I think you missed a key part of what I was saying: the physical/mental signaling of the person being lifted is exactly what is producing the effect. Using measures to prevent that signaling (e.g. through a chair or backboard) would not be helpful in studying the effect.
You might have missed that because you were thinking about how someone could be lifted up. It might help for me to explain that in the example I gave, the person being lifted extends their arms while pointing down at the floor in either case. So, lifted or not, they have extended arms, pointed at the floor. Clearly there is a detectable difference in the way they hold their arms, since people lifting (or trying to lift) them detect a difference, but the difference is not immediately visible to this casual observer.
Apart from the idea of eliminating cooperation, I think you make a good suggestion. I have thought several times about different ways of demonstrating this technique (blindfolds, nonverbal signals, randomized cues determining where to “focus ki”). I like some of the ones you came up with.
However, with all that said, one does not need to conduct research to critique it, and I believe my critiques of current acupuncture research stand fully as I have expressed them!
Ben Goldacre likes to emphasise that the placebo effect is based in the social and cultural relationship between patient and therapist. As you’ve noticed, we’re describing the same thing with different words.
Medical folk usually don’t think of “placebo” as a pejorative, BTW. We’re all too aware of the vital importance of placebo in healing; medicine is still to a large degree the art of amusing the patient while we wait for the body to heal itself.
Even when applying a non-placebo treatment, the placebo effect is still a vital part of the healing process. Much of the time, placebo accounts for the majority of the effect. Pharmacology and psychology work in concert.
A lot of the popular alt-med methodologies are popular precisely because they strongly invoke the genuine healing power of psychology. Their practitioners work on developing an excellent bedside manner; they devote substantial time to each client; they encourage the client to use them as an informal counsellor during the session; etc.
It’s time consuming and requires skill, but it works. Having an empathetic person be closely attentive to you for an hour or so makes people feel better, for obvious reasons. Adding some superficially dramatic medical theatre to that helps even more; done right, it can trigger the brain into accelerating the body’s innate healing abilities.
But you can invoke that power without spending a fortune on snake oil or buying into an objectively false theory of physiology. That’s where the problem lies.
My grandmother was a nurse, and liked to tell us about hypochondriacs who’d repeatedly come into the hospital insisting that they had persistent headaches or mysterious pains that no medicine could fix. They would often be given a bottle of sugar pills labeled Obecalp (placebo spelled backwards) and the doc would say “take one every morning”. My grandma says it worked every time.
Two sugar pills are a more effective painkiller than one sugar pill.
Salt water injections are even better still.
Red sugar pills are better stimulants, blue sugar pills are better sedatives.
Sham spinal surgery (cut 'em open, do nothing, stitch 'em up) works.
Pacemakers improve heart function after they’ve been implanted but before they’re switched on.
Changing what you tell people about a drug you’ve given them can alter the rate at which the physical molecules are eliminated from their bloodstream. You can literally talk someone into an overdose.
Perhaps. I read into your statement the point that what was being signaled was “I am too heavy to lift,” whereas I was arguing that what is being signaled is “My arms are not stable enough to be used to lift me.” I was more trying to eliminate that as a variable than trying to prevent signaling entirely.
I disagree.
Your entire argument rests upon two things.
The first is the fact that acupuncture works. Fine. That’s a valid point. However, it’s still worth researching whether it works better than, say, massage therapy, or chiropractic, or homeopathy, or other therapies which are theorized to work through the placebo effect. My personal preference is, between similarly effective therapies, to choose the least invasive one. Acupuncture is extremely invasive (as any time you pierce the skin you’re increasing the risk of infection, however marginally), so unless it can be proven to be significantly more effective than alternatives, I think it should be avoided. And “sham acupuncture” is a good way to determine whether the invasive part of acupuncture is, in fact, necessary.
The second is this statement:
…Which you spend no time trying to prove or defend. Your aikido example is a very good illustration of how communication is not necessarily verbal, but it does not seem to tie “communication” back to acupuncture.
I mean, don’t get me wrong, it’s all worded very poetically. I appreciate the imagery. But you offer no convincing evidence for why sticking needles into someone for pain relief isn’t surgery.
I can’t help feeling that the two “sides” are talking past each other here.
“Acupuncture doesn’t work!” might not mean “No one derives any benefit from it” or “It’s not associated with better outcomes” or even “It’s all in your mind!” But if its efficacy derives not from anything that the needles do in or to bodies, then in what sense is it acupuncture that’s working?
If we take a group of people scheduled for, say, stomach surgery, and administer the appropriate acupuncture to half of them while they’re undergoing surgery (and if they don’t know that’s part of the plan), and if they do better or recover more quickly or whatever than the half who didn’t receive acupuncture, then I think everyone would agree, “Yep—acupuncture has its own intrinsic benefit. We might not understand how, but it looks like it’s doing something.”
But if the control and experimental groups have similar outcomes, then we’d say, whatever benefit acupuncture has been shown to provide in other contexts has nothing to do with the acupuncture itself. It’s “only” triggering some other processes that are themselves very effective,
If I tell you that this rock is magic and that when you hold it and say these magic words, it summons an invisible unicorn that will come to your assistance, and if you perform better on a test (or whatever) after you do it—which you might!—what would we conclude? Probably not that magic stones or unicorns are real. Something happened to you. You performed better. But what’s responsible for that? Does it matter? To some people, it matters.
Hm. Yes, I think you don’t have to worry about the “your arm isn’t stable enough” issue – in the standard Ki Aikido version I’ve seen in a few different dojos, people tend to report “he (or she) is too heavy to lift!” after trying sincerely. I’ve tried sincerely, and had that experience. The experience is similar to trying to lift a much more heavy object which is just as solid as the arms that I could lift a moment before.
As far as where my argument rests, that may be so. I’ll address those points!
My argument is that: a) the impact of acupuncture is derived from a pattern of communication between practitioner and patient, rather than through a linear forcing mechanism of the type which makes most drug or surgery interventions effective, and therefore b) the effectiveness of acupuncture is best measured by comparing the clinical effectiveness of a particular style of acupuncture to the clinical effectiveness of another therapy.
You start with the idea that we should compare the effectiveness of different therapies:
I totally agree with you! In fact, my recommendation for acupuncture research is that it focus on clinical effectiveness, contrasted with the effectiveness of other therapies in comparable populations. There are schools of acupuncture which use extremely fine needles, and even some which “needle off the body”. There are also schools which use different theories of point location. These too should be assessed, and if one of them (or sham acupuncture) is more cost effective (including side effects) than a standard biomedical treatment or some other “alternative” therapy, it should be used. If not, we should darn well publish those results!
That is only my consequential claim, however. My primary claim, that acupuncture is not like drugs or surgery, seems obvious to me, but I will try to go through my reasoning.
If you hang out with acupuncturists and talk with them about their practice, they might describe their experience of trying one thing, monitoring the effect (in reports of patient sensation or in changes in the patient’s pulse) and then trying something else. Some points work better for a condition in one patient; other points might work better for a very similar-appearing condition in a different patient. The acupuncture may theorize based on the language of meridians and chi, but fundamentally they are in an empirical search process in which they look for a pattern of response which they and the patient interpret as health.
This is different from a surgical or pharmaceutical intervention. Of course, a doctor may also be trying a series of therapies to resolve a disease condition, but the surgical intervention itself forces a change in the patient’s body, and then relies on the patient’s body to subsequently heal. A pharmaceutical intervention is often less linear in its causality, but many drugs do directly force changes in the body’s chemistry. (A major critique of standard research models that I’ve heard from neuroimmunologists of my acquaintance is that they neglect the circular causality in body systems and ignore many of the chains of effects from interventions, but let’s leave that aside as a side track).
So on one hand we have acupuncture needling, which may or may not produce a significant change in the body, depending on how the body responds to the signal. On the other we have surgery or pharmaceutical interventions, which force changes, regardless of whether the body then heals as a result.
Maybe this is simply a matter of degree – after all, the needle forces some cascade of response when it stimulates or penetrates the body. I do not think so. My metaphor of talk therapy is a better one, in my opinion. Words do physically stimulate the body when they are spoken in one’s presence, but the body’s response is mostly driven by the meaning that the body-mind makes of those words! As far as I can tell, the way people respond to acupuncture much more closely matches the communication model than it does the linear forcing model.
If you don’t find this convincing, then I may not be able to effectively share my reasoning! I hope, however, I’ve been clear.
I would say, @Footface, that if you got much better results with your unicorn rock story than people would get using other stories then we would have zero proof that unicorns are real, but we could certainly say that “Unicorn Rock Therapy” works.
If you are a met met…
Genetic biomarkers of placebo response: what could it mean for future trial design?
The methionine form of the enzyme is three to four-times less efficient at catabolizing dopamine than the valine form. Consequently Met/Met homozygotes have higher levels of prefrontal dopamine relative to Val/Met and Val/Val. Met/Met individuals have a greater tendency to seek and appreciate rewards as well as to confirm new information based on their prior beliefs (confirmation bias).
We recently demonstrated that Met/Met individuals have significantly greater placebo responses than Val/Met and Val/Val and that the response is highest when treated by a warm, caring practitioner.
I think people get hung up on mechanism. There’s no “thing” that we can find when we look for “chi” or “ki”. There’s no separate form of energy that anyone’s detected. Meridians may be related to connective tissue, or may not, but there’s certainly not an easily detected physical structure that matches the meridian charts. So… what the hell? Is it bullshit? What’s happening?
I was telling @nimelennar that I don’t think acupuncture forces a specific response. It’s not like “we removed the spleen; now there’s no more spleen.” It’s not a needle-activated wiring system where certain circuits get turned on or off in a mechanical way. It’s much more like a conversation. So no, there’s no such thing as chi, but yes, acupuncture as therapy is effective.
You are saying that if (some form of) acupuncture is effective while people are conscious, but not while they’re unconscious, then the needling isn’t doing anything.
I wonder if that experiment has been done, but I think we might find one of the following:
Acupuncture isn’t as effective as the other interventions we are using,
Acupuncture is more effective than the others while people are conscious, but not while they’re unconscious, or
Acupuncture is more effective in both conscious and unconscious people.
We agree that in case 1 we shouldn’t consider acupuncture to be effective and in case 3 we should consider it to be in effective. We might disagree about case 2. I think that any method of treatment that reliably produces healing in a cost-effective way should be widely adopted! The mechanism is very interesting, but not important to deciding whether to use it.
The problem is that SSRIs don’t work any better than the older non-selective tricyclics. Pharmacology researchers settled on a neurotransmitter-reductionist consensus and went down a dead-end alley. Eventually they realised that it was a dead end, and raising serotonin is neither essential nor sufficient, but the research expenses had to be recouped somehow (remember the old “Talking-to-Prozac” marketing push, “everyone should take it, like children’s vitamins”?)
I don’t have the luxury of time to source this right now but maybe someone reading the description will recognize the story.
Back in the 16th century or so, they conducted an experiment on a condemned criminal. He was told that he would be executed by exsanguination. They blindfolded him, made a small incision on his neck, but not enough to cause significant harm. Then they positioned a jug of warm water so that the stream would run down the criminal’s neck, starting from the incision site. When a sufficiently large amount of water ran out of the jug, the criminal died.
A moral question arises. If the effects of acupuncture (whatever they are) are placebo, or endorphin release, or the outcome of a therapeutic conversation, then all the impedimentia of qi and meridians and specific spots is charlatanry. As indeed is the idea of a “qualification” in acupuncture. Anyone could perform it, given only a quick course in needle hygiene and “where not to poke”.
But does the placebo work better with a higher fee, justified by the therapist’s qualification, even if it is fraudulent?
If you stop talking about nonsense like qi and all the rest of it—including the idea that there’s an extensive medical-ish discipline someone can be an expert in—will acupuncture still trigger the same whatever-it-is?
Well, my so-called osteoarthritis was cured by rolfing, not acupuncture. And I wasn’t actually trying to cure the arthritis symptoms at the time - I completely believed the doctors who told me it was incurable and that all my bones would ache forever. But I didn’t want the make the story three times as long so I skipped some details :).
I actually had planned to try acupuncture first - I was literally dialing the phone book of alternative medicine - but none of the accupuncturists picked up their phones when I called, and I got fixed before I looped through the entire directory back to accupuncture. So I’ve never actually had it, myself.
I understand the point of meta-analysis (and have read more than a few!) but I believe actual experimentation is superior evidence to reading about other people’s experiments.
YES. Thank you for the post on funnel plots, that was very interesting!
@Max_Blancke, I wrote a response to the Mayo Clinic’s non-explanation of the hit-or-miss efficacy of SSRIs but then I discovered that @Wanderfound said what I wanted to say only he did it better, so I am deleting my less useful response and pretending I never wrote it.
I get migraines, which are believed to be related to serotonin levels. Several of the drugs I have been prescribed affect serotonin. So I have looked into it a little, and SSRIs are a dangerous crapshoot, that you should only take after careful consideration; they can permanently alter parts of your brain chemistry that simply aren’t fully understood yet.
There are several classic variations on the theme. In my very limited and unscientific experimentation, I discovered that young children are less influenced by a test subject “making themselves heavy” than a mature adult is. This is the total opposite of the result I was expecting!
