Fun with a local homeopath

Note: Prinsloo has edited his website in light of our criticisms, but the version of his site that I responded to is still available on Google cache. 

A Pretoria-based homeopath, one Dr. JP Prinsloo, has taken on some local skeptics, including Owen and Angela. I'll have more to say about him in the next while, but for the moment I want to do three things: point to Owen's superb (and damn funny) response, address one of Prinsloo's arguments and demonstrate he misinterprets the medical literature on homeopathy.

In a section of his website "Answering the Skeptics", Prinsloo makes the follow argument:

Let me begin this page by stating quite emphatically that;

It is against my principles to debate the validity and efficacy of Homeopathy with ignorants (sic).

On this page, reference to the word ignorant (sic) shall mean: Any so-called scientist or "expert" that expresses him/herself on the subject of Homeopathy, it's validity or efficacy, but who -

* Is not a qualified Homeopath;
* Has not studied Homeopathy to the extent that a Homeopath does;
* Has not conducted extensive research on Homeopathy in accordance with the scientific principles of Homeopathy under the supervision of a qualified Homeopath;
* Does not possess sufficient experience in the practical application of Homeopathy in a clinical setting;
* Who is not registered as a Homeopathic Practitioner in South Africa and / or does not meet the requirements for such registration;
* Who is not an expert on applied Homeopathy.(*)

With respect to Homeopathy, that is an ignorant (sic) in my opinion and someone not worthy of my time.

(*) Howard Stephen Berg, The World's Fastest Reader, defines an expert as "someone who has read at least 25 books on a particular subject".

This is a really bad argument. But first, even if we accept these absurd requirements, there is a person who, as a former homeopath, fulfills these criteria and is nevertheless a prominent and respected critic thereof: Edzard Ernst. The key point, though, is that people self-select into homeopathy, so saying only homeopaths are qualified to say anything about it is a transparent attempt to shield it from criticism. Are only astrologers possibly qualified to say anything about astrology? Shall we dismiss all criticisms of parapsychology unless it comes from a qualified parapsychologist? Am I an ignorant (sic) for dismissing the flat earth theory despite not having read 25 books about it? Of course not; doing so would unnecessarily cede the field to the woos. Prinsloo simply misunderstands how and when to defer to experts. (A topic I'm currently writing a lengthy post about, by the way). Furthermore, the most relevant question about homeopathy is: does it work? Do large well-designed double-blind placebo-controlled trails demonstrate that it has a statistically significant clinically significant effect? That is, when you take care not to fool yourself, does homeopathy work? (Hint: the answer is no). And, as Simon also points out in a comment to Owen's post, the most relevant expertise in answering that question is in research methodology and medical statistics. Is Prinsloo a qualified medical statistician? Has he read 25 books on medical research methods and statistics?

Prinsloo also manifestly misunderstands the medical literature. (Alternatively, he's a lair -- but that would be uncharitable. Keep Hanlon's Razor always in mind). In "Homeopathy in Perspective" (based on a journal article of Prinsloo's apparently), he states:

A state of the art meta analysis reviewed 186 studies, 89 of which fit pre-defined criteria, showed that patients taking homeopathic medicines were 2.45 times more likely to experience a positive therapeutic effect than placebo.(19)

Sounds promising. So let's follow the reference, shall we? "Are the clinical effects of homoeopathy placebo effects?" (The Lancet, 2005). Prinsloo has a slight problem: this study simply doesn't conclude what he says it does. Here is an excerpt from the Discussion section:

We assumed that the effects observed in placebo-controlled trials of homoeopathy could be explained by a combination of methodological deficiencies and biased reporting. Conversely, we postulated that the same biases could not explain the effects observed in comparable placebo-controlled trials of conventional medicine. Our results confirm these hypotheses: when analyses were restricted to large trials of higher quality there was no convincing evidence that homoeopathy was superior to placebo, whereas for conventional medicine an important effect remained. Our results thus provide support for the hypothesis that the clinical effects of homoeopathy, but not those of conventional medicine, are unspecific placebo or context effects.

Prinsloo continues:

Another meta-analysis reviewed 107 studies of homeopathic medicines, 81 of which (77%) showed positive effect. Of the best 22 studies, 15 showed efficacy. The researchers concluded: "The evidence presented in this review would probably be sufficient for establishing homeopathy as a regular treatment for certain indications." Further, "The amount of positive evidence even among the best studies came as a surprise to us." (20) 

And what is this meta-analysis (sic)? "Clinical Trials of Homeopathy" (BMJ, 1991). Prinsloo does report accurately on the details, but then conveniently ignores the authors' conclusion: "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." (It continues to say that there is a legitimate case for further research). Nor does Prinsloo mention that this study (which was a systematic review, not a meta-analysis) has subsequently been rubbished. The positive result in this study, it seems fair to conclude, was due to inappropriate weightings of trail quality (the exclusion of peer-review, for one) and biased selection. Moreover, subsequent better designed systematic reviews (like this one by Ernst) have concluded "there was no homeopathic remedy that was demonstrated to yield clinical effects that are convincingly different from placebo".

More on Prinsloo later...

-----------
Shang, A., Huwiler-Müntener, K., Nartey, L., Jüni, P., Dörig, S., Sterne, J., Pewsner, D., & Egger, M. (2005). Are the clinical effects of homoeopathy placebo effects? Comparative study of placebo-controlled trials of homoeopathy and allopathy The Lancet, 366 (9487), 726-732 DOI: 10.1016/S0140-6736(05)67177-2