescribe or otherwise recommend CAMs to patients knowing they are nothing more than placebo in order to harness those “non-specific effects” that you yourself can’t seem to employ in your own patient care.

    Part of that is the fact that the majority of placebo effects is indeed reporting error and bias – thus when looking at clinical trials and determining that, say, acupuncture had no better result than placebo but that it was indeed significantly better than no treatment at all or even standard of care, you cannot claim there was anything actually done since most of that will be the reporting error and bias. Everything Benedetti has said confirms exactly that – and it has been what I have been saying all along.

    I have also always said that there is indeed a real effect from placebo. Which makes no sense, until one realizes that the effect is from the active intervention of the therapeutic interaction. Benedetti’s work and what he cited further confirms this and even elucidates the precise neurophysiological mechanisms for it. Some of it I already knew, some of it I assumed, and some of it was new. But none of it is contrary to anything I have been saying. So yes, my mind is open. And I have refined my thinking, but found no reason to fundamentally change my stance. Certainly not to align with yours and believe that recommending CAM is at all reasonable.

    And yes, I tossed a little ad hominem your way. But that is only a logical fallacy when it is not relevant to the argument. And my remarks were indeed very relevant. I’ve acknowledged all along the real effects of placebo – but I’ve said that they are clinically untenable to employ in the manner which you advocate. Prescribing it outright as a recommendation to CAM is unethical. And your argument that we cannot harness those effects as well as CAMsters do is lazy. So don’t get all in a huff because I have called you out and you have no defense against it.

    maintaining that the placebo does nothing but now that we know it probably does “do” something we will call it something else — an “active intervention” in your selection of words. How the hell does that clarify anything?

    It clarifies things very nicely. For example, it reconciles the notion that by definition a placebo is completely inert yet we find real physiological changes with placebo. In fact, Benedetti himself notes that there is indeed these two very different phenomena and that delineation is very important. If you’d listened to the podcast you should have noted how much time they devoted to the definition. My only sticking point is exactly what nomenclature to apply and when – the precise terms I’ve used don’t quite jive with some of what Benedetti says, but the concepts I describe do. If you can’t follow the discussion, then I am sorry – it is complex and I’ve made it as clear as I can.

    A simpler and more obvious viewpoint is to accept that the placebo “does” nothing in itself except to the extent that certain characteristics of it can add to all other aspects of the therapeutic evnironment in eliciting non-specific patient responses.

    Simpler? Yes. Better? No. The sCAMsters love making things simple and dumbing them down so as to obfuscate the fact that their modalities don’t actually work, or if they do only through the very limited and highly variable psychogenic placebo effects. And on top of that, we know better now and don’t have to say it so vaguely.

    You may suggest that my mind is not open to insight. I’d suggest that yours isn’t adept enough to follow such complex nuances. And for whatever reason, you seem content that as a physician I would not be able to employ these non-specific effects you so dearly love since you claim that only the sCAMsters can really do it well. So to that I say “what utter humbug!” Don’t project your limitations onto me. Try and discuss the actual science instead of making these nebulous claims that there “must be something” to acupuncture and CAM placebos that we can’t harness and settle for simplistic explanations of placebo that don’t let us suss out the differences.

  51. pmoran says:

    @pmoran,

    If you want to accept some CAM treatments because of subjective benefits, to be logically consistent wouldn’t you have to accept the subjective benefits produced in the victims of every quack and snake oil salesmen? Some of them are very good at those placebo effect-generating interactions. Should we go back to using Perkins’ tractors? George Washington used them and got subjective placebo benefits.

    Harriet, I have already expressed the opinion that the mainstream should not use placebos/CAM and “accept” it in any systematic way. Why keep harking back to that? (Actually a great many doctors admit to using them, but we might need to look at the individual contexts before knowing how condemnatory to be of that. I know of some heinous examples with cancer and of course will violently oppose that.)

    There are two separate issues here. What does the science say about the strength of placebo potential?

    I suggest, after an extremely detailed look at the evidence (much of which I have posted here without there being any rebuttal) that that question is not yet finally resolved, apart from there being, in general, no effect on underlying disease processes.

    The available data is otherwise consistent with a range of potencies of placebo depending upon context, as you might expect from something so dependent upon patient states of mind and psychological inputs. It is not unfeasible that in favorable settings it can extend into the range of efficacy of commonly used drugs.

    The second issue is: how should our understanding of the placebo affect our policies and attitudes towards CAM?

    There are obviously going to be considerable differences of opinion on that and I make no apology whatsoever for having certain leanings that could only be resolved by clear evidence that the public derives no benefit at all from CAM.

    For this reason I am bound to react when “shruggies” and “quackademics” are portrayed as somehow betraying science. I ask “which science are they betraying?” if the sole objective of scientific medicine is patient welfare.

    There are many different elements feeding into my leanings including historical and evolutionary perspectives.

    To answer your question (above) what you say is brodaly true. As I have said, safety and patient welfare should dominate our response to CAM. The pseudoscience is about evenly spread and is rarely as vulnerable as we think.

    We actually have enough information about most forms of CAM to be able to offer cost/risk/likely non-specific benefit judgements similar to those that guide mainstream treatment selection if patients ask our opinion. I would happily state in any information provided that I think “X” probably “works” as placebo, but would prefer a less loaded term.

    But I admit there is yet much to be thought through.

    1. Harriet Hall says:

      @pmoran,
      “I have already expressed the opinion that the mainstream should not use placebos/CAM and “accept” it in any systematic way. Why keep harking back to that?”

      Because even though you don’t argue for systematic acceptance, you keep suggesting that there might really be something valuable there, that prescribing a placebo can sometimes actually benefit patients. It seems to me you want to leave a loophole for a doctor to prescribe unproven treatments. If placebos offer real benefits in clinical practice, it seems to me it would logically follow that Perkins’ tractors would be as acceptable as acupuncture, and that overt quackery might be just as justified as any unproven or insufficiently tested treatment.

  52. pmoran says:

    It clarifies things very nicely. For example, it reconciles the notion that by definition a placebo is completely inert yet we find real physiological changes with placebo.

    Well for Pete’s sake don’t confuse things by calling it an “active intervention”.

    Part of that is the fact that the majority of placebo effects is indeed reporting error and bias – thus when looking at clinical trials and determining that, say, acupuncture had no better result than placebo but that it was indeed significantly better than no treatment at all or even standard of care, you cannot claim there was anything actually done since most of that will be the reporting error and bias

    Well sham acupuncture has in some hands produced effect sizes well into the moderate range. Explain how you know that is “mostly” due to reporting error and bias, especially if you also expect neurophysiological responses from sham acupuncture in its capacity as an “active intervention”?

    Look, I actually do understand what you mean. I (metaphorically) helped write the sceptical book on this stuff.

    It is very difficult to distinguish reporting bias from “true” placebo responses in any kind of clinical study and we have no certainty yet as to what the neurophysiology means, and possibly even whether such studies will stand the test of time and replication.

    Nevertheless there remains substantial plausibility to placebo influences. It would be astonishing if they were not “real” to some significant extent and in some settings.

    We are thus not entitled to allow certain biases to fill in the gaps in our knowledge. I think we skeptics tend to do that. I know I have.

  53. nybgrus says:

    Well for Pete’s sake don’t confuse things by calling it an “active intervention”.

    How does that confuse things? It is an active intervention. I am delineating and sussing out the inert substance from the active psychological treatment associated with it. That is why my argument from the beginning has been to call “placebo effect” the reporting bias and other stuff that isn’t “real” and separate that clearly from the actual part of the therapeutic ritual that effects tangible change – what I have started calling the psychogenic placebo effects. I’ve made the analogy to CBT since that is an active psychological intervention – and I think it is apt.

    Explain how you know that is “mostly” due to reporting error and bias, especially if you also expect neurophysiological responses from sham acupuncture in its capacity as an “active intervention”?

    Because the data demonstrates it to be so. Comparing studies on placebo with studies using placebo you find that to be evident. But moreso, my argument was that since we know the reporting error and bias to be prevalent and factoring in things like the Hawthorne effect, when looking at RCTs where some intervention works “no better than placebo” the only reasonable conclusion is that it doesn’t work.

    Nevertheless there remains substantial plausibility to placebo influences. It would be astonishing if they were not “real” to some significant extent and in some settings.

    You consistently lose me. Of course there are “real” and significant effects. I’ve said that all along. The entire point of this post is to demonstrate that. It is your recommendation for how that may be applied that I find fault with.

    And I am not the only one. You keep coming back and saying “Harriet, I have already expressed the opinion that the mainstream should not use placebos/CAM and “accept” it in any systematic way. Why keep harking back to that?” – perhaps, Peter, because that is what you keep saying in your arguments. I consistently see it. Dr. Hall sees it. Dr. Gorski sees it. And we have all commented on it numerous times. Your stance is very unclear and wishy washy. Every time one of us tries to pin down what you are saying, you claim that isn’t it. Perhaps instead of questioning why we are so consistently questioning you, you should exam your rhetoric and make sure it is conveying what you actually mean to say. Not that this is a popularity contest, but I have received consistent feedback from the commentariat here and at neurologica that my stances and explanation are usually quite clear, firm, and explicit. The same simply cannot be said for you.

    We are thus not entitled to allow certain biases to fill in the gaps in our knowledge. I think we skeptics tend to do that. I know I have.

    Honestly, I am not keen to attack you personally, but it seems to me and others here that it is you that has some biases being interjected.

    You said to Dr. Hall:

    …for having certain leanings that could only be resolved by clear evidence that the public derives no benefit at all from CAM.

    Except that is not how to prove a stance. You need to prove that the public does derive benefit from CAM in order to condone its existence. The onus is not on us to prove it has no worth at all and otherwise tolerate it. Furthermore, we have demonstrated clearly that CAM is, for all intents and purposes, placebo wrapped up in a ritual. So any utility from CAM is either clinically irrelevant or unethical. You say:

    It is not unfeasible that in favorable settings it [placebo effects] can extend into the range of efficacy of commonly used drugs.

    I fully agree and I think everyone here would as well (at least to some degree). I concede that I did not think that very likely prior to my recent foray into Benedetti’s work, but I will absolutely grant you that data demonstrates exactly what you claim in that statement.

    But that does not make a feasible argument for the existence of CAM. As I said above, whatever placebo effects may be harnessed by CAM will either have negligible effect size or be part of that narrow subset which does have significant and real physiological effects. Obviously for the latter we can easily and comfortably eschew it’s use and find no justification. But you claim, “what about the real effects of size comparable to drugs?” The problem there is ethical – not one of efficacy.

    You continue to stick to the finest points of “what if” when everything demonstrates that those are academic questions – we needn’t have them answered to have a firm stance on CAM.

    Pure placebo is unethical to use in medical practice – even if it would make your patient better. And so no matter how you slice the effects up, CAM is pure placebo and unethical to tolerate its existence.

  54. PharmScep says:

    “Everything Benedetti has said confirms exactly that – and it has been what I have been saying all along.”

    Benedetti stated that his research and other placebo research has no known use clinically at this point and it may not. He also said that at this point it is primarily a means to study the brain.

    For anyone to stretch this fact to utility in the clinical arena would be a grand stretch of the imagination, which is what you seem to be attempting.

  55. pmoran says:

    Explain how you know that is “mostly” due to reporting error and bias, especially if you also expect neurophysiological responses from sham acupuncture in its capacity as an “active intervention”?

    Because the data demonstrates it to be so. Comparing studies on placebo with studies using placebo you find that to be evident.

    Again, what data?

  56. nybgrus says:

    @pharmscep:

    He (and Campbell) note that the mechanisms of psychogenic placebo effects as he outlines them are not useful clinically in and of themselves – i.e. pure placebo has no place clinically. That is why I am edified in my stance on CAM. However, they both also note that knowledge of the mechanisms can and should be used in conjunction with actual medicine in a clinical setting – that is what I am advocating.

    Note that Campbell discusses specifically the utility of telling patients what medicine they will be getting, how it works, and what to expect. For certain patients using adjectives like “powerful pain reliever” which is true indeed, can further heighten the effects. Furthermore, knowledge of these mechanisms can be used to avoid counter-transference in health care professionals – as Campbell points out again sometimes patients feel better immediately and since the HCW knows the medicine can’t have acted that fast believe that the patient was lying the whole time. Understanding these mechanisms will do good service and dispel such notions.

    So no, I do not think I am stretching any facts here at all, and as I said – my overall stance on the topic hasn’t really changed; just my understanding has deepened.

  57. nybgrus says:

    @pmoran: The gestalt of it Peter. Once again, you seem paralyzed to draw inference on convergence unless there is a single specific paper or meta-analysis to your liking. As Dr. Hall pointed out, you seem to want to keep your loop-hole for woo-of-choice until someone writes a peer-reviewed paper titled “Meta Analysis Demonstrates Why All CAM is Not Useful At All: a Paper for Peter Moran.”

  58. PharmScep says:

    @nybgrus…I was speaking more in the direction of pmoran…I realize the points you are making and agree…when I say “placebo” is not clinically useful at this point – what I mean is that literally giving someone a sugar pill, an herbal, accupuncture or sham acupuncture for actually …I see no clinical utility to such things…when I was talking about stretching, I was talking about the apparent “loophole” you are noticine in pmoran’s comments….

    “Meta Analysis Demonstrates Why All CAM is Not Useful At All: a Paper for Peter Moran.” lol, sorry pmoran, that is kinda funny ;)

  59. PharmScep says:

    correction *actually…I meant to say for an actual disease in an actual treatment…I see no clinical evidence to suggest that in any way I care to use a sugar pill or something of that sort for treating patients…but the knowledge that benedetti spoke about definately is interesting to make sure we all speak to patients correctly…..

  60. pmoran says:

    Pharmscep, actually the notion of the placebo preceded the evolution of the modern clinical study. Healers throughout history commented on the ability of the human mind to both cause illness and affect it. Doctors noted that a saline injection sometimes seemed to work as well as a normal dose of morphine. This is how it all began.

    One reference to the use of the word goes back to 1811. Placebo-controlled trials were rare before the middle of the last century.

    Note also that the objection of most of the mainstream to the use of placebo is not based on it not “working”. Whether it “works” has been on the back burner and subject to non-evidence-based convenient opinion for a long time because it was held, with some justification, that it is unethical and counterproductive for doctors to knowingly employ placebos.

    CAM, however, poses different questions. The ethical considerations don’t apply to most conditions of its use, and it is mainly being employed when there are deficiencies in mainstream care. We can also now predict with some confidence that users derive small and possibly sometimes large benefits from it,and that it helps with deep human needs including an inclination towards compusive consumption of medicines when ailing.

    So I ask “what approaches to it will maximize any benefits while also reducing the harm?”

    I am not sure yet that I have any answers but feel I am posing a legitimate question based upon a more clear-sighted than average understanding of what we are up against it.

    Benedetti is being cautious, avoiding controversy in that interview. You should also read this paper he has also co-written ( with Ted Kaptchuk, one of SBM’s “bete noirs” on placebo matters. )
    http://www.bioethics.nih.gov/departmentpubs/Miller%202010%20-%20Lancet.pdf

  61. Harriet Hall says:

    @pmoran,

    “The ethical considerations don’t apply to most conditions of its use, and it is mainly being employed when there are deficiencies in mainstream care.” I think that has been shown not to be true, by studies previously cited here.

    “We can also now predict with some confidence that users derive small and possibly sometimes large benefits from it,and that it helps with deep human needs including an inclination towards compusive consumption of medicines when ailing.”

    No we can’t, at least not in a clinical setting without prior conditioning. And if we could, it would apply equally to users of fraudulent quack products. Again, what about Perkins’ tractors? If patients derived these benefits from them, would you use them?

  62. pmoran says:

    @pmoran: The gestalt of it Peter. Once again, you seem paralyzed to draw inference on convergence unless there is a single specific paper or meta-analysis to your liking.

    Again — bluster and ad hominem, and no answer..

    I at least expected some clarification of what you meant by this presumed “mis-speak”:

    “Comparing studies on placebo with studies using placebo you find that to be evident .

    How is anyone to interpret that? Have you been out celebrating something?

    Would it help if I rephrased the question? — “what kind of evidence allows you to attribute quite substantial effect sizes from sham interventions in some studies “mostly” to reporting error and bias, (especially now we have evidence of potentially helpful neurophysiological responses to placebo)?”

    One reason for being careful here is that the blithe assumption that patient’s reported benefits are not “real” is one way of immediately putting such claimants off-side, when what we should be about here is doing everything we can to regain the trust that we have lost for various reasons. It is doubly tragic if we are also making judgments that are contrary to our own science.

  63. pmoran says:

    Sorry about the formatting — I am sure you can work out who said what.

  64. pmoran says:

    “The ethical considerations don’t apply to most conditions of its use, and it is mainly being employed when there are deficiencies in mainstream care.”

    I think that has been shown not to be true, by studies previously cited here.

    Studies show that by far the majority of use of CAM is complementary to mainstream care, which implies that problems are at least partially persistent.

    It also stands to reason that if the mainstream had an effective and safe, permanent cure for any condition CAM options would die out completely, if only from a declining market.

    “We can also now predict with some confidence that users derive small and possibly sometimes large benefits from it,and that it helps with deep human needs including an inclination towards compulsive consumption of medicines when ailing.”

    No we can’t, at least not in a clinical setting without prior conditioning.

    I think the mere act of having taking action about a problem can allow the mind to focus elsewhere than on the symptoms and induce an “it’s not so bad now” response. The conditioning aspect is only part of the placebo, and we get plenty of that from our mothers. There is also likley to be an in-built survival instinct to respond to placebo.

    And if we could, it would apply equally to users of fraudulent quack products. Again, what about Perkins’ tractors? If patients derived these benefits from them, would you use them?

    Our scientific minds have a low tolerance of paradox, but the quirks of the human mind do permit the likelihood that treatments offered with fraudulent intent will help some people. To which I also say: Rats! Damn! (etc.)

    Nevertheless the only likely scenario in which a doctor would confront Perkin’s Tractors is that a patient came along saying they had been recommended to her as producing dramatic relief of their condition, and should she try them?

    Let me turn this around onto you. This patient has a non-life-threatening but troublesome condition. She is already on maximum medical treatment, or any new evidence-based treatment that might be tried carries the risk of serious side effects.

    The patient is a simple soul who will probably go and try the treatment at the urging of her friend whatever we say, but is merely seeking our aquiescence. The treatment is cheap and safe. A soupcon of placebo might well help her over a current bad patch.

    What would you do?

  65. Harriet Hall says:

    @pmoran,

    ” if the mainstream had an effective and safe, permanent cure for any condition CAM options would die out completely”

    I disagree. I can’t imagine that CAM would ever die out. Belief systems would still make some patients reject the mainstream treatment. Just one example from my local area: a chiropractor rejected antibiotics for his son’s meningitis and treated him only with manipulation. (He died.)

    ” the mere act of having taking action about a problem can allow the mind to focus elsewhere than on the symptoms”

    I absolutely agree; but I don’t think that justifies prescribing placebos.

    “What would you do?”

    I thought I had already made clear what I would do. I would tell her there was no scientific evidence that it worked but I would acknowledge that some people thought it had helped them and I would have no objection to her trying it and I would ask her to report back and let me continue to manage her care. What I would not do is suggest it or recommend it in the first place, which I assume you would not either. From what you have said previously, I got the impression that you would suggest things like acupuncture, whereas I would not.

  66. nybgrus says:

    @PharmScep:

    It seemed to me that the comment was directed at me. Sorry for the miscommunication.

  67. nybgrus says:

    Again — bluster and ad hominem, and no answer..

    No bluster or ad hominem. Just frustration. You are asking me to answer a question which is something that the body of blogging by Gorski and Novella is attempting to cover. You want me to cite some specific piece of evidence or a meta analysis that neatly answers your question. That is disingenuous and you should know that. Hence my comment about the paper that should be written for you.

    It also stands to reason that if the mainstream had an effective and safe, permanent cure for any condition CAM options would die out completely, if only from a declining market.

    Dr. Hall covered it already but I think it is a vastly salient point to hammer in on. I believe that through proper education and a firm stance from our profession, that CAM use will decline – but never go away. It is fundamentally rooted in ideology. The parallels between CAM and religion are so many it truly astounded me as I learned more and more. So there will always be the true believers and there just isn’t much we can do about that. But then there are the gullible, the credulous, and especially the lay person who just doesn’t know better that are being mislead and duped by the sCAMsters. So trying to assert that CAM will go away once medicine has the answer is completely wrong. Dr. Hall’s example was quite apt – antibiotics are that solution you say would make CAM go away. Sure didn’t in that case.

    Whether it “works” has been on the back burner and subject to non-evidence-based convenient opinion for a long time because it was held, with some justification, that it is unethical and counterproductive for doctors to knowingly employ placebos.

    With some justification? I’m at a loss here. Either you really have a poor grasp of medical ethics or you are trying to bend things to fit into your loophole you keep wishing to push forth.

    The ethical considerations don’t apply to most conditions of its use, and it is mainly being employed when there are deficiencies in mainstream care.

    They don’t apply to CAM? Do you even realize the ethical ramifications of what you have just said? By logical extension I could say that since bride burning is not subject to the same ethical concerns in India as in Australia, I could go ahead and utilize that convenient rite to off my spouse. Maybe CAM doesn’t keep a good ethical house, but to use that as justification for your loophole to look the other way is egregious. And we’ve already touched upon the whole “CAM only exists where medicine fails argument.” I honestly can’t believe you’ve read this blog and been a physician and can still hold that view with how many times that has been demonstrated to be false.

    Benedetti is being cautious, avoiding controversy in that interview. You should also read this paper he has also co-written

    Benedetti is staying within the science and factually stating that placebo in isolation has no place in clinical practice and that it can only be used to enhance and augment already evidence based treatments. The paper says exactly the same thing, even though I can tell when Kaptchuk had a hand in it because he overstepped a few times what I would consider reasonable. The ultimate conclusions and discussion of the ethical principles is exactly in line with what all of us here have been saying, however. Perhaps you should read the paper again – especially the “Ethical principles of enhancing placebo effects in clinical care” section. I’ll point out that the little spiel in there about how to tell a patient with low back pain to use acupuncture was most likely written by Kaptchuk and mirrors a lot of what you have said – and is still wrong. That is inentionally misleading. We are supposed to be trained to a higher standard and therefore must hold ourselves to that higher standard. I am not stupid enough to think a patient will reasonably interpret me saying that as “Acupuncture is BS and only works through placebo” – the authority I would wield as a physician would give the imprimatur of legitimacy and that sort of intentional misleading is, IMO, unethical. It is very different to say “Hydromorphone is a very powerful pain killer and will work really fast” because those are true statements, even if you wish to quibble about the strength of my qualifiers and adjectives.

    “what kind of evidence allows you to attribute quite substantial effect sizes from sham interventions in some studies “mostly” to reporting error and bias, (especially now we have evidence of potentially helpful neurophysiological responses to placebo)?”

    As I said, to answer this question would be to go over a large swath of blogging that Gorski, Novella, Hall and others have done. The kind of evidence is understanding of things like the Hawthorne effect, that regression to the mean, natural fluctuation, reporting bias, observer error, study design flaws, and researcher bias all will always factor into a study. Then take into account the generally small and transient effect sizes we see in RCT placebo, combine the known mechanisms as outlined by Benedetti, and you should be able to see that in the vast majority of cases the majority of placebo effect is indeed simply bias and error. The utility of Benedetti’s (and other’s) work in placebo mechanisms is to demonstrate to us that even in those cases, the results are still highly variable, usually not of large effect size, and most importantly limited to very specific types of pathologies. So knowing all of this, and nowing that error and bias will always enter into the calculus of an RCT, when you find that acupuncture works for back pain the same as placebo, you can be reasonably certain that at least a significant portion of that has nothing to do with the psychogenic placebo effects and thus whatever effect size you do see must be larger than the true effect size. We can quibble over whether that constitues over 50% or not, but that doesn’t matter because you still haven’t resolved the ethical issues inherent to the conversation.

    One reason for being careful here is that the blithe assumption that patient’s reported benefits are not “real” is one way of immediately putting such claimants off-side, when what we should be about here is doing everything we can to regain the trust that we have lost for various reasons.

    As I’ll demonstrate in just a bit, you seemlessly flow from study results, to population level effects, to individual patient level interactions and the questions, answers, and responses at each level are different. Keep things straight Peter.

    It is doubly tragic if we are also making judgments that are contrary to our own science.

    And it is infinitely more tragic if we also knowingly ignore a high ethical standard in our pursuit of making the patient feel better at all costs. I won’t burden the conversation with the tales you know all to well before medical science felt that need.

    I think the mere act of having taking action about a problem can allow the mind to focus elsewhere than on the symptoms and induce an “it’s not so bad now” response.

    Lets stick the science, until you need to make a statement that feels good to say and has no bearing on the conversation. I can be kind and distract my patients from their symptoms just fine. I don’t need to refer them to some CAM quack to do that for me, thank you very much.

    Our scientific minds have a low tolerance of paradox, but the quirks of the human mind do permit the likelihood that treatments offered with fraudulent intent will help some people

    Which is why I am employing the term “psychogenic placebo effects” because that dispels the paradox and makes everything very clear.

    Let me turn this around onto you.

    There is no need. You have done this tack before a few times and each time it has been asked and answered quite handily by Dr. Hall and myself as well. And this is that example I spoke of above where you transition the argument between study, population, and individual as if it were all the same argument. It is not. On the flipside, my partner’s friend who is a very “holistic” person in that semi pejorative sense we might ascribe to it truly believed in acupuncture. She asked me my thoughts on it. I gulped hard, since I actually quite liked her and so did my partner, so upsetting them was not something I was interested in. She saw my trepidation and asked me to continue. After about an hour of talking, and explaining things as I am wont to do here, she was impressed and thanked me. And has since never used acupuncture. Would I have an hour with a patient? Not in one go. But over the course of my therapeutic relationship? Certainly. And I am willing to work at that to do my best.

  68. pmoran says:

    Nybgrus As I said, to answer this question would be to go over a large swath of blogging that Gorski, Novella, Hall and others have done. The kind of evidence is understanding of things like the Hawthorne effect, that regression to the mean, natural fluctuation, reporting bias, observer error, study design flaws, and researcher bias all will always factor into a study. Then take into account the generally small and transient effect sizes we see in RCT placebo, combine the known mechanisms as outlined by Benedetti, and you should be able to see that in the vast majority of cases the majority of placebo effect is indeed simply bias and error. The utility of Benedetti’s (and other’s) work in placebo mechanisms is to demonstrate to us that even in those cases, the results are still highly variable, usually not of large effect size, and most importantly limited to very specific types of pathologies. So knowing all of this, and nowing that error and bias will always enter into the calculus of an RCT, when you find that acupuncture works for back pain the same as placebo, you can be reasonably certain that at least a significant portion of that has nothing to do with the psychogenic placebo effects and thus whatever effect size you do see must be larger than the true effect size. We can quibble over whether that constitues over 50% or not, but that doesn’t matter because you still haven’t resolved the ethical issues inherent to the conversation. etc —

    That tells me nothing I didn’t know and have not echoed myself on occasions.

    Nevertheless, much of the above will be controlled for in head-to-head comparisons that show substantial effect sizes for various sham treatments over time when compared to no treatment groups or patients on a waiting list for treatment — not “small and transient”, and even detectable within the rather bland psychological environment of the typical clinical trial .

    Those “responses” must be mainly due to either placebo responses or reporting biases.

    I am asking you, or anyone else, how you know how much is which? Remember in the NEJM study 80% of the effect of a bronchodilator on the symptoms of asthma could be reproduced by placebo. There are many examples allowing for substantial “effects” from that combination of phenomena.

    It is clear that you, like most sceptics and myself at one time have a strong wish to diminish the significance of placebo influences within medicine, so as to leave CAM with no place to hide. That requires you to have an answer for my question, by which I mean direct evidence.

    This is, after all a science-based forum which means that there are few limits to the scientific rigor that can be demanded of you.

    It is not enough to quote the opinion of others or rummage around for explanations which suit that purpose.

  69. pmoran says:

    Harriet:” if the mainstream had an effective and safe, permanent cure for any condition CAM options would die out completely”

    I disagree. I can’t imagine that CAM would ever die out. Belief systems would still make some patients reject the mainstream treatment. Just one example from my local area: a chiropractor rejected antibiotics for his son’s meningitis and treated him only with manipulation. (He died.)

    You are right, of course. There will always be nutters.

    Nevertheless most CAM users just want to get better and they will get to hear of excellent results from the mainstream just as readily as they respond to the testimonials of CAM.

    I got the impression that you would suggest things like acupuncture, whereas I would not.

    If acupuncture does not also test your resolve it should, for two reasons.

    Firstly, even its sham versions seem able to produce substantial effect sizes, and possibly worthwhile results even in difficult conditions such as chronic back pain at least in some populations.

    This we might expect, since it incorporates many of the features that we would expect of an ideal placebo — elaborate ritual, invasiveness, regular hands-on contact with the therapist, perhaps a little mystery.

    The ethical arguments against the placebo have been formulated with the assumption that any benefits from them can be readily forgone, but what if they can help when nothing much else will, or if they can perform as well as safely as some drugs, as some preliminary studies suggest?. .

    Secondly,acupuncture is an “impure” placebo, in that it does some things that might be expected to have useful physiological effects, the enforced periods of relaxation, distractant and counterirritant effects from the needling? — also the expectation of needling may also be a more potent trigger of endorphin release than ordinary placebos.

    Those effects may well be temporary but they could facilitate a resetting of pain tolerance levels.

    1. Harriet Hall says:

      @pmoran,

      “they will get to hear of excellent results from the mainstream just as readily as they respond to the testimonials of CAM”
      But they won’t understand the difference.

      “The ethical arguments against the placebo have been formulated with the assumption that any benefits from them can be readily forgone”
      No, the assumption is that the ends don’t justify the means if you have to lie to patients, and that the benefits can be obtained in other ways without lying.

      ” the expectation of needling may also be a more potent trigger ”
      Yes, it probably is. That doesn’t mean a more potent placebo trigger is ethical. As weing pointed out, sham surgery is an even more potent trigger.

      “Those effects may well be temporary but they could facilitate a resetting of pain tolerance levels”
      This is nothing but speculation.

      I don’t understand why you keep grasping at straws to justify recommending acupuncture. I can’t help but wonder whether something happened in your life or in your clinical experience that created a bias.

  70. nybgrus says:

    You admonish me to stay within the science and you reference Benedetti and you accuse me of essentially cherry picking and just so stories.

    Then you go ahead and ignore one of the most stressed and salient points that Benedetti made – you cannot evaluate the effect size of a placebo arm from an RCT since it is not designed to actually do that. I’m not seeking to minimize it any more than Benedetti himself states that they are minimal.

    Remember in the NEJM study 80% of the effect of a bronchodilator on the symptoms of asthma could be reproduced by placebo.

    Oh I remember quite well, and there were 30 comments afterwards demonstrating each and every aspect of that study which gives very good and plausible reasoning as to why the majority of that effect would be anything but real benefit. You seem to have forgotten that.

    You are arguing from ignorance and hedging your bets in a way that I and others here disagree with. You also play fast and loose with ethics.

    The ethical arguments against the placebo have been formulated with the assumption that any benefits from them can be readily forgone, but what if they can help when nothing much else will, or if they can perform as well as safely as some drugs, as some preliminary studies suggest?

    The ethical arguments state that they must be foregone. As I’ve illustrated above, you can’t even accurately determine what the actual effect size is from an RCT. So you are arguing for the bending or outright breaking of ethical standards in order to hope that the effect size of a placebo arm from an RCT is comprised in majority by real benefit. Yet you have absolutely no basis to assert that, and there are myriad reasons why that assertion can be considered almost certainly false.

    So you haven’t even established that they can effect real benefit, but even if they can they still cannot be employed ethically. I could also suggest that forcibly removing someone’s kidney for a transplant would be beneficial to a large group of people that would otherwise have no science based recourse for treatment. We could even set up a government system in which the majority of the populace voted to enact such a law. It would still not be ethical.

    Secondly,acupuncture is an “impure” placebo

    What does that even mean?

    The counterirritant effects are BS. Have you ever had acupuncture? I have. The needles are so thin you don’t feel them going in. And as Novella pointed out in a study, you would have to vibrate them at 50hz to achieve any sort of physiological action.

    Enforced relaxation? Once again, you have just described just about every CAM out there – and at a practical level you have not described all of acupuncture, just the model you would likely employ if you were to use it.

    Distractant? Once again, a nurse, physio, volunteer, or even a physician can be just as fine (and according to Benedetti even better) distractant. I used to volunteer in a peds ward when I was in high school – all I did was talk to kids who were in hospital and they felt much better. I see no reason to send them to an acupuncturist who believes he is moving your qi around imaginary meridians to achieve what a normal human conversation can.

    The expectation of needling may be a more potent trigger? For someone who continually admonishes me to stay within the science you sure like to make a lot of assumptions not supported by the data.

    Getting into long conversations with you becomes maddening. You keep looking for that ethical loophole (and I’m not the only one to notice), you switch between RCT data, populations, individuals, and neuroscience studies on placebo as if they were all the same thing (and you still haven’t addressed that), and then you chastise me for going out of the breadth of science and data and then postulate “well maybe this could possibly work because….” And whenver you get directly confronted, you just pop in a quick “oh sure, well there will always be nutters but….” as if that actually addresses the issue raised.

  71. weing says:

    pm,

    “If acupuncture does not also test your resolve it should…”

    How about sham arthroscopy for knee osteoarthritis? Would you recommend it as nothing else has helped?

    “The ethical arguments against the placebo have been formulated with the assumption that any benefits from them can be readily forgone, but what if they can help when nothing much else will, or if they can perform as well as safely as some drugs, as some preliminary studies suggest?”

    The placebo effect is inherent to the practice of medicine whether you like it or not. SBM is about benefiting the patient above and beyond the placebo. I would prefer resources be allocated to finding out what will really help the patient. If all you offer is the placebo, remember that Pavlov’s dog would stop salivating if all you did was ring the bell and stopped reinforcing with the meat.

  72. pmoran says:

    Then you go ahead and ignore one of the most stressed and salient points that Benedetti made – you cannot evaluate the effect size of a placebo arm from an RCT since it is not designed to actually do that. I’m not seeking to minimize it any more than Benedetti himself states that they are minimal.

    Nybgrus, your understanding of this field is minimal. Benedetti is referring to placebo controlled drug trials wherein it is true that a myriad of influences can be operative in the placebo arm. He is attacking the simplistic idea that all that is measured in the placebo arm of those represents placebo influences as indeed many sceptics do.

    Only those RCTs which compare placebo interventions to no treatment (or waiting list) can bring us close to measuring the combined influence of reporting biases and placebo.

    Those studies ARE designed to “do that”, as best we can, and those studies ARE permissive of worthwhile effects with substantial effect sizes, no matter how much name-dropping you like to offer to support your case and no matter how much support you think you are getting from other contributors. I know that most of those have not looked at the available evidence in depth either and that they are also simply echoing prevailing, cosy, sceptical dogma.

    There is also no way that Benedetti regards placebo influences as being as trivial as you wish them to be. He regards his studies as entrenching the feasibility of significant placebo influences, not minimising them as you seem to want them to do.

    What I have said in my last post still goes.

  73. nybgrus says:

    Be wary of being the lone maverick, pmoran.

    I concede my knowledge of the field is minimal – I am the first to admit I am learning. And yet those many times my superior in experience and knowledge seem to agree with my assessment.

    And to top that off, it is painfully obvious to me that you are attempting to contort evidence and insert gaps in knowledge to justify unethical patient care.

    Weing said it well: SBM is about benefiting the patient above and beyond the placebo

    I am advocating for using placebo as the minimum of interaction regarding patient care. To augment patient care. Not to be patient care.

    Let me be clear – no matter what the actual benefit, even if it outright cures cancer, using placebo in isolation must be and is unethical. Using it in concert with science based medicine is the only way – not recommending acupuncture because you (or EBM) is out of ideas.

  74. jmcohen87 says:

    nybgrus, I wanted you to see this link:
    http://www.youtube.com/watch?v=h4MhbkWJzKk

  75. JPZ says:

    @weing

    “SBM is about benefiting the patient above and beyond the placebo.”

    Awesome summary!

  76. daedalus2u says:

    I have to agree with Harriet and nybgrus.

    The interview was about trying to understand placebo effects and how to use them to best advantage while delivering “standard of care” treatment, not trying to use placebo effects to supplement or replace “standard of care” treatment. SBM practitioners need to understand placebo effects so that they can deliver the best possible “standard of care” SBM treatments, not so they can add a dose of placebo on top of whatever else they do for their patients.

    A practitioner of SBM can’t know what a CAM practitioner will do because there is no data and/or no rational thought behind CAM. Will the CAM practitioner do something benign like reiki, or something horrific like the Lupron Protocol or the Gonzales Protocol, or as Harriet mentions use spinal adjustment to treat meningitis? To send a patient to a CAM practitioner is to send them to someone who is incompetent and does not realize they are incompetent.

    A treating physician has a duty of care that can’t just be abdicated. A placebo CAM treatment could just as easily have adverse as beneficial effects. I know that if my physician recommended a CAM treatment to me, that would adversely affect my expectations that my physician was competent and knew what he/she was doing. I don’t expect perfection from my physician, but I do expect honesty and more important intellectual honesty and ethical treatment. Being told to do something that I knew was bogus would reduce what ever positive placebo effects I got from being treated competently.

  77. DevoutCatalyst says:

    The problem I had with CAM providers is they ALL dissed modern medicine in some way or other, they ALL let on that they had a superior means of knowing — better than doctors. Insidiously they eroded my confidence in real medicine. CAM provides placebo at what cost?

  78. nybgrus says:

    @jmcohen:

    Great video. As I said – I am not super familiar with Goldacre, but everything I have seen, I do love – including this one.

    I’ve actually commented here a few times that the publish or perish and economic pressures associated with generating positive studies is very bad. And the fact that no one wants to publish negative studies is just as bad. I think that all data should be published, and negative studies should unashamedly and clearly say so in their conclusions since we can often learn just as much from knowing what doesn’t work as we can from knowing was does work. And I have said this long ago, both here and in personal communications.

    Goldacre hits the nail on the head when he says that all science needs to be utterly transparent. The problem is that the social, legal, and especially financial pressures on pharmaceutical companies make that nigh impossible for them. I’m not excusing it, simply recognizing the real basis for it. I used to do molecular pharmacology research and read The Billion Dollar Molecule back in my undergrad days and IMO until we change the way we think of health care and its delivery, I don’t see a solution except vigilance such as we see here on SBM.

    If we legislate changes to force such things, then either we’ll completely stifle pharmaceutical innovation or the companies will go belly up. My more libertarian friends want to open up all health care, top to bottom, to free market forces, but that is simply not a tenable model. It is not a system that functions as a free market model, and as Goldacre demonstrated quite aptly, when aspects are opened to free market forces, we see clever ways to manipulate data and alter the perception of evidence. Society at large simply doesn’t have its priorities right on these issues.

    And as a closing side note, since I did notice where else you’d put the video, Goldacre’s discussion in no way changes the critical analysis of MY in regards to treating the flu.

  79. nybgrus says:

    @Dr. Hall – exactly what I have been trying to say. The ends do not justify the means.

    @daedalus: Exactly – I’ve been trying to argue that with some 1st year students. When you send someone off to a CAM practitioner you are literally sending them to a black box where even if you have a reasonable assumption it will be placebo with no harm, that is really a hope and nothing more.

    And as devoutcatlyst points out, and in my own experience having an entire degree in the field of alternative medicine (called medical anthropology, but trust me – absolutely identical), the vast majority of CAM revolves around bashing EBM (referred to with malice as “Western reductionist medicine”) and eroding the trust of patients. I’ll say it till I am blue in the face – my med anthro profs literally yelled at how corrupt and inferior “Western medicine” and how “reductionism” was evil and patently wrong for patient care. The notion of “holism” was taken to the point where any departure from it was an immediate failure – something that I have seen echoed in a less strident form in a few of my med school lectures. One professor even stated that herbals work because of synergy and that science has attempted and failed to reduce it down and thus herbals are superior and we cannot expect to know mor ethan that.

  80. daedalus2u says:

    nybgrus The reason they are so strident is because once you take away the mysticism and the mumbojumbo, they have nothing. This reminds me of a Nietzsche quote:

    “Mystical explanations are thought to be deep; the truth is that they are not even shallow.”

    Or as Pauli said: “Not even wrong.”

  81. pmoran says:

    Harriet: I don’t understand why you keep grasping at straws to justify recommending acupuncture.

    I thought I was stopping just short of that and had explained that placebo influences remain of importance to me in how I behave in OTHER dealings with CAM (in its more benign and possibly helpful forms).

    I am, I suppose, prepared to defend doctors who might “recommend” a trial of acupuncture under certain circumstances as I suspect you would too.

    A doctor who suspects that it really might work in some mysterious way and tries it out with a difficult patient problem may be guilty of being a trifle scientifically gullible but it can hardly be seen as a major ethical transgression. And it is not as though a lot of medical bodies and personages (including the otherwise sceptical Edzard Ernst) have not supported acupuncture for some conditions.

    So what is so outrageous about any use of acupuncture that it provokes the reactions we are seeing here?

    This what I mean by medical scepticism becoming a competitive sport. It risks being marginalized into having even less influence over events than it has displayed so far, because no one understands what all the fuss is about, while also wondering why doctors don’t see it as such a big deal that some of their own methods are not that well supported by science. (Oh, I know all the counterarguments, but if people here have not yet twigged to the fact that the CAM “war” is all about perceptions let them start right now).

    Ethical considerations aside, the general run of mainstream doctors would not be very effective in using placebo-style medicine, which is another reason why I am not so enthusiastic about that.

    They would feel obliged to make similar comments to yours : “I would tell her there was no scientific evidence that it worked –” when for maximal placebo responses what patients really need to hear from trusted advisers is “I have had some amazing results with this!”.

    CAM practitioners will be able to say that with their hands on their hearts, possibly in part because of potent placebo responses in some patients, not merely through the many and varied illusions that we know occur within daily medical practice.

    My point with Nybgrus is that we have no secure basis on which to judge otherwise but he assumes we have simply because others have said so.

    This may well be partly why some perfectly honest CAM practitioners can still believe their methods work even after “studies show they don’t”. Why cannot we sceptics bring ourselves to acknowledge this, and react in the more measured way that such an understanding might warrant?

    Instead they usually get accused of wanting to undermine science or having more despicable agenda. Some of them may merely be trying to make sense of the medical world, as are we with perhaps with perhaps a little more scientific sophistication.

    That is the kind of thing I have in mind.

  82. weing says:

    “This may well be partly why some perfectly honest CAM practitioners can still believe their methods work even after “studies show they don’t”. Why cannot we sceptics bring ourselves to acknowledge this, and react in the more measured way that such an understanding might warrant?”

    We know that it is very difficult for someone to acknowledge that they were wrong. This has been described very well by Carol Tavris in her “Mistakes were made, but not by me.” in the case of recovered memory therapists, prosecutors, and others. Just look at the increase in belief when the world doesn’t come to an end as predicted by their cult leaders. I agree we shouldn’t deride it. I think we need to recognize this default mechanism that we are all susceptible to.

  83. Harriet Hall says:

    @pmoran,
    “I am, I suppose, prepared to defend doctors who might “recommend” a trial of acupuncture under certain circumstances as I suspect you would too.”

    I would make allowances for doctors who believed they were recommending it on good evidence, but I wouldn’t “defend” them. I would ask those doctors to look at the evidence more closely,and I would certainly not recommend it myself.

    Edzard Ernst used to say acupuncture was effective for pain, but in his most recent systematic review of systematic reviews he found otherwise.

    We can understand CAM practitioners and accept that they are sincere and they really believe they are helping their patients in ways that scientific medicine can’t. But that doesn’t mean we should give them a free pass. When they are wrong, and when they mislead patients, we need to speak out.

  84. nybgrus says:

    This may well be partly why some perfectly honest CAM practitioners can still believe their methods work even after “studies show they don’t”. Why cannot we sceptics bring ourselves to acknowledge this, and react in the more measured way that such an understanding might warrant?

    So your argument boils down to “We should be nice to them [sCAMsters] because they truly believe they are helping people and they maybe could possibly be helping out a little bit via placebo effects?” That’s what I am getting from your latest post. Well, that and an assertion that you don’t think we as science based physicians can actually utilize psychogenic placebo effects to augment real medicine. To which I say to you – clearly, it is not I that needs to look at the science better.

    Instead they usually get accused of wanting to undermine science or having more despicable agenda.

    They are accused of such and this is demonstrated with evidence to be true. You are confusing intentional and knowing malice with an ideological agenda – they lack the former, for the most part, but are certainly guilty of the latter.

    the general run of mainstream doctors would not be very effective in using placebo-style medicine,

    They shouldn’t be using placebo-style medicine at all. They should be augmenting actual medicine with psychogenic placebo effects to improve outcomes. And if they aren’t good enough at it…. train them to be! Your argument is tantamount to saying, “The average run of the mill neurosurgeon isn’t very good at alleviating radiculopathy pain, so lets send patients to a chiro.” Um, no. Train the surgeons better.

    A doctor who suspects that it really might work in some mysterious way and tries it out with a difficult patient problem may be guilty of being a trifle scientifically gullible but it can hardly be seen as a major ethical transgression.

    A trifle gullible? OK maybe, though I’d leave off the qualifier and feel the statement is more accurate. As for the ethical transgression – of course not. You are fighting straw men again. If you truly and genuinely believe it has benefit, you aren’t ethically wrong, you are just wrong. But once you know better – which you claim to – then you are ethically wrong.

    My point with Nybgrus is that we have no secure basis on which to judge otherwise but he assumes we have simply because others have said so.

    Put simply and accurately – you are wrong. We do have a secure basis to judge and make positive claims. And your insistent claim that I use appeal to authority is quite irksome and tired. Especially considering that I started out much like Edzard Ernst – I believed homeopathy worked, I thought reductionism was evil, I believed “the placebo effect” was extremely powerful, and I believed in “mind-over-matter” sorts of things. That was when I assumed so simply because others have said so. Then I did this thing called “follow the evidence” and learned the deep basis of the scientific method and came to my own conclusions. When I reference the esteemed authorship here it is not an appeal to authority. It is in direct response to your accusations that I am a neophyte incapable of making accurate conclusion and/or to demonstrate that I need not answer a question because the depth and breadth of it is not only enormous, but has already been answered by the others. It is not to say that I am right because they say so. It is to say I am right, they agree, and if you want a good answer their’s is sufficient. There is a distinct difference between appeal to authority and deferral to expertise and consensus.

  85. jmcohen87 says:

    @nybgrus

    I don’t say MY is effective, but we don’t have good evidence for its alternative and it’s way more expensive…So why not go with the cheaper kind?

  86. pmoran says:

    Nybgrus Let me be clear – no matter what the actual benefit, even if it outright cures cancer, using placebo in isolation must be and is unethical.

    Good grief! You see what extremities can arise when any group, even supposedly science-based medical sceptics get together to feed off each other, and try to outdo each other, even by holding on to perceived orthodoxy in what would be, to say the least, an extraordinarily contentious area of medical ethics?

    Put simply and accurately – you are wrong.

    You cannot even advise me what kind of evidence directly refutes the existence of significant placebo responses within some medical settings.

    AND I know why. There is none. I have been studying this field for decades and if anything the evidence is firming up in support of such non-specific influences — the modern neurophysiological studies, those showing that placebo suggestions can reduce opiate requirements, and very sizeable effect sizes from designated “placebos” when compared to the natural progress of some conditions.

    Also — snap! I was once one of those wanting to argue against the importance of placebo influences.

  87. pmoran says:

    Weing:I agree we shouldn’t deride it. I think we need to recognize this default mechanism that we are all susceptible to

    Exactly. The only reason we know where CAM is going wrong is that we have been there, got that T-shirt, and quite recently, too.

  88. nybgrus says:

    @jmcohen:

    Except for the fact that MY is not an alternative – it doesn’t work as well as the oseltamivir in the study, and the mechanism is not elucidated though it is clear it has nothing to to with viral load. Goldacre’s question was not regarding the efficacy or MOA of oseltamivir it was questioning whether hard end points such as complications are reduced on a population level. Substituting MY makes absolutely no sense.

  89. nybgrus says:

    @pmoran:

    Good grief! You see what extremities can arise when any group, even supposedly science-based medical sceptics get together to feed off each other, and try to outdo each other, even by holding on to perceived orthodoxy in what would be, to say the least, an extraordinarily contentious area of medical ethics?

    Yes, sometimes ethics is a bitch. As I said, the ethics are clear on the use of placebo in isolation. Just as they are clear that forcibly harvesting a kidney to save a life is unethical – both patients will live, one of whom would have died, yet you aren’t arguing for breaching ethics in that case. You are arguing to the wind:

    the modern neurophysiological studies, those showing that placebo suggestions can reduce opiate requirements, and very sizeable effect sizes from designated “placebos” when compared to the natural progress of some conditions.

    [Insert exasperated tone] I agree with you! Pyschogenic placebo effects CAN and DO effect ACTUAL physiological CHANGE. I am NOT arguing that, NOR is anyone else here. It is the UTILIZATION of placebo in isolation as a form of CAM that is the central issue – which is unethical. The issue only slightly second there is the notion that in most cases the placebo effects we see in RCTs are mostly artifact, etc etc.

    I’m sure you’ve been studying this for decades. But you can’t seem to hold a solid line of conversation. This has become muddled enough that there is no point in continuing. Till the next time, pmoran.

  90. daedalus2u says:

    PM and nybgrus; to resolve this ethical dilemma is exactly why I am trying to get my nitric oxide producing bacteria accepted.

    I have posted a link to this many times, but now that there seems to be general recognition among the SBM crowd that there is an actual physiological placebo effect, maybe there can be some thought as to how that physiological placebo effect might be triggered pharmacologically.

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

    Presumably the physiological placebo effect has to do with allocation of metabolic resources. We know those resources are finite, we know that resources used for one thing (healing a wound) can’t be used for something else (running from a bear), we know that organisms allocate them, we know that under stress wounds do not heal as well, we know that wounds do heal under stress, we know that wound healing is a very complex process of many thousands of steps (transcription, protein synthesis, cell mobilization, phagocytosis, autophagy, etc, etc, etc), occurring in a precise order so as to achieve high fidelity repair of exquisitely complex tissues. If the repair remains high fidelity, then the myriad processes are still working together “in sync”. If they are still working together “in sync”, then the rate the processes are happening must be a consequence of the “in sync” control system that is co-regulating the myriad processes “in sync”.

    We know that resource allocation into the fight-or-flight state must be mediated centrally by the CNS where fight-or-flight is triggered. It is the detection of an image of a bear chasing you on your retina, decoded by your visual cortex and interpreted as a bear about to eat you that triggers the physiology that allocates all available resources into fight-or-flight. If you escape from the bear, it will be the centrally mediated bear-detection neuroanatomy that will give the “all clear” and allow the fight-or-flight state to be replaced with something else.

    If you have a benign treatment that can pretty reliably trigger a placebo-like effect, then it is a pretty good treatment to add to just about every other treatment.

    If someone is really stressed out, and is in the fight-or-flight state to some degree, what they need is a way to get out of the fight-or-flight state. There isn’t a good pharmacological way because the fight-or-flight state is an extremely complex state from deep evolutionary time. It is extremely robust, it exhibits hysteresis and cannot be easily hacked into and thwarted pharmacologically.

  91. jmcohen87 says:

    @ nybgrus and everyone else at SBM

    I want to get some opinions about this man’s claim to halt and REVERSE Alzheimers. Please watch all of the videos and give him a fair chance. AND BE OPEN MINDED FOR ZEUS’S SAKE.

    http://www.youtube.com/watch?v=j1FmK4582mA

    This guy – steven fowkes – isn’t your average alternative quack. He knows his stuff.

  92. jmcohen87 says:

    http://www.youtube.com/watch?v=0cqz11RknsQ&NR=1

    I’m curious what SBM and its readers have to say about these videos…
    Quackery or maybe these people are on to something…?

  93. Harriet Hall says:

    @jmcohen87,

    Why should we waste our time watching videos from unreliable sources? For instance, Steven Fowkes has only published 3 articles on PubMed, and they are all in Medical Hypotheses and don’t even address Alzheimer’s. He may talk a good talk, but he has not published any evidence to support his claims. If he could really halt and reverse Alzheimer’s, we would know about it from other sources.

    These people have a track record of presenting a lot of testimonials and speculation, but little else. Why don’t you watch the videos yourself and then research the claims on PubMed? If you find any credible evidence, you can share it with us.

  94. jmcohen87 says:

    @ Harriet Hall

    That’s a pretty limiting and smug way to assess claims made by someone. I would call it groupthink. Steven Fowkes is by any standard an intelligent man and he says he’s been searching for 40 years to try and figure out how to treat AD and he says he finally cracked the code and it isn’t worth 10 minutes of your time to see if he is on to something? I say 10 minutes because if he started spewing misinformation I would then say it’s a sufficient reason to pull the plug.

    “Credible evidence” according to your standard generally takes hundreds of millions of dollars – if not more – to establish. That money can never be recouped without making a patent on the product being investigated. So by the nature of the system it will be exceedingly rare for you to find any “natural treatment” (any treatment which can’t be patented) scientifically proven, even if the treatments abound.

    It’s hard for me to understand that you wouldn’t listen to what he says, even if only to expand your cognitive horizons and see if you could spot a critical flaw in his thinking. Maybe it will spur your own thinking in a direction it hasn’t been before. Or perhaps you can even learn something that you didn’t know!

  95. jmcohen87 says:

    I think it would be a good article to analyze what he says and evaluate it. Attacking completely discredited things like homeopathy and energy medicine is already boring.

  96. daedalus2u says:

    I can’t get the youtube where I am now, but I found his blog

    http://www.projectwellbeing.com/category/alzheimers-disease/

    and I am not impressed.

    He doesn’t mention exercise which is known to be quite important. He doesn’t cite any references or any data. He doesn’t suggest mechanisms by which these interventions would work. He doesn’t provide any way for people who are knowledgeable to check his reasoning.

    If he knows how to reverse Alzheimer’s, why doesn’t he tell us how he knows. What chain of facts and logic led him to the conclusion that these interventions would reverse Alzheimer’s?

    The interventions he is talking about would not cost hundreds of millions to test. They are all over the counter or wholesome foods, their use does not even require an IRB. It does require something to test, the bloggers at SBM can’t figure out if these interventions would reverse Alzheimer’s without looking at data that tests that hypothesis. If this guy doesn’t have the data, and doesn’t have a chain of facts and logic that is compelling (that would be a chain of facts and logic that constitutes a reasonable prior plausibility), how can anyone evaluate it?

    I am especially not impressed by this, or rather I am very negatively impressed.

    http://www.projectwellbeing.com/steve-fowkes/355/

    where he brings up mercury and loss of glutathione as the “cause” of Alzheimer’s. This is not correct. There is no excess mercury in the brains of individuals with Alzheimer’s. Autism is not associated with mercury, with leaky guts, SIDS is not associated with vaccination. The way he talks about physiology shows a profound misunderstanding of what is going on. He does not understand phosphorylation, or thiol physiology.

    He doesn’t have even a modest insight as to what is wrong in Alzheimer’s or how to fix it.

    There are infinitely many things that are wrong. No one has the time, or the inclination to debunk them all.

  97. weing says:

    jmcohen87,

    I did listen to the first video. He makes claims but does not give any evidence to back up these claims. If he has studies backing up these claims, he is not providing them. How can I tell if the mechanisms he describes are true? Because he says so? I belong to the group that says “Show me.”It is up to him to present evidence that his claims are correct. Then we can evaluate the evidence and his claims further. I also have to agree with D2U’s impression of his understanding physiology. Given his misunderstanding of physiology, the likelihood of any mechanism he describes being correct is virtually nil, and would be due to just dumb luck.

  98. nybgrus says:

    @jmcohen:

    You need to understand how science works better. First off, having a robust article printed in a highly respected journal is not that hard. A 9 year old was published as lead author in JAMA. Just recently an article was published in the Journal of Neurourology and Urodynamics on how needing to pee really bad makes your concentration go down. So please disabuse yourself of this notion that getting published in a respected journal is hard or that it takes millions of dollars to get preliminary data into a respected journal.

    So if Mr. Fowkes has anything to offer the scientific community, he can very easily publish a small pilot study and go from there. If it bears merit, I can assure you that many a lab would be very interested in persuing it and if it is indeed novel and useful, his publication will be proof positive that he was the one to have discovered it.

    Anytime someone is the lone maverick, claiming the have the cure to something (especially if that something is a heavily researched and high prevalence pathology), and instead of doing the hard work to put together a good paper on the topic runs around on youtube and has a blog claiming persecution, I can pretty much guarantee you that there is absolutely nothing to the claim.

    You are right, that this response was an example of groupthink – insofar as this group actually thinks.

    And as daedalus demonstrated since he had the time and wherewithal, there is absolutely no validity to this claim, nor any of the thousands of other such similar lone maverick claims.

    Bear in mind we don’t dismiss these ideas because of some conspiracy theory you may have cooked up in your head. It is because we understand how science works, what the pathophysiology involved is, and what the current state of scientific understanding is. And with that kind of background, it is easy enough to dismiss. But the real crux is exactly as Dr. Hall and daedalus have spelled out – we don’t need to nor have an onus to chase down every random “promising” lead. The onus is on the person or group involved to demonstrate that validity.

  99. Why am I always hearing things like “That money can never be recouped without making a patent on the product being investigated. So by the nature of the system it will be exceedingly rare for you to find any “natural treatment” (any treatment which can’t be patented) scientifically proven, even if the treatments abound.”

    I think this is possibly a misinterpretation of patent law. From a patent lawyer found on the web.

    “naturally occurring substances, scientific principles, and algorithms are not patentable in and of themselves.  For example, a plant extract is not patentable, but a method of extracting it or of using it may be patentable.”

    Also consider this Reuters article discussing research that indicates 2/3rd of drugs produced in the last 25 years have come from nature. Drugs like taxol are based on plants and are patentable.

    http://www.reuters.com/article/2007/03/20/environment-drugs-nature-dc-idUSN1624228920070320

Comments are closed.