Homeopathic Cough Medicine

Translated to English: If you include the piss-poor quality trials while making desperate excuses for their lack of results, pretend that other well-run trials with poor results would have been "good" if they hadn't stuck to accepted rules, and rapidly shift the goalposts to make a failed trial a barely ceded pass, homeopathy measures up.

How can one get something out of nothing?

:D

M.
 
Burden of proof, Rodney.

If you think she's wrong, get a statistician who disagrees with her.

You won't, though, because, as I've repeatedly said, you're playing to the fence-sitting lurkers, and not to us at all.

If I were arguing that the 83 of 107 homeopathy trials that produced results above placebo (cited in the 1991 British Medical Journal article) conclusively demonstrated that homeopathy is efficacious, the burden would be on me. But I am not arguing that. Rather, I am arguing that homeopathy needs more study. Linda, on the other hand, is arguing that -- notwithstanding the 83 positive homeopathy trials -- homeopathy has been discredited based on a new statistical model that uses a "pre-test probability" subjective estimate for a homeopathic medicine being efficacious. The burden is on her or one of her supporters to demonstrate that this new model is applicable to homeopathy trials.
 
If I were arguing that the 83 of 107 homeopathy trials that produced results above placebo (cited in the 1991 British Medical Journal article) conclusively demonstrated that homeopathy is efficacious, the burden would be on me. But I am not arguing that. Rather, I am arguing that homeopathy needs more study. Linda, on the other hand, is arguing that -- notwithstanding the 83 positive homeopathy trials -- homeopathy has been discredited based on a new statistical model that uses a "pre-test probability" subjective estimate for a homeopathic medicine being efficacious. The burden is on her or one of her supporters to demonstrate that this new model is applicable to homeopathy trials.
Point #1: If homeopathy wants to be treated as conventional medicine, then it must accept conventional analysis, so of course the method applies, unless you are suggesting it doesn't apply to conventional medicine, either. In which case, fls' standing trumps yours, at least for me.

Point #2: You have missed a step. You asked her to do the analysis. She did. You have a problem with the analysis (which really amounts to: you don't understand it). You won't demonstrate a problem with the analysis.

Point #3: You are making a claim, though you are veiling it, intentionally I think. You are claiming her analysis is (a) invalid (b) inapplicable to homeopathy. Demonstrate your claim.

You've demonstrated nothing in this thread, Rodney. Nothing at all. Your questions have been answered. Now you shift goalposts and burden of proof.
 
What's the point? Remember the "evaluating psychics' accuracy" thread? you asked for the opinions of some doctors. When we got them, they didn't agree with you, so you decided they weren't good enough.
But they didn't agree with each other, either. ;) Also, I'm still waiting for Huntsman's doctors to weigh in.

Actually, they did agree with each other, they just answered different parts of the multi-faceted questions you came up with. Questions which we suggested to be split up in different parts to avoid confusion, but which you insisted be asked the way you composed them.

If anyone's interested, it's another perfect example of Rodney constantly moving the goalposts around, giving us new challenges, having those challenges met and then coming up with various excuses for why he decides he can ignore the evidence we just presented.

The current thread is following the exact same pattern.
 
Point #1: If homeopathy wants to be treated as conventional medicine, then it must accept conventional analysis, so of course the method applies, unless you are suggesting it doesn't apply to conventional medicine, either. In which case, fls' standing trumps yours, at least for me.

Point #2: You have missed a step. You asked her to do the analysis. She did. You have a problem with the analysis (which really amounts to: you don't understand it). You won't demonstrate a problem with the analysis.

Point #3: You are making a claim, though you are veiling it, intentionally I think. You are claiming her analysis is (a) invalid (b) inapplicable to homeopathy. Demonstrate your claim.

You've demonstrated nothing in this thread, Rodney. Nothing at all. Your questions have been answered. Now you shift goalposts and burden of proof.
Have you ever taken a statistics course, Garette? If so and you still have your old textbook, open it and see if you can find any mention of Linda's contention that "negative studies don't matter." Also, see if you can find any justification for applying a "pre-test probability" to the results of a controlled study of the efficacy of medicine. According to Linda's model, even if a bias-free study finds that a homeopathic medicine was effective above placebo with a probability of 95%, that probability is reduced to only 0.15%. Does that make sense to you? And if it does, why isn't that same probability applied to non-homeopathic medicines? (You inquired: "If homeopathy wants to be treated as conventional medicine, then it must accept conventional analysis." But Linda's analysis is completely unconventional!)

Try actually following the logic of the arguments, rather than assuming that Linda is the "skeptic" and I am the "woo" with regard to statistical analysis of homeopathy studies.
 
Have you ever taken a statistics course, Garette?
No.

Rodney said:
--snip--
Does that make sense to you?
Two part answer:

1. Only vaguely, as I am not a statistician. As such, I am forced to determine levels of credibility of the presenters, one side being you, the other being fls, exarch, and others. fls' credentials alone would have me leaning her way. Add my experience with you indicating quite clearly how you avoid answering questions, shift goalposts, and pretend arguments that refute your assertions never happened, and it's a slam dunk.

2. You'll recall in a much earlier post how fls talked about context, meaning the robust history of conventional medicine research and the paucity of vigorous homeopathic research. That makes sense to me.

Rodney said:
And if it does, why isn't that same probability applied to non-homeopathic medicines?
See my point number 2 above.

Rodney said:
(You inquired: "If homeopathy wants to be treated as conventional medicine, then it must accept conventional analysis." But Linda's analysis is completely unconventional!)
You are mixing two different meanings of "conventional." I don't know if you're doing it intentionally.

"Conventional" as I used it meaning mainstream medicine which does not accept complementary medicines because of their lack of evidentiary foundation.

"Conventional" as you used it meaning "standard usage, regardless of discipline."

I don't know if her analysis is conventional in the sense of not being usual, standard, or common statistical method. But I do know that she has said it is. While I admit I could be wrong, she has far more credibility than you right now.

Rodney said:
Try actually following the logic of the arguments,
I do. And when the technical aspects get beyond my understanding (as they have here), I choose my experts carefully.

Rodney said:
rather than assuming that Linda is the "skeptic" and I am the "woo" with regard to statistical analysis of homeopathy studies.
I don't assume Linda is a skeptic in regard to this. I assume I'm not being lied to regarding her qualifications, and based on the clarity of her (non-technical) posts, I assume she is not only open minded but has the best interests of her patients at heart and that she would adopt any validated method for treating them.

Likewise, I don't assume you are the woo. I have reached the reasoned conclusion that you will not change your position, regardless of evidence. I have reached the conclusion that you will espouse claims regarding homeopathy and Edgar Cayce and the like regardless of the lack of valid evidence supporting them. Whether you do this because of a legitimate belief on your part, I do not know.
 
No homeopathy is a set of well defined claims.

1. If a substance causes a problem then if you dilute it by say 10x it will cause the body to remove that symptom from itself.
2.The more you dilute it the more effective it is.
I am afraid this is ill-considered. For example, I can make the valid claim that an antibiotic is effective against a particular microbe; but that doesn't mean the antibiotic combats any other particular microbe.

Homeopaths make thousands of claims. For example, they will say that tinselweed 10X will treat the vapors in a woman of mellow nature, and who prefers cold weather. Whereas, a nervous man who likes warm weather can be cured of the vapors by young-girl's tears 5C. (Young-girl's tears is/was a homeopathic treatment!?)

Each of these is a therapeutic claim that requires verification. No single study, or meta-analysis of 100 studies, can prove all of the thousands of claims.

That is why I challenged Rodney to provide one example of a homeopathic remedy that is known to be reliable. That is why he ignores my challenge- he can't cite one example.

As this has never been shown to be true the fundamental processes that it is supposed to work by has nothing backing it up.
This is true- the homeopathic notion has no backing; but bleevers say that's not the same as disproven. Grasping at straws.
 
Last edited:
I agree that Linda has made a major contribution to this thread; however, I find her statistical analysis of homeopathy studies invalid. If you think I'm wrong, see if you can find a statitistician who agrees with her.

Isn't Linda is talking about a Bayesian type analysis, and there are plenty of statisticians who would agree with that.

I'm no doctor, but I assume that conventional medicines have a relatively high prior probability of effectiveness before clinical trial because only those that are promising on the basis of animal studies, test-tube studies, computer modelling etc get to the clinical trial stage.

You can't compare that to testing substances selected at random in case they impart magical properties to water with no prior reliable evidence or logical reason to think they do. The prior probability is completely different in each case.

'Conventional' medical substances if selected and tested with no good prior reason to expect effectiveness would also have an extremely pre-test probability.
 
And yet homeopaths will regularly as clockwork prescribe exactly the same remedy for the same set of symptoms for absolutely anybody, regardless of any other supposed contributing factor. And very often without the much-vaunted "individual taking" they pride themselves on that supposedly sets them above ordinary medicine.

Take a look at most of the homeopathy forums - this is common practice. The procedure goes like this:

1) Patient says "I have symptoms A, B and C."

2) Homeopath selects random potion and says "Buy this potion and take it, and call me again if it doesn't work."

3) Time passes.

4) Patient says "Didn't work. What now?"

5) Go to step 2.

Note this is a closed loop. The only exits are for the patient to get better of their own accord (i.e. natural healing), or they die (of disease or old age, whichever comes first). Or they get wise and go see a REAL doctor who knows what they are about.
 
And yet homeopaths will regularly as clockwork prescribe exactly the same remedy for the same set of symptoms for absolutely anybody, regardless of any other supposed contributing factor. And very often without the much-vaunted "individual taking" they pride themselves on that supposedly sets them above ordinary medicine.

Take a look at most of the homeopathy forums - this is common practice. The procedure goes like this:

1) Patient says "I have symptoms A, B and C."

2) Homeopath selects random potion and says "Buy this potion and take it, and call me again if it doesn't work."

3) Time passes.

4) Patient says "Didn't work. What now?"

5) Go to step 2.

Note this is a closed loop. The only exits are for the patient to get better of their own accord (i.e. natural healing), or they die (of disease or old age, whichever comes first). Or they get wise and go see a REAL doctor who knows what they are about.
I nominated this.
 
Ah, delicious irony.

The suggestion that clinical trials in general and homeopathy trials in particular are subject to dependent probabilities is nonsensical and unbelievable. But the suggestion that water has the ability to cure disease if at some point it came in contact with, but no longer contains any other chemicals, is eminently plausible.

Linda
 
And yet homeopaths will regularly as clockwork prescribe exactly the same remedy for the same set of symptoms for absolutely anybody, regardless of any other supposed contributing factor. And very often without the much-vaunted "individual taking" they pride themselves on that supposedly sets them above ordinary medicine.
And remember that the subject of this thread is OTC homoeopathic cough medicines - homoeopathic nostrums that are intended to treat a specific condition without individualisation, and without any homoeopathic consultation. If the homoeopathic consultation and individualisation are, as is claimed by homoeopaths, essential to homoeopathy, why are they not denouncing OTC homoeopathic medicines as frauds?
 
Isn't Linda is talking about a Bayesian type analysis, and there are plenty of statisticians who would agree with that.

I'm no statistician, but I don't think that this would necessarily belong to Bayesian statistics.

Linda provided a set of initial conditions (# of chemicals; # of ailments) and from there set out a series of steps that explained how she reached her final result. There didn't seem to be anything too speculative about her premises (if anything just a bit arbitrary, as would be expected) or her reasoning, it seems fairly standard to me.

Mind you, it certainly would have to be on the borderline.

Someone please correct me if I'm wrong about any of the above.
 
Ah, delicious irony.

The suggestion that clinical trials in general and homeopathy trials in particular are subject to dependent probabilities is nonsensical and unbelievable. But the suggestion that water has the ability to cure disease if at some point it came in contact with, but no longer contains any other chemicals, is eminently plausible.
Linda
I didn't say that homeopathy was eminently plausible, but either you test it fairly or you don't. It is illogical to conduct a controlled study of a homeopathic medicine, find that it has performed significantly better than placebo, and then downgrade the results by applying a low pre-probability percentage to them. If you are convinced that homeopathy cannot work, testing is irrelevant.
 
This model was developed as a way to calculate the PPV of a research study. If you want to know more, you can read the paper, but it is quite technical. I’m willing to try to explain it, but it’s not light reading. <g>

I thought I'd take a look at this, but the link isn't working for me. If you would provide the specifics of the paper, I can probably get hold of it. Since I'm a professional statistician and the opinion of a professional statistician was requested, I thought I might try to contribute an opinion regarding the analysis.
 
I'm no statistician, but I don't think that this would necessarily belong to Bayesian statistics.

Linda provided a set of initial conditions (# of chemicals; # of ailments) and from there set out a series of steps that explained how she reached her final result. There didn't seem to be anything too speculative about her premises (if anything just a bit arbitrary, as would be expected) or her reasoning, it seems fairly standard to me.

Mind you, it certainly would have to be on the borderline.

Someone please correct me if I'm wrong about any of the above.

The first link she provided in post 475 regarding the basic approach she used is a Bayesian analysis.
 
Rodney cites a 1991 analysis that does not support evidence for efficacy as evidence that homeopathy may work. How convoluted can argument get? It really sounds like the "thinking" of a chiropractor. Here is an analysis of some other reports that do not support efficacy:
http://www.acsh.org/healthissues/newsID.632/healthissue_detail.asp

I am still waiting for a study that supports efficacy of a homeopathic product (that is certified not to have been adulterated with active amounts of drugs).

Clearly, Rodney doesn't have that evidence or we would have seen it.
 
I thought I'd take a look at this, but the link isn't working for me. If you would provide the specifics of the paper, I can probably get hold of it. Since I'm a professional statistician and the opinion of a professional statistician was requested, I thought I might try to contribute an opinion regarding the analysis.

Thanks Beth.

The paper is "Why Most Published Research Findings Are False" by John Ioannidis, at PLoS Medicine.
http://medicine.plosjournals.org/perlserv/?request=get-document&doi=10.1371/journal.pmed.0020124

If that doesn't work....browse for the August 2005 issue of PLoS Medicine at www.plos.org.

I'm happy for the chance to elaborate on what I was thinking (if necessary/desirable).

Linda
 
It is Bayesian. I didn't mean to confuse anyone.

Linda
 

ISF - Join now!

Every member here is approved by hand. No bots, no spam, just people who care about evidence and honest debate.

Membership is free!

Create your free account

Back
Top Bottom