Can we please start a discussion on acupuncture, here. Should all players get needled before tournaments or should it remain voluntary?

http://www.usgo.org/news/2010/12/couldn ... ight-help/

Acupuncture is a sham. "Couldn't hurt, might help?" That's exactly the way sham medicine is sold to people.

_________________
My blog about Macs and more: Kirkville

That's an amazing leap of false logic

What? The fact that it's a sham (which is true), or the fact that sham medicine often uses the "couldn't hurt" argument (which is true as well)?

_________________
My blog about Macs and more: Kirkville

Many people are willing to pay 100-200% markup just because a product has white plastic or a fruit logo. Doesn't mean it's a sham, everybody is happy. Also, even if the scientific basis is similar to homeopathy, placebo effects are significant, so it does work for some people.

They won the tournament. So, it didn't hurt

Well, it's your opinion that it's a sham, which isn't the same as being fact. A lack of evidence for something working doesn't mean it doesn't work, and I'm not aware of a large number of papers showing a null effect - absence of evidence isn't evidence of absent. There are even some (including recent reviews on specific health issues) that actually appear to show a positive effect in practice of acupuncture, though this is outside the field of science that I spend much of my time. I don't believe in it particularly, but I haven't spent a great amount of time looking for its possible effectiveness - however, I don't disbelieve in it for the same reason. You can't prove a negative, but you can systematically and repeatedly demonstrate a null in bigger and better conducted studies until you finally have to conclude there's nothing to look for. That hasn't yet been done for acupuncture as far as I know.

But no, the false logic was saying "just because sham medicine and acupuncture can be referred to as couldn't hurt, might help, acupuncture is a sham". If I have a headache, I might drink some water - if I'm dehydrated and that's causing my headache, that will help. If it's something else, it probably won't. However, it probably won't hurt, so "couldn't hurt, might help" could be applied to that drink of water - does that make water helping with headaches a sham too?

If you want to read about acupuncture, and the supposed longevity of the technique, its results, etc., this is a good document:

http://www.centerforinquiry.net/uploads ... _Paper.pdf



I didn't say that. Re-read what I said:



I didn't say that because they can be referred to using that expression meant that acupuncture is a sham. I said that the title of the article uses the same type of "slogan" that is used for sham medicine. I merely stated that acupuncture is a sham.

As for any placebo effect, that is certainly possible. However, I don't see that as a logic to allow people to sell such a treatment. (I feel the same about homeopathy, which is also a sham.)

Regarding any possible "proven" successes of acupuncture, this blog (http://www.sciencebasedmedicine.org/?cat=8) has a number of articles showing how poor such studies are. For example, this one about acupuncture as a treatment for amblyopia (http://www.sciencebasedmedicine.org/?p=9030) is particularly interesting, and shows how poor studies can end up getting publicized. In fact, the article's conclusion is a nice, concise answer to the question:

"Acupuncture is ultimately a shell game of preliminary unreliable results and misinterpreted non-specific/placebo effects."

_________________
My blog about Macs and more: Kirkville

Your grasp of language is good enough to be aware that most readers will take the two sentences as tied with some level of reasoning logic. If these really are two totally independent sentences with no intended association, why post them?



Blogs are not science, regardless of whether those authoring it are qualified medical doctors. Most such sites I've found, particularly ones that spend their time referring to pseudoscience and quacks, are as poorly reasoned as the "Homeopathy cures cancer because it made my dog better" brigade. I consider neither of those links particularly convincing, and have a strong suspicion having now done another, slightly more detailed Pubmed search, that they are just as guilty of cherry-picking their null studies as the other side are in advocating papers that find an effect. That's not to say I don't thoroughly approve of an in depth analysis of methodologies and conclusions of papers, as it is absolutely essential in my opinion for the proper progress of science. However, it is equally important to scrutinise papers that agree with your opinion as it is to scrutinise those that don't, and that doesn't appear to be the case on this blog.

Even reviews published in highly rated peer reviewed journals are not above being simply wrong, but they at least have passed a few valid tests of credibility to get there. A blog has virtually zero credibility in the scientific community, and rightly so.

i hurt my ankle badly and was limping for 5 years.
i went to korea and had two session of niddle penetrating from one side to other side (ankle), since then i was able to walk straight.

is it a sham?
many doctors could not fix my problem but two 5 min session fixed it 100%.
i was non believer at first but had to give in before the result.

there are many things that you can not explain why it works but many times it is true.

_________________
"The more we think we know about
The greater the unknown"
Words by neil peart, music by geddy lee and alex lifeson

Actually, the people writing there are very well respected in the medical community. Just because it's a "blog" doesn't mean they're not serious. I think that's a very poor assumption on your part.

_________________
My blog about Macs and more: Kirkville

If you re-read my post, I never said they aren't serious. Most long term blog writers are, or they would stop. I also did not say that they lacked competence in their field. However, the credibility statements isn't an assumption, it's just the way science works. There are all sorts of controls to make sure that research that is used to inform policy and guidance has passed a few semi-rigorous tests on relevant criteria. For all we know, the article you linked to could have picked one study in the face of 20 other studies published on the same association, and it was the only one that didn't find a statistically significant positive association in a well controlled double-blind environment. They say that's not the case, and anyone is free to take their word for it (I suspect that's not the case too, but the example stands). Science cannot be based on blogs, regardless of the respect the authors have in their particularly environment.

Zero credibility is not the same as "they talk nonsense", or "the content of their blog is wrong". It's a statement that the material they post specifically in their blog has not passed any formal scientific assessment of veracity or accuracy. As a result, it has a very low weight when compared to peer reviewed research ... even when compared to the studies they themselves are criticising. If their view is a better, more completely argued case than the hypothesis put forward in the study, then this is sad, but it is also just life. To increase the weight of their analysis, it should be published as a correspondence in the journal of the original paper, or they should get together and do a formal meta-analysis or literature review and get that published in a peer reviewed journal. For all we know, the content they have written would not pass muster, and there's no way of finding out because anyone can post on a blog.

Isn't this putting the burden of evidence in the wrong place? I think that, until acupuncture has been proven effective (beyond the placebo effect) in a sufficient number of controlled double blind studies, the null hypothesis should be: "Acupuncture doesn't work"



I didn't read Kirk's post that way, but I guess you could read it that way. To me, he seemed to be making two statements, without saying that one implied the other.

[quote="HermanHiddema"]
Isn't this putting the burden of evidence in the wrong place? I think that, until acupuncture has been proven effective (beyond the placebo effect) in a sufficient number of controlled double blind studies, the null hypothesis should be: "Acupuncture doesn't work"
[quote]

Of course. This is the case for any medication/therapy/technique/cure/etc. Just because people believe it works doesn't mean it works (in part because of placebo, in part because, in many if not most cases, the body heals itself anyway).

BTW, just because I posted a link to a blog and an article doesn't mean that _only_ bloggers contest the value of acupuncture. I think it's obvious that they are just representative of the general medical opinion (though if you get your medical information from Oprah, it may be skewed in the other direction).

_________________
My blog about Macs and more: Kirkville

No, as there's no burden of evidence required. You are correct that this is the null hypothesis, and for it to be "true" then yes, the burden of evidence is on demonstrating it (or rather, consistently demonstrating the falseness of the positive hypothesis). Stating explicitly that something doesn't work does need a body of evidence failing to demonstrate that it does work. However, there's a big but semantically important difference between "it doesn't work", and "there's no strong evidence that it does work". The latter requires no burden of evidence, and it needs to be shown to be incorrect by the burden of evidence on the other side before such treatments should be brought into clinical practice.

In practice then, without good evidence, people can choose to use it or not, but I don't believe that clinical practices should employ it as a technique when they have better supported methods that they can use instead. If the scientific literature is suddenly filled with papers demonstrating the efficacy of acupuncture, then its use can be reassessed, until then "no good evidence that it works" is enough to not use it. However, if you want to make statements like "it doesn't work", or "it's a sham", then you choose to take the burden of evidence onto yourself to support it, as you've made a definitive statement just as strong as "it does work".



Fair enough, I'm happy with the possibility that I was in the minority (maybe even a minority of 1) for misinterpreting it

An analogy to put across my point:

Most people are now pretty happy that smoking causes lung cancer. Not just associated, but actually causes it. At the moment, the precise causal mechanisms are not really understood (as is the issue with a number of cancers), but the evidence is sufficiently strong that it's accepted by the scientific and medical establishments. That is thanks in part to the work of Sir Richard Doll and others in the 50s pooling large quantities of epidemiological data and collating some experiments involving nicotine and tar and rats.

Does this mean that prior to this, say, the 1940's, that the null hypothesis "smoking does not cause lung cancer" was true because sufficient evidence to suggest otherwise hadn't yet been published? The answer obviously is "no", because smoking does cause lung cancer. At that point, the state of science said "there's not enough evidence to support smoking causing lung cancer" - neither "yes" nor "no" - and it required continued research to push it one way or the other. There is a burden of evidence required for both yes and no, but there is also a neutral middle ground of "no-one knows" that is the default position, for which no burden is required.

There is the more contentious issue is whether treatments in the "no-one knows" should be allowed to be sold as treatments or paid for services. My personal feeling is that the tax payer should not be contributing national contributions to a state funded implementation of non-demonstrated treatments, but I also believe people should have the right to spend their personal money on it if they want to.

Attention conservation notice: a bunch of claims about evidence and rational credence/confidence that aren't specifically about acupuncture at all. Mostly disagreeing with topazg.

1) Implications are rarely at issue in scientific practice. What we're interested in are inductive support relations, which are not a matter of implications (quick: how many times does the sun have to rise before that implies that it will rise again tomorrow?).

Similarly, if a treatment (and I am not saying this is true of accupuncture, because of my ignorance) was entirely propounded by known con-men with a regard of selling fraudulent treatments, that would not imply anything about its efficacy, but the rational assumption would be that it does not work.

2) Leaving con-men to the side, if I take some random procedure, say poking at a spot on your elbow with a candy cane, then absent any further evidence, your credence that it has therapeutic benefits should be very low. You can say "you should just withhold judgment" but there's another compelling way to describe things.

You do not merely say "there's no evidence that it's effective" but you should have a reasonably high degree of belief that it's not. After all, there is an indefinite set of things you can do to a person, most of which have no therapeutic effect. If candy cane poking was not recommended by some positive evidence, it is just another randomly selected treatment with nothing to recommend it.

Your belief will lack something called resiliency. Although you have a reasonably high confidence that candy cane poking is ineffective, your belief will not persist if you get contrary evidence. Given one (good) study that suggests effectiveness, your credence that it's effective will be much higher. The contrast is a treatment that's been extensively studied and found to be ineffective. That belief is resilient given one good looking study that suggests it works, you'll tend to believe that it was simple chance. You will require multiple studies to substantially update your belief.

Note that your probability estimate that the treatment is effective can be the same in both cases. In each case, you might think that there's a 10% chance that the treatment is effective. What differs is the conditional probabilities that it's effective, given new evidence about its efficacy.

Ordinary language uses words like "confidence" to cover both credence and resiliency (and probably more things besides). There's a big advantage to distinguishing between them.

3) Obviously the question is not about whether what's true changes. It's about what is reasonable to believe. We won't get very far asking if "smoking doesn't cause lung cancer" was true in 1940--we'll get further asking what scientists should have believed.

_________________
Occupy Babel!

I don't know, are you sure you disagree with me? I pretty much agree with everything you've just written

So, as an analogy, people should be free to be, for example, scammed by Bernard Madoff, who told them that they were going to make money when, indeed, they lost money? It comes down to something similar: either you want the truth, or you want to allow anyone to claim anything, and, as long as it money spent by individuals, tough titty?

_________________
My blog about Macs and more: Kirkville

Kirkmc: I think the Madoff analogy is badly flawed. In his case, there were legal requirements concerning disclosure and account keeping, and he was misrepresenting everything there--he claimed to be buying and selling stocks and other assets, producing a certain return, where nothing of the sort was happening.

_________________
Occupy Babel!