View Single Post
Old 11th December 2007, 06:25 AM   #79
fls
Penultimate Amazing
 
Join Date: Jan 2005
Posts: 10,226
Originally Posted by JamesGully View Post
Sorry, this meta-analysis says exactly what I said it does. I said that the Cochrane report called the results of the treatment of influenza "promising." I made it clear that its use in the prevention of the flu is not proven, though this has not been its use anyway.
This is what you said:

"Skeptics can no longer say that homeopathic medicines are proven to be effective in treating a specific condition.

If anyone is asked, does Oscillococcinum have an effect on influenza or influenza-like syndrome, one must say YES."

I assumed the first sentence was a typo, otherwise we could all end the conversation in agreement. The meta-analysis does not support your claim that Oscillococcinum has an effect on influenza. It was qualified as "might".

Quote:
In reference to "severe sepsis," you wrote

As for the two groups...both groups had a life-threatening condition in which other studies show that 40-70% of the people die. In 6 months, 50% of those given the placebo died, while only 25% of those given a homeopathic medicine died. How you can say that these significant differences are just what you would expect from NO treatment is a statistical improbability (thus, the impressive p-value). These results not only show an effect, but they show some long-term effects of homeopathic medicines. Thank you. He shoots; he scores.
Badly Shaved Monkey has provided a demonstration that shows statistically significant differences can be expected most of the time if your initial group shows heterogeneity in the absence of any differences in treatment, such as special water. I have also experienced this problem in actual practice (except that I recognized it was a problem I needed to deal with rather than manipulating it to my advantage to claim something that isn't true).

But now you have another problem. Here you are claiming that the effects of homeopathy are long term. That they don't show up in the short term is something to be expected. However, your fibrositis references all depend upon homeopathy having no long term effects, as they are all cross-over studies. If you have any suspicion that the effects carry over from one treatment period to another, you cannot analyze them as though they do not.

So you face a choice. Either cling to your sepsis study as proof of homeopathy's effectiveness or your fibrositis studies, but they cannot both be true.

Quote:
The BMJ study was double-blind, placebo-controlled WITH a cross-over (you cannot do better than that). The second study is worthy of your attention because not only was there greater CLINICAL improvement in the homeopathic treated patients (some of you can call that "subjective," except when improvement in YOUR own symptoms is observed), the researchers also conducted EEG readings while the patients were given their dose or a placebo via olfaction. Yeah...this study will blow your mind...and maybe your nose too...

Scientific Evidence

Researchers in England found that patients with fibromyalgia were (obviously) a varied group with differing symptoms but that there was one homeopathic medicine, more than any other, that seemed to be indicated for a certain sub-set of fibromyalgia patients (Fisher, 1986). This medicine, Rhus toxicodendron (also called Rhus tox) was found to be indicated in 25% of fibromyalgia patients.

The researchers found 30 patients who seemed to fit the symptoms of Rhus tox, and they were given a homeopathic dose of this medicine, 6C (this dose is considered a “low potency,” that is, it is a dose that generally does not have long-term effects). The researchers found that there was a significant degree of improvement in the reduction of pain and tender points and improved sleep when the subjects were taking the homeopathic medicine, as compared to when the subjects were taking a placebo.

The results of this experiment were so significant that the researchers found that there was a 5 in 10,000 chance (P<.005) of this occurring from chance.

This study, therefore, strongly suggests that homeopathic medicines can be effective in reducing the pain and tender spots and in improving sleep in some patients suffering from fibromyalgia.
The researchers performed a faulty analysis. When an appropriate analysis was performed, there were no statistically significant differences found. I am surprised that you don't seem to be aware of this.

Quote:
In addition to this study, a more recent study published in the highly respected journal, Rheumatology (published by the British Society for Rheumatology) also found statistically significant results from homeopathic treatment. Researchers from the University of Arizona in collaboration with homeopaths conducted a double-blind, randomized, placebo-controlled trial with 62 fibromyalgia patients (Bell, Lewis, Brooks, et al., 2004). Patients were randomized to receive an oral daily dose of an individually chosen homeopathic medicine in LM potency (or a placebo). Patients were evaluated at baseline, 2 months, and 4 months.

The study found that 50% of patients given a homeopathic medicine experienced a 25% or greater improvement in tender point pain on examination, as compared to only 15% of those who were given a placebo experienced a similar degree of improvement (P=0.008). After 4 months, the homeopathic patients also rated the “helpfulness of the treatment” significantly greater than did those who were given a placebo (P=0.004). It is therefore not surprising that the study also showed that the average number of remedies recommended by the homeopaths was substantially higher to those in the placebo group as compared with the real treatment group.
If you look at table 2 in this study, you will see that there are no statistically significant differences between the two groups on any of the measures. The researchers elected to adjust the values for some of the differences in baseline measures, and it is the adjusted values that show statistically significant differences. That these two analyses lead to divergent results weakens your ability to draw conclusions from the data. And it invalidates any claims of certainty.

Quote:
One special additional feature of this trial was that the first dose of medicine was given by olfaction (by smell) and that both groups were monitored with EEG. The researchers found that there was a significant and identifiable difference in the EEG readings in patients who were given the real homeopathic medicine as compared to those given the placebo (Bell, Lewis, Schwartz, et al, 2004a).

Each patient had three laboratory sessions, including at baseline, at 3 months, and at 6 months after initial treatment. The researchers found that the active treatment group experienced significant increases in the EEG relative alpha magnitude, while patients given a placebo experienced a decrease in this measurement (P=0.003).

The combined evidence of clinical improvement along with objective physiological response from the homeopathic medicine makes the results of this trial of additional significance.
I don't have access to the full text of this study. Until I see otherwise, I have to be suspicious that this study suffers from the same problems as the other two. And I find this line confusing - "...sniff alpha-1 and alpha-2 increases at 6 months versus baseline [did not correlate with] clinical outcomes in the active group." Are they saying that the EEG findings were unrelated to the clinical outcomes? Wouldn't that weaken the idea that the homeopathic treatment influenced either outcome?

Quote:
REFERENCES:

Bell, IR, Lewis, DA, Brooks, DJ, Schwartz, GE, Leis, SE, Walsh, BT, and Baldwin, DM, Improved Cilnical Status in Fibromyalgia Patients Treated with Individualized Homeopathic Remedies Versus Placebo, Rheumatology, January 20, 2004:1111-7.

Bell IR, Lewis Ii DA, Lewis SE, Schwartz GE, Brooks AJ, Scott A, Baldwin CM. EEG Alpha Sensitization in Individualized Homeopathic Treatment of Fibromyalgia. Int J Neurosci. 2004;114(9):1195-1220.

Peter Fisher, Alison Greenwood, EC Huskisson, et al., “Effect of Homoeopathic Treatment on Fibrositis (Primary Fibromyalgia),” BMJ, 299(August 5, 1989):365-6.

So, Linda, is this drivel too, and are these journals drivel, and is everyone but you and your buddies on this list drivel? I think that the problem is that some people have a problem in seeing patterns.
Humans see pattern wherever they look, even when there is no connection to the elements. I think the problem is that while we recognize that the appearance of a pattern does not necessarily mean that there is a connection, many people have trouble letting that idea go. The "patterns" that you attribute to homeopathy are the same patterns that we expect to see from chance, bias (including cognitive), and a little fraud. There doesn't need to be a true treatment effect from homeopathy to generate the pattern that we see.

Quote:
A 3-pointer and I was fouled on the shot.
Two of these studies were outright negative. One study showed "significant" results that would be expected at least 50% of the time in the absence of treatment effects. That you consider this a "slam dunk" reflects very poorly on the veracity of your claims.

Linda

Last edited by fls; 11th December 2007 at 06:26 AM.
fls is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top