Homeopathy works...

[Mojo]

...as well as placebo!

Brien et al. Homeopathy has clinical benefits in rheumatoid arthritis patients that are attributable to the consultation process but not the homeopathic remedy: a randomized controlled clinical trial. Rheumatology, published online November 13, 2010.

Quote:

Conclusion. Homeopathic consultations but not homeopathic remedies are associated with clinically relevant benefits for patients with active but relatively stable RA.

There's an accompanying editorial by Edzard Ernst.

Quote:

Yet Brien et al. argue that the placebo effect of the consultation with a homeopath is specific to homeopathy and ‘dependent on the ritual of the collaborative and highly individualized consultation necessary to identify a homeopathic remedy and the associated symbolic meaning response for that patient’. Proponents of homeopathy tend to defend homeopathy in that way. More critical minds might, however, see things differently. They would doubt whether ineffective therapies can be vindicated through the non-specific effects they generate. They would also warn against the double standard this would establish. A useless surgical operation, for instance, does not become useful and recommendable because it generates a host of non-specific effects which are typical of that setting.

Their argument would also seem to be contradicted by a recent systematic review (Nuhn T, Lüdtke R, Geraedts M. Placebo effect sizes in homeopathic compared to conventional drugs - a systematic review of randomised controlled trials. Homeopathy. 2010 Jan;99(1):76-82) which concluded that "placebo effects in RCTs on classical homeopathy did not appear to be larger than placebo effects in conventional medicine."

[Red3]

So homeopathy still doesn't work then? I'm shocked.

[quarky]

It is a placebo.

[Mojo]

Covered by the Torygraph here, by the way. They give the last word to the super-special placebo effect:

Quote:

Dr Sarah Brien, the study’s lead author, said that while previous research had suggested homeopathy could help patients with rheumatoid arthritis, the study provided the first scientific evidence to show such benefits were “specifically due to its unique consultation process”.

But by no means the first scientific evidence to show that it ain't the remedies.

[Mojo]

It'll be interesting to see if the homoeopaths try to spin this, or just ignore it.

[Emet]

Editorial link did not work for me. This appears to work:

http://rheumatology.oxfordjournals.o...gy.keq265.full

[Red3]

Originally Posted by Mojo

It'll be interesting to see if the homoeopaths try to spin this, or just ignore it.

I'm voting for spin.

[Mojo]

Originally Posted by Emet

Editorial link did not work for me. This appears to work:

http://rheumatology.oxfordjournals.o...gy.keq265.full

Thanks - I'd messed up the formatting. Fixed now.

[saizai]

AFAICT the takeaway is just: regular doctors should get better at their bedside manner, interview skills, etc - because it's not just about getting info, the interaction itself has beneficial clinical effect if you're nice.

[quarky]

Anyone have any links to a double blind with placebo against placebo? It would be interesting, and has probably been done. If it was anything like a fair coin toss, one could expect placebo A, or placebo B to have a slight edge.
The kind of edge that gets people elected. 51/49

[Rolfe]

Originally Posted by saizai

AFAICT the takeaway is just: regular doctors should get better at their bedside manner, interview skills, etc - because it's not just about getting info, the interaction itself has beneficial clinical effect if you're nice.

It's not that simple. Homoeopaths regularly and systematically lie to their "patients". They tell them that they can cure them, that they'll work together to find the right remedy, and that the remedy will provide "a deep and lasting cure".

Being lied to often makes people feel better in the short and even the medium term. Unfortunately, doctors are not allowed to lie to their patients in this way.

Rolfe.

[Mojo]

Originally Posted by saizai

AFAICT the takeaway is just: regular doctors should get better at their bedside manner, interview skills, etc - because it's not just about getting info, the interaction itself has beneficial clinical effect if you're nice.

My "regular doctor" has a splendid bedside manner. I usually feel better while he's still talking to me.

[zeno2712]

Originally Posted by Red3

So homeopathy still doesn't work then? I'm shocked.

Phew! Just think of all those text books that won't have to re-written. And the Noble prizes not returned...

[paiute]

Originally Posted by Mojo

It'll be interesting to see if the homoeopaths try to spin this, or just ignore it.

"Homeopathy has clinical benefits in rheumatoid arthritis patients...." according to new scientific study published in peer-reviewed journal!!

[Paul C. Anagnostopoulos]

Originally Posted by Rolfe

Being lied to often makes people feel better in the short and even the medium term. Unfortunately, doctors are not allowed to lie to their patients in this way.

I think we should let doctors lie. And pharmaceutical companies, too.

Just sayin'.

~~ Paul

[Mojo]

Originally Posted by Paul C. Anagnostopoulos

I think we should let doctors lie. And pharmaceutical companies, too.

And sleeping dogs?

[Mojo]

Originally Posted by Red3

I'm voting for spin.

Just noticed this. Note the comments, e.g.:

Quote:

What the study shows is that HOMEOPATHY WORKS. ie: Patients with arthritis who choose to visit a homeopathic doctor, have a homeopathic consultation and take a homeopathic medicine derive benefit from this.

[Delphinium]

I'm going for spin too, they'll ignore it was the consultation process and attribute the improvement to homeopathy.

Doctors generally have good enough bedside manner and practice patient centred care, but they just can't beat someone that has the time to fluff around for an hour or more.

[fls]

I realize we're just joking around, but can I put in a request that we stop perpetuating the idea that homeopaths have better beside manner or that their patients are more satisfied with the therapeutic encounter than they would be from a doctor? We've talked about this before, and as far as I can tell, the evidence suggests otherwise. This is their last remaining justification for their presence, so why concede to it?

Linda

[Deetee]

Linda, perhaps you could help.

I noticed that this study gave incidence rates for adverse events, those for homeopathy arms totting up to be more than those in the placebo arms.
- Table 7.
http://rheumatology.oxfordjournals.o...gy.keq234.full

The authors say the differences are not statistically significant, but that may be because they only compare individual subgroups.

If one just counts those on placebo as compared to those on homeopathic remedy, you can see that 32 patients on placebo experienced 95 events (2.96 per patient) but 45 patients on a homeopathy remedy experienced 187 events (4.15 events per patient).

Now I think that may be a statistically significant difference, but am unsure what stats test to do to look at it. (the one I tried said p=0.03)

Even if a proper test is not quite statistically significant, it does rather point against the homeopaths claims that their remedies are completely harmless.

If you have a study which not only demonstrates the remedy is no better than placebo, but is also more harmful than placebo what possible justification can there be in making people take it?

Perhaps a whole new era of homeopathic management is imminent, where patients go through the consultation procedure, and get given a plastic yellow bath duck at the end of it, rather than some magic pills. This would be so much cheaper, and the NHS might even reconsider the "cost-benefit" of the whole idiocy. It would also chop Boiron's and Nelson's and Helios's profits in one fell swoop.

[Mojo]

Also being discussed on Science-Based Medicine:

http://www.sciencebasedmedicine.org/?p=8269

[Mojo]

Originally Posted by fls

I realize we're just joking around, but can I put in a request that we stop perpetuating the idea that homeopaths have better beside manner or that their patients are more satisfied with the therapeutic encounter than they would be from a doctor? We've talked about this before, and as far as I can tell, the evidence suggests otherwise. This is their last remaining justification for their presence, so why concede to it?

See the Homeopathy paper referred to in the O/P, which seems to confirm your point (although that may not have been the intention behind it):

Quote:

CONCLUSIONS: Placebo effects in RCTs on classical homeopathy did not appear to be larger than placebo effects in conventional medicine.

[Badly Shaved Monkey]

Rather than post the same thing in two places, I just posted this at SBM. The basic point being that I suspect any effect from consultation via the Hawthorne Effect is almost by definition likely to be temporary, almost an artefact of the study itself rather than something that is exploitable for long-term gain.

[Fnord]

Homeopathic elixirs do work better than a placebo (e.g., "Sugar Pill"), but only when used to treat dehydration.

[Belz...]

Homeopathy works:

The less you do it, the better off you are. Dilution principle.

[fls]

Originally Posted by Deetee

Linda, perhaps you could help.

I suspect not.

Quote:

I noticed that this study gave incidence rates for adverse events, those for homeopathy arms totting up to be more than those in the placebo arms.
- Table 7.
http://rheumatology.oxfordjournals.o...gy.keq234.full

The authors say the differences are not statistically significant, but that may be because they only compare individual subgroups.

If one just counts those on placebo as compared to those on homeopathic remedy, you can see that 32 patients on placebo experienced 95 events (2.96 per patient) but 45 patients on a homeopathy remedy experienced 187 events (4.15 events per patient).

Now I think that may be a statistically significant difference, but am unsure what stats test to do to look at it. (the one I tried said p=0.03)

I'm not sure what to do with the limited information we are given. Maybe you could treat them as relative rates. The comparison you made was not one of the comparisons the authors made. When looking at the effect of homeopathic remedies they compare groups 2 and 4 vs. 3 and 5, which gives a less marked difference (3.97 events per patient instead of 4.2).

Quote:

Even if a proper test is not quite statistically significant, it does rather point against the homeopaths claims that their remedies are completely harmless.

If you have a study which not only demonstrates the remedy is no better than placebo, but is also more harmful than placebo what possible justification can there be in making people take it?

I suspect that this result, like the reported 'positive' results, merely reflect what happens when you take advantage of post-hoc data dredging.

What is interesting is that this study was negative for an effect from the homeopathic consultation, yet it is being reported as though it were positive.

Quote:

Perhaps a whole new era of homeopathic management is imminent, where patients go through the consultation procedure, and get given a plastic yellow bath duck at the end of it, rather than some magic pills. This would be so much cheaper, and the NHS might even reconsider the "cost-benefit" of the whole idiocy. It would also chop Boiron's and Nelson's and Helios's profits in one fell swoop.

Rubber ducky, you're the one...
You make bath time lots of fun...



Linda

[!Kaggen]

The question I have is why are humans inherently predisposed to respond to a placebo?
I suppose one could argue that there is a survival value in an action that convinces ourselves we are doing something which might have a positive outcome.
Is there any evidence though in other animals that they might entertain counter-factuals and that this has positive survival value?

[Mojo]

Originally Posted by !Kaggen

The question I have is why are humans inherently predisposed to respond to a placebo?
I suppose one could argue that there is a survival value in an action that convinces ourselves we are doing something which might have a positive outcome.

Or possibly it is a carry-over from "Mummy will kiss it better" - babies cry when they need attention, but if they carried on crying when given attention there would be less incentive for the parents to provide the attention, so babies that cry, but stop when comforted are more likely to receive the attention they need than babies that just cry.

[Kuko 4000]

How was the benefit of the patients measured, self reporting of pain relief or something more specific?

[!Kaggen]

Originally Posted by Mojo

Or possibly it is a carry-over from "Mummy will kiss it better" - babies cry when they need attention, but if they carried on crying when given attention there would be less incentive for the parents to provide the attention, so babies that cry, but stop when comforted are more likely to receive the attention they need than babies that just cry.

In my experience that does not apply during suicide hour

[Deetee]

Originally Posted by fls

I suspect not.

I'm not sure what to do with the limited information we are given. Maybe you could treat them as relative rates. The comparison you made was not one of the comparisons the authors made. When looking at the effect of homeopathic remedies they compare groups 2 and 4 vs. 3 and 5, which gives a less marked difference (3.97 events per patient instead of 4.2).

I suspect that this result, like the reported 'positive' results, merely reflect what happens when you take advantage of post-hoc data dredging.

What is interesting is that this study was negative for an effect from the homeopathic consultation, yet it is being reported as though it were positive.


Rubber ducky, you're the one...
You make bath time lots of fun...



Linda

Ta.

You'll find a little fella who's cute and yella,
and chubby....

Rubber ducky I'm awfully fond of you!

[Deetee]

Originally Posted by Kuko 4000

How was the benefit of the patients measured, self reporting of pain relief or something more specific?

The paper lists several primary and secondary outcome measures under the section called "outcome assessment". I am not very familiar with them, but they appear to be well validated methods in standard use, so at least the authors didn't invent their own assessment score.

[Deetee]

Here is a good review by David Colquhoun
http://www.dcscience.net/?p=3695

[Kuko 4000]

Originally Posted by Deetee

The paper lists several primary and secondary outcome measures under the section called "outcome assessment". I am not very familiar with them, but they appear to be well validated methods in standard use, so at least the authors didn't invent their own assessment score.

What I'm trying to pin down is how big of an effect (other than self reports of feeling better in the short term) only talking, listening and believing something can possibly have to a patient that suffers from rheumatoid arthritis or any other condition

[Red3]

Originally Posted by Mojo

Just noticed this. Note the comments, e.g.:

Typical...

[Kid Eager]

I went to add a response to the comments on Pulse, but couldn't write anything that came across as civil. All drafts were variant of "how fanarkling stupid do you have to be to misread the report that badly?", so I didn't comment at all...

KE