Acupuncture works - as placebo

[Asolepius]

This BBC item is an exercise in how to mislead with a headline. As Simon Singh would say, there is not a jot of evidence from this review that acupuncture does anything over and above a placebo. Frankly I have given up complaining to the BBC about such grossly misleading reporting, as they have flagged me as a troublemaker and they don't reply now.

 

[Sideroxylon]

The headline just needs a small change by moving the right quotation mark left a bit: Acupuncture 'works' for headaches

 

[Mojo]

The problem seems to be that the BBC doesn't understand that there's a difference between acupuncture and simply sticking needles into people at random (or at least that's what acupuncturists claim).

 

[Ivor the Engineer]

Which review is the BBC article referring to? I can't find anything online from the Cochrane Collaboration reviewing acupuncture for the treatment of headache which has been published recently.

 

[Ivor the Engineer]

The problem seems to be that the BBC doesn't understand that there's a difference between acupuncture and simply sticking needles into people at random (or at least that's what acupuncturists claim).

From the BBC article:

The scientist leading the review said the results showed that putting needles into particular locations might not be that important.

 

[Mojo]

Indeed, but it is a basic principle of acupuncture that it is important where you put the needles, because they claim it works by regulating the flow of qi along meridians. Therefore acupuncture is not the same as merely sticking needles in at random.

ETA: even if it appears to work no better.

 

[Ivor the Engineer]

So if the title of the article was changed to something along the lines of:

'Medical practitioners, perceived by patients as competent in the medicinal uses of needling, sticking needles in patients helps prevent headaches'*

you would be happier?









*Yes, I'm not going to give up my day job to go write headlines.

 

[Spud1k]

It can be a bit of an odd one to call that one. For instance, I personally think that using acupuncture to kick smoking or to stop travel sickness is probably just placebo effect. But then having said that, merely giving someone something to focus on in those situations has been shown to have an effect. In that context, would it apply to say it 'works' even if it is just a placebo?

My mum the physio uses it, but most often as a means of releasing tensed-up muscles by targeting them directly. Regardless of the explanation, that one seems to work as well.

 

[RationalVetMed]

*Yes, I'm not going to give up my day job to go write headlines.

As sound-bites go it leaves something to be desired but it gets the point across!

I'm doing an interview on local radio tomorrow and I'm told this article will be mentioned (amongst other CAM related subjects) so if anyone has any hints or tips for a rank amateur as to how not to look a complete idiot on air I'm all ears.

Yuri

 

[Zeuzzz]

Prayer works - as placebo

Meditation works - as placebo

Laetrile cures cancer - as placebo

Anything works - as placebo

 

[blutoski]

It can be a bit of an odd one to call that one. For instance, I personally think that using acupuncture to kick smoking or to stop travel sickness is probably just placebo effect. But then having said that, merely giving someone something to focus on in those situations has been shown to have an effect. In that context, would it apply to say it 'works' even if it is just a placebo?

The standard way to answer this question is to compare to nontreatment. In most situations, placebo appears to work about as well as no treatment.

Generally, when we're testing things like stop-smoking or weight loss, it's important to pay attention to intention-to-treat, not just the people who obtain results.

 

[Roma]

Well I would like to know if it works or does not work. Plain and simple as that may sound I am sick and tired of being passed off from one doctor to another and never getting any answers or relief for the severe pain I've had about twenty years.
The hospital pain clinic told me last year that there was nothing that could be done but said they would refer me to an acupuncture clinic.
So am I just being passed off again to do something that will be a complete waste of my time and money? I have better things to do.

 

[Spud1k]

The standard way to answer this question is to compare to nontreatment. In most situations, placebo appears to work about as well as no treatment.

Generally, when we're testing things like stop-smoking or weight loss, it's important to pay attention to intention-to-treat, not just the people who obtain results.

I think that about nails it. The NHS offers the 'stop smoking' to pregnant women (or at least used to) because in spite of being horribly unproven, it's one of the few nicotine-free ways beyond going cold turkey. I personally think the logic is that even if it is a crock, the placebo effect will still mean that in some cases, less cigarettes are smoked and more babies are born healthy overall, which makes sense in an NHS way.

 

[Spud1k]

Well I would like to know if it works or does not work. Plain and simple as that may sound I am sick and tired of being passed off from one doctor to another and never getting any answers or relief for the severe pain I've had about twenty years.
The hospital pain clinic told me last year that there was nothing that could be done but said they would refer me to an acupuncture clinic.
So am I just being passed off again to do something that will be a complete waste of my time and money? I have better things to do.

If you're asking 'does it work' in the modern medicine proven-by-double-blind-clinical-studies sense, the answer would be a definite no. But then sometimes you do get some people swearing by it.

Myself, I'd call myself an agnostic on this topic. I don't buy into the whole energy lines mumbo jumbo but it apparently the body does respond (at least locally) to having pins shoved into it and twiddled in some ways that aren't completely understood, so it may be good for something. It's biggest problem is that no-one has yet developed a truly convincing placebo (it's a bit difficult to pretend you're sticking a needle into someone when you're not) and as such, hasn't had the sort of development that you would normally get through doing blind trials and stuff.

 

[Asolepius]

If you're asking 'does it work' in the modern medicine proven-by-double-blind-clinical-studies sense, the answer would be a definite no. But then sometimes you do get some people swearing by it.

Myself, I'd call myself an agnostic on this topic. I don't buy into the whole energy lines mumbo jumbo but it apparently the body does respond (at least locally) to having pins shoved into it and twiddled in some ways that aren't completely understood, so it may be good for something. It's biggest problem is that no-one has yet developed a truly convincing placebo (it's a bit difficult to pretend you're sticking a needle into someone when you're not) and as such, hasn't had the sort of development that you would normally get through doing blind trials and stuff.

It's really important to consider the trends here. Until recently we haven't had credible placebos for acupuncture. Now, two main placebos are being used:

1. `Stage dagger' needles, which touch the skin but don't penetrate because the needle retracts into the handle. These have been tested to see if people can tell the difference, with quite convincing results.
2. Needling outside the meridians.

Neither technique is ideal, and neither can be double-blinded, only single-blinded. But the point is that as the placebo control improves, the effect size decreases. Non-treatment controls in the past showed big effects for acupuncture, but sham controls show effect sizes that fall below statistical significance. If we ever do develop a fully credible double-blind technique for acupuncture, I would predict that all effects will disappear.

I should mention the imaging studies that show effects on certain brain areas in response to needling. I'm not a neurologist, but I suspect that there is little or no consistency between the studies. It really isn't surprising that needling triggers a response in the brain. The question is, is it therapeutically relevant?

Overall, my position has shifted over the last year or so. Before that, I was fairly confident that acupuncture was effective for some types of pain and for post-op nausea and vomiting (PNV). I am now much less convinced regarding pain, as it can increasingly be explained by placebo effect. For PNV, it's well known that pressure on the wrist will help, and anaesthetists commonly do this. It's a curious neurological effect but it's not acupuncture.

 

[Spud1k]

I'm not 100% convinced by the 'stage dagger' needles, partly because they won't create the same sensation (in particular to the 'twiddling' of the needles) and partly because they will still exert pressure, which some people claim should also have an effect.

I just want to be clear about this; the effect I am most open-minded about is being able to help release muscle spasms locally (i.e. not by sticking pins in the opposite side of the body). I've heard different explanations as to why this might be the case, most of them very hand-waving, but regardless, it does appear to have an effect. In the case of headaches, what a lot of people perceive as pain inside the head is sometimes caused by tension in the muscles around it, so I could see how it could help there.

 

[Asolepius]

I'm not 100% convinced by the 'stage dagger' needles, partly because they won't create the same sensation (in particular to the 'twiddling' of the needles) and partly because they will still exert pressure, which some people claim should also have an effect.

My point is not that sham needles are 100% convincing, but that every time we develop a better placebo the effect size decreases. By extrapolation, it seems likely that the hypothetical perfect sham would show no effect.

I just want to be clear about this; the effect I am most open-minded about is being able to help release muscle spasms locally (i.e. not by sticking pins in the opposite side of the body). I've heard different explanations as to why this might be the case, most of them very hand-waving, but regardless, it does appear to have an effect. In the case of headaches, what a lot of people perceive as pain inside the head is sometimes caused by tension in the muscles around it, so I could see how it could help there.

I'm not aware of any studies on muscle spasm. Sure, relieving spasm could relieve headache, but I can't perceive a mechanism for acupuncture relieving the spasm.

 

[Ivor the Engineer]

My point is not that sham needles are 100% convincing, but that every time we develop a better placebo the effect size decreases. By extrapolation, it seems likely that the hypothetical perfect sham would show no effect.

<snip>

No effect compared to no treatment, no effect compared to acupucture, or both?

 

[Asolepius]

No effect compared to no treatment, no effect compared to acupucture, or both?

Clearly the placebo effect is going to stay the same. What changes is the difference between `real' acupuncture and sham acupuncture, as the sham gets more convincing. It's a very good demonstration of the effect of patient expectation on outcome. Expectation of benefit changes perception of pain. The more theatrical the procedure the stronger the effect.

That's an idea. What if we were to invent a totally new fake treatment that had a highly complex and impressive procedure? Then we could give different groups of patients the treatment, each group having a different level of drama and complexity. I predict that the effect size would be greatest with the most theatrical treatment. OK, so what? We already know this from giving patients placebo pills of different colours. The red ones work better I believe.

 

[Ivor the Engineer]

I suggest conducting a trial of needling vs. conventional pharmacological treatment for pain where the patients are told both treatments are unlikely to be particularly effective and are administered by miserable physicians with poor communication skills.