Badchemist’s Blog » Homeopathic Education Services

Homeopathic Education Services – Part 1

NB: I started writing this months ago and just found it languishing in my drafts. I was planning on going through all of the Dana Ullman’s articles but I’ll see how it goes.

I found little gem earlier today. As these articles are supposed to ‘educate’ I feel it’s only fair for me to go through them and help educate people about homeopathy.

The first, A Condensed History of Homeopathy, I’ll skip – the history of homeopathy is dull. So straight on to, A Homeopathic Perspective on Health and Healing in the 21st Century. The first oddity in this article is this:

Each new technology, however, brings with it new problems and does not necessarily solve old ones. For instance, despite the many advances in medicl [sic] testing, most medical tests are accurate only 90% of the time. Thus, if a physician recommends a battery of 20 medical tests, only 36% of the patients would receive accurate results

I don’t now how they’ve calculated these figures but using the equation for a binomial distribution:

$P_p (N | n) = \frac{N!}{n!(N-n)!}p^n(1-p)^{N-n}$
the value would actually be nearer 12%.

Binomial distribution for number of accurate tests

The accuracy of this value isn’t what annoys me though, its the use of the value to suggest that conventional tests and, by implication, medicine is very poor indeed. From the values used to produce the chart above 99.7% of people would have 14 or more tests return accurate results. From these it should be possible to perform the diagnosis and if there is some ambiguity a test may be rerun – the chance of an incorrect result twice being only 1%.

Later in the article we get treated to the well known canard that homeopathy has a good evidence base.

New research has been published in highly respected scientific journals on homeopathy which shows that the small doses of medicines that homeopaths use definitely have action,(6) and a meta-analysis of clinical studies which was published in the British Medical Journal found the medicines to be particularly effective in treating allergies, arthritic conditions, migraine headaches, common infections, and rheumatoid arthritis.(7)

Reference 6 is for Dana Ullman’s book which I wouldn’t waste my money on. Reference 7 is to this paper in the BMJ in 1991. The conclusions read thus:

At the moment the evidence of clinical trials is positive but not sufficient to draw definitive conclusions because most trials are of low methodological quality and because of the unknown role of publication bias. This indicates that there is a legitimate case for further evaluation of homoeopathy, but only by means of well performed trials.

Considering there are far more recent and stricter meta-analyses that have all found homeopathy to be no better than placebo (see this article by Ben Goldacre for more detail).

Dana then falls back to that other great piece of evidence for homeopathy