Acupuncture placebo better than acupuncture

[Skeptic Ginger]

Could it be possible, that the people that do not belive in Acupuncture are searching for confirmation?
Because in studies where there are also is a group that gets scientific medicine, and the result shows, the Acupuncture AND the sham Acupuncture both seem to be better than scientific medicine, i see alot handwaving from the people that "know" it doesnt work.

In defense of the disbelievers, one starts with a couple problems here. First, the proposed mechanism of action is bogus. No "chi" has ever been detected and no significance of acupuncture points has yet been described. No mechanism of action has been proposed. This after thousands of years of use and more recently, plenty of opportunity to investigate such mechanisms. The most we have is referred pain pathways and evidence stimulating certain nerve pathways can affect pain perception.

The second problem is it is up to the proponents of acupuncture to prove or at least demonstrate it works. It is not up to the skeptics to assume it works unless disproved.

Since there are studies and articles of anecdotal evidence giving acupuncture the generous benefit of the doubt, I'd say that skeptics have gone out of their way in this case to find confirmation studies. They simply are not there.

 

[Skeptic Ginger]

....it does suggest that psychological factors can influence pregnancy in IVF.

Linda

Which suggests the subjects knew the difference between the sham and the actual acupuncture in this study. Interesting.

 

[DC]

Dictator Cheney,

If you could provide a link to the page that includes the guidelines for NSAID use in the Empfehlungen zur Therapie von Kreuzschmerzen, I think I could clear up this issue by looking at a google translation.

Linda

http://www.akdae.de/35/64-Kreuzschmerzen-2007-3Auflage.pdf

they only have the newest guidelines online, back then they was using olders one, i dont know how what was achanged.

Those are the guidlines used in germany for therapies and pharmaka for Lower back pain (afaik), in Acupuncture they also mention the GERAC studie.

They say correctly that the goal of the studie, to show significant better performing of Acupunctur vs sham Acupuncture, failed. But they also point out that both the real and the sham acupuncture did perform better than the standard therapy. Theyr conclusion is that there is no evidence that acupuncture works.

So the organisation that makes those guidelines, is writing about that studie in the newest guideline, the studie claimed to have based the standard therapy on the guidelines that come from that organisation.

So i would think that they would point out that the standard therapy was not followed correctly.

But im sure the guidlines from Germany and the USA are diffrent.
i cant translate those guidelines, i hope the link will help solving it.



i also read about another studie where i was not able to get full documents, only journals that reported about the studie.

does anyone know this journal?
http://www.anesthesia-analgesia.org/

 

[DC]

In defense of the disbelievers, one starts with a couple problems here. First, the proposed mechanism of action is bogus. No "chi" has ever been detected and no significance of acupuncture points has yet been described. No mechanism of action has been proposed. This after thousands of years of use and more recently, plenty of opportunity to investigate such mechanisms. The most we have is referred pain pathways and evidence stimulating certain nerve pathways can affect pain perception.

The second problem is it is up to the proponents of acupuncture to prove or at least demonstrate it works. It is not up to the skeptics to assume it works unless disproved.

Since there are studies and articles of anecdotal evidence giving acupuncture the generous benefit of the doubt, I'd say that skeptics have gone out of their way in this case to find confirmation studies. They simply are not there.

i am not really sure what the TCM claim about the mechanism is in detail. I doubt they know it nor that their claim will be correct about the mechanism, when indeed something is working about it.

But there are scientists that try to find out more about it, by using modern methods.

And alsong no real evidence can be produced, one should stay sceptic, but should not claim to know its not working.
There are many people that tryed it, and belive or maybe even know it works for some things. like backpain.

 

[Jeff Corey]

i am not really sure what the TCM claim about the mechanism is in detail. I doubt they know it nor that their claim will be correct about the mechanism, when indeed something is working about it.

Of course not, since it doesn't work.

"But there are scientists that try to find out more about it, by using modern methods.

And alsong no real evidence can be produced, one should stay sceptic, but should not claim to know its not working.
There are many people that tryed it, and belive or maybe even know it works for some things. like backpain.[/QUOTE]

But they are wrong.

 

[fls]

http://www.akdae.de/35/64-Kreuzschmerzen-2007-3Auflage.pdf

Okay. It was too big for a google translation, but I translated the relevant bits and it looks like NSAIDs are recommended as per the usual evidence-based use. From the paper, it looks like patients in the conventional arm were on NSAIDs as per the guidelines.

Giving NSAIDs for a maximum of two days a week looks like a relatively unrelated guideline, specific only to the study, to use as a rescue medication for acute pain in all three groups.

Linda

 

[Jeff Corey]

Which suggests the subjects knew the difference between the sham and the actual acupuncture in this study. Interesting.

Like you could tell if someone poked you with a pointed stick, rather than fresh fruit?
http://www.youtube.com/watch?v=piWCBOsJr-w

 

[Skeptic Ginger]

i am not really sure what the TCM claim about the mechanism is in detail. I doubt they know it nor that their claim will be correct about the mechanism, when indeed something is working about it.

But there are scientists that try to find out more about it, by using modern methods.

And alsong no real evidence can be produced, one should stay sceptic, but should not claim to know its not working.
There are many people that tryed it, and belive or maybe even know it works for some things. like backpain.

DC, I have looked into acupuncture research very thoroughly and have concluded that currently there is no evidence supporting it's efficacy. If a patient finds acupuncture successfully treats his or her pain, and especially if that patient subsequently avoids narcotic pain medications, then I would do nothing to discourage that patient from using acupuncture. But I could not recommend someone try it on the possibility they would get a placebo benefit. That is not justified by the evidence.

 

[DC]

Of course not, since it doesn't work.

"But there are scientists that try to find out more about it, by using modern methods.

And alsong no real evidence can be produced, one should stay sceptic, but should not claim to know its not working.
There are many people that tryed it, and belive or maybe even know it works for some things. like backpain.

But they are wrong.[/QUOTE]

how do you know?

University of Jena Germany Published a studie and claim to have proven with that double blind study, that acupuncture works painkilling. and that it is not a placebo effect.

I cant find it in english.
but it was in the "Anesthesia & Analgesia" (2004; 98: 141-147) journal.
whish claimed to be an accepted and peer reviewed journal. But i dont know.
Nor cant i find a full documents on the study.

 

[DC]

DC, I have looked into acupuncture research very thoroughly and have concluded that currently there is no evidence supporting it's efficacy. If a patient finds acupuncture successfully treats his or her pain, and especially if that patient subsequently avoids narcotic pain medications, then I would do nothing to discourage that patient from using acupuncture. But I could not recommend someone try it on the possibility they would get a placebo benefit. That is not justified by the evidence.

Did you also read about that 2004 studie from the Jena University?

 

[DC]

Winfried Meissner, MD*, Thomas Weiss, MD, PhD, Ralf H. Trippe, PhD, Holger Hecht, Clemens Krapp, MD*, and Wolfgang H. Miltner, PhD

Departments of *Anesthesiology and Intensive Care and Biological and Clinical Psychology, Friedrich Schiller University Jena, Jena, Germany

Address correspondence and reprint requests to Winfried Meissner, MD, Department of Anesthesiology and Intensive Care, Friedrich-Schiller-University Jena, 07740 Jena, Germany. Address e-mail to [email protected].

The effect of acupuncture on pain perception is controversial. Because late amplitudes of somatosensory evoked potentials (SEPs) to noxious stimuli are thought to correlate with the subjective experience of pain intensity, we designed this study to detect changes of these SEPs before and after acupuncture in a double-blinded fashion. Sixteen volunteers were anesthetized by propofol and exposed to painful electric stimuli to the right forefinger. Then, blinded to the research team, the acupuncture group (n = 8) was treated with electric needle acupuncture over 15 min at analgesic points of the leg, whereas the sham group (n = 8) received no treatment. Thereafter, nociceptive stimulation was repeated. SEPs were recorded during each noxious stimulation from the vertex Cz, and latencies and amplitudes of the N150 and P260 components were analyzed by analysis of variance. P260 amplitudes decreased from 4.40 ± 2.76 µV (mean ± SD) before treatment to 1.67 ± 1.21 µV after treatment (P < 0.05), whereas amplitudes of the sham group remained unchanged (2.64 ± 0.94 µV before versus 2.54 ± 1.54 µV after treatment). In conclusion, this double-blinded study demonstrated that electric needle acupuncture, as compared with sham treatment, significantly decreased the magnitudes of late SEP amplitudes with electrical noxious stimulation in anesthetized subjects, suggesting a specific analgesic effect of acupuncture.

IMPLICATIONS: This double-blinded study demonstrates that electric needle acupuncture, as compared with sham treatment, significantly decreases the magnitudes of late somatosensory evoked potential amplitudes with electrical noxious stimulation in anesthetized subjects, suggesting a specific analgesic effect of acupuncture.

http://www.anesthesia-analgesia.org/cgi/content/abstract/98/1/141

 

[DC]

http://www.uni-jena.de/img/unijena_...iopsych/Literatur/Meissner_AnesthAnalg_04.pdf

the full study paper in english of that small (to small IMHO) study from the Uni Jena.

 

[Jeff Corey]

"Then, blinded to the research team, the acupuncture group (n = 8) was treated with electric needle acupuncture over 15 min at analgesic points of the leg, whereas the sham group (n = 8) received no treatment...]

It appears that they have no idea of what a double blind study is. And since you must have read this before quoting it in your post, you don't either.

 

[Jeff Corey]

http://www.uni-jena.de/img/unijena_...iopsych/Literatur/Meissner_AnesthAnalg_04.pdf

the full study paper in english of that small (to small IMHO) study from the Uni Jena.

I assume you mean too small. No, that is a common error. The results would be quite impressive if the study were properly designed wiith an appropriate placebo control group.

Studies with large numbers of subjects can achieve statistical significance with trivial effects or effects from a hidden confounding variable, like a small file drawer effect in a ganzfeld study. Successful replications of small studies is much more convincing.

If you meant that the print in study was too small to read, I'd agree.

 

[fls]

http://www.uni-jena.de/img/unijena_...iopsych/Literatur/Meissner_AnesthAnalg_04.pdf

the full study paper in english of that small (to small IMHO) study from the Uni Jena.

Yeah, too small. The result seems to depend upon a few unusually high readings that happened to be sorted into the acupuncture group.

Linda

 

[fls]

I assume you mean too small. No, that is a common error. The results would be quite impressive if the study were properly designed wiith an appropriate placebo control group.

How would you have set up the placebo control group?

Studies with large numbers of subjects can achieve statistical significance with trivial effects or effects from a hidden confounding variable, like a small file drawer effect in a ganzfeld study. Successful replications of small studies is much more convincing.

There is a happy medium.

Small numbers suffer from increasing the relative possibility of false positives (by decreasing the number of true positives). They are also susceptible to hetergeneity, which becomes a problem if the outcome is dependent upon outliers - it's easy for the outliers not to end up distributed evenly between the two groups.

Linda

 

[Jeff Corey]

Sham needles. http://www.acupuncturetoday.com/archives2002/apr/04needle.html
As to n and power, "However, if too many observations are used (or if a test is too powerful with a large
sample size), even a trivial effect will be mistakenly detected as a significant one. Thus,
virtually anything can be proved regardless of actual effects.(from indiana.edu/stat-math), as I'm sure you know. Since I can't read the article, did they use a t-test? The t is pretty robust according to Alan Boneau's Monte Carlo studies.
http://www.garfield.library.upenn.edu/classics1986/A1986D597700001.pdf

 

[Dancing David]

I think it's simply awkward wording.

Linda

It looked to me as though they only got NSAIDs twice a week, but I was wrong. Most likely.

 

[fls]

Sham needles. http://www.acupuncturetoday.com/archives2002/apr/04needle.html

How could that have made a difference? The anaesthesiologist and research team couldn't see the acupuncture area, the patient was anaesthetized, and the acupuncturist had nothing to do with the data collection or analysis.

As to n and power, "However, if too many observations are used (or if a test is too powerful with a large
sample size), even a trivial effect will be mistakenly detected as a significant one. Thus,
virtually anything can be proved regardless of actual effects.(from indiana.edu/stat-math), as I'm sure you know.

As I said, there's a happy medium. I doubt that you are suggesting that the next step up from a power of 15% to detect a medium-sized effect is a power of 95%. I am not suggesting anything but an appropriate balance of alpha and beta.

And your advice to replicate small studies is one of the ways one would go about creating the appearance of an effect when one is not present. (See Ioannidis)

Since I can't read the article, did they use a t-test? The t is pretty robust according to Alan Boneau's Monte Carlo studies.
http://www.garfield.library.upenn.edu/classics1986/A1986D597700001.pdf

They did an ANOVA. Robust "unless the heterogeneity of either form or variance is so extreme as to be readily apparent upon inspection of the data..."

Linda

 

[Jeff Corey]

How could that have made a difference? The anaesthesiologist and research team couldn't see the acupuncture area, the patient was anaesthetized, and the acupuncturist had nothing to do with the data collection or analysis...

"Sixteen volunteers were anesthetized by propofol...", from this it isn't clear to me that the subjects were unconscious. "Anesthesia" refers to " Total or partial loss of sensation, especially tactile sensibility."