Acupuncture study

[David Bailey]

Barry'll like this:
BBC NEWS | Health | Needles 'are best for back pain'

Especially the bit saying:

But the Archives of Internal Medicine study also suggests fake acupuncture works nearly as well as the real thing.

Another quote:

After six months 47% of patients in the acupuncture group reported a significant improvement in pain symptoms, compared to 44% in the sham group, and just 27% in the group who received conventional therapy.

So in other words, acupuncture is almost certainly a placebo effect only...

[ducasi]

So in other words, acupuncture is almost certainly a placebo effect only...

Not necessarily – the fact that any needle therapy is almost twice as effective as conventional therapy suggests that there could be something more than placebo happening.

And it looks like acupuncture knows better what to do with the needles once they are stuck into you.

[Barry Shnikov]

Not necessarily – the fact that any needle therapy is almost twice as effective as conventional therapy suggests that there could be something more than placebo happening.

And it looks like acupuncture knows better what to do with the needles once they are stuck into you.

Surprising to me that you need a statistics lesson.

Of the people who had what they thought was needle therapy, 44% who didn't actually have needle therapy said they 'felt better', and 47% of those who actually did have needle therapy said they felt better.

There isn't a statistically significant difference between 44% and 47%, that is to say unless the sample group is in the tens of thousands the difference falls within the experimental uncertainty area.

Therefore treating people with acupuncture and pretending to treat people with acupuncture achieves the same effect, which is about twice as successful as 'conventional' therapy.

It's difficult to imagine a stronger argument in favour of the placebo effect.

Let's not forget, also, that 56% and 53% reported something other than 'a significant improvement'. Most of us would be hoping for better success percentages than that from whatever medical treatment we get.

[Dreadful Scathe]

To demonstrate this to Ducasi, I am now performing voodoo acupuncture by prodding needles into a small linux penguin, with the word linux scored out and "Mac" written underneath. He can tell us if he feels better or worse than before I started. Prodding needles into the penguins bum....now.

[Andy McGregor]

So in other words, acupuncture is almost certainly a placebo effect only...

To me this proves that people have misplaced faith in acupuncture. If they believed they were getting acupuncture they believed their condition was improved. There must have been a placebo group for the conventional therapy. How did that group fare?

As somebody who has been involved in clinical trials for many years I'm always amazed at how well placebo works. At one time there was even a drug that doctors could give that was a placebo - worked brilliantly for many patients and had very few side-effects. Unfortunately it was withdrawn for ethical reasons. Doctors now have to resort to sub-therapeutic doses if they want a placebo effect.

[ducasi]

Of the people who had what they thought was needle therapy, 44% who didn't actually have needle therapy said they 'felt better', and 47% of those who actually did have needle therapy said they felt better.

No, the 44% had a "sham acupuncture" needle therapy, but it could still be called some sort of needle therapy.

There isn't a statistically significant difference between 44% and 47%, that is to say unless the sample group is in the tens of thousands the difference falls within the experimental uncertainty area.

Here are the numbers... "verum vs sham, 3.4% (95% confidence interval, –3.7% to 10.3%; P = .39)".

This indeed shows that the difference between the two results is within the confidence interval, but until there are more tests, I'd say these result suggest that "verum" acupuncture is possibly better than "sham" acupuncture.

Therefore treating people with acupuncture and pretending to treat people with acupuncture achieves the same effect, which is about twice as successful as 'conventional' therapy.

Indeed.

It's difficult to imagine a stronger argument in favour of the placebo effect.

That's a big leap. Maybe sticking needles into people does actually produce a physiological response which can alleviate back pain.

From the article...

The researchers ... say their findings suggest that the body may react positively to any thin needle prick - or that acupuncture may simply trigger a placebo effect.

Without more research both conclusions are valid.

Let's not forget, also, that 56% and 53% reported something other than 'a significant improvement'. Most of us would be hoping for better success percentages than that from whatever medical treatment we get.

And it looks like conventional medicine isn't providing it...

[ducasi]

To demonstrate this to Ducasi, I am now performing voodoo acupuncture by prodding needles into a small linux penguin, with the word linux scored out and "Mac" written underneath. He can tell us if he feels better or worse than before I started. Prodding needles into the penguins bum....now.

I wondered why my Mac had crashed!

[stewart38]

Surprising to me that you need a statistics lesson.

Of the people who had what they thought was needle therapy, 44% who didn't actually have needle therapy said they 'felt better', and 47% of those who actually did have needle therapy said they felt better.

What do you mean by 'needle therapy' , read the report again

-----------------------------------

Although needles were used in the sham therapy, they were not inserted as deeply as in standard acupuncture. Neither were they inserted at points thought key to producing a therapeutic effect, or manipulated and rotated once in position.

After six months 47% of patients in the acupuncture group reported a significant improvement in pain symptoms, compared to 44% in the sham group, and just 27% in the group who received conventional therapy.

---------------------------------


What it suggest is needles work but we are not very good (yet) of knowing how or where to put them. You could argue how or where you use them is of no relevance based on these results end of stats lesson

[Dreadful Scathe]

No, the 44% had a "sham acupuncture" needle therapy, but it could still be called some sort of needle therapy.

Hmm I suppose you have a point, but its not much of a boost for acupuncture practitioners when there's proof that sticking random needles in a random order into someone is equally as good for them. It does sound like proof to me that acupuncture itself is entirely worthless - its supposed to be a skilled practice after all.

[stewart38]

Hmm I suppose you have a point, but its not much of a boost for acupuncture practitioners when there's proof that sticking random needles in a random order into someone is equally as good for them. It does sound like proof to me that acupuncture itself is entirely worthless - its supposed to be a skilled practice after all.

The group is far too small to make sense of the variation found i.e. 47% acupuncture 44% sham (but with needles)

You could only say it better then no needles ?

A 47% /44% variance is better then most house odd in Vegas hardly ‘insignificant ‘ but again it depends on numbers

[David Bailey]

It does sound like proof to me that acupuncture itself is entirely worthless - its supposed to be a skilled practice after all.

Yes, it sounds like there's practically zero effect from the "study" of acupuncture - so in other words, all this stuff about primary, pathways is clearly rubbish.

There was a guy on the radio talking about this last night, and he was coming up with the standard rubbish arguments pro-acupuncture (e.g. "A billion Chinese can't be wrong" - I guess that means we should forget this democracy nonsense on that basis then ).

I hadn't caught his name, but assumed he was an industry representative (as the other interviewee was Ben Godlacre of Badscience.net - I was then shocked at the end of the piece to realise that the "pro-acupuncture" guy was actually a government representative... - that's my money he's wasting

What this study says to me is:
- Sticking needles in people may well have some objective beneficial result
- Or, it may be that placebos have a good result
- But this result can be achieved by anyone
- So training people to be acupuncturists is a total waste of money.

[Aleks]

Hmm I suppose you have a point, but its not much of a boost for acupuncture practitioners when there's proof that sticking random needles in a random order into someone is equally as good for them. It does sound like proof to me that acupuncture itself is entirely worthless - its supposed to be a skilled practice after all.

I sincerely hope that all the hours I spent learning the meridian and 5 element theory to get my degree isn't invalid. Maybe I just needed to learn it to prove I'm capable of retaining large quantities of (useless) information.

[stewart38]

Yes, it sounds like there's practically zero effect from the "study" of acupuncture - so in other words, all this stuff about primary, pathways is clearly rubbish.

There was a guy on the radio talking about this last night, and he was coming up with the standard rubbish arguments pro-acupuncture (e.g. "A billion Chinese can't be wrong" - I guess that means we should forget this democracy nonsense on that basis then ).

I hadn't caught his name, but assumed he was an industry representative (as the other interviewee was Ben Godlacre of Badscience.net - I was then shocked at the end of the piece to realise that the "pro-acupuncture" guy was actually a government representative... - that's my money he's wasting

What this study says to me is:
- Sticking needles in people may well have some objective beneficial result
- Or, it may be that placebos have a good result
- But this result can be achieved by anyone
- So training people to be acupuncturists is a total waste of money.

People and stats and what is ‘insignificant’


Ok

You have a roulette table assume 18 black and 18 red and one green 0 (some have two but assume 1)

Assume 10,000,000 turns of the roulette will

What’s the chance of getting a red ?

What’s the house take per turn (long run)


If you shifted these ‘odds’ by 9% what would/could happen ? (47 and 43)

[David Bailey]

People and stats and what is ‘insignificant’

4% of 1,200 is not trivial, but it's not massively conclusive either, IMO.

And one obvious explanation for the variance might be that the "sham" acupuncturists didn't "fool" some of their patients, so the placebo effect was less pronounced - so the "proper" acupuncturists were more convincing liars, basically.

But even if "proper" acupuncture were objectively shown to be a massive 4% better than "sham" acupuncture, that's not exactly a ringing endorsement.

[Tessalicious]

What this study suggests is simply that 'puncture' is useful in general, it doesn't have to be 'acu'.

Fortunatly, I don't have to worry too much either way - I don't let anyone near me with a needle or any size, shape or accuracy if I can help it...

[cms]

I sincerely hope that all the hours I spent learning the meridian and 5 element theory to get my degree isn't invalid. Maybe I just needed to learn it to prove I'm capable of retaining large quantities of (useless) information.

completely agree! i have used acupuncture to great effect on many people - and not just patients! my ex boyfriend had cerebral palsy, ran marathons and had knee pain. after reading a similar case study i tried it on him and the results were amazing! so it does appear to work............
but then - whats this one study compared to the many others that have been done over the years

C x

[David Bailey]

but then - whats this one study compared to the many others that have been done over the years

Such as?

[Barry Shnikov]

That's a big leap. Maybe sticking needles into people does actually produce a physiological response which can alleviate back pain.

That isn't what acupuncture teaches though, is it?

It teaches that it alters the flow of chi around the body.

Sticking needles into people to alleviate pain is explainable under conventional medicine.

[Barry Shnikov]

This indeed shows that the difference between the two results is within the confidence interval, but until there are more tests, I'd say these result suggest that "verum" acupuncture is possibly better than "sham" acupuncture.

Well, you'd be wrong to say that, because your own paragraph states that the tests actually do not show that.

[Barry Shnikov]

And it looks like conventional medicine isn't providing it...

Conventional medicine can't cure ebola, but that's no reason to listen to somebody who says extract of his toe cheese - or anything else - will do the job.

We need to start with the assumption that it may be impossible to cure some ailments, and then go on to see what does work, rather than from the assumption that all ailments are curable and if conventional medicine won't do it let's rush into the arms of the quacks.