DC's Improbable Science

Truth, falsehood and evidence: investigations of dubious and dishonest science

DC's Improbable Science header image 2

Homeopathy in Intensive Care

February 15th, 2007 · 4 Comments

Using potassium dichromate to treat patients in intensive care (rather than to clean the glassware)?

No, that isn’t a joke. The respectable journal, Chest, official journal of the American College of Chest Physicians, published an article that purported to show that homeopathic potassium dichromate (i.e. water) was a useful way to treat patients in intensive care. [Frass M, Dielacher C, Linkesch M, et al. Influence of potassium dichromate on tracheal secretions in critically ill patients. Chest 2005; 27:936–941].

The title and abstract don’t mention the word ‘homeopathy’ at all. Potassium dichromate, like all hexavalent chromium compounds, is very toxic, but luckily for the patients there was no potassium dichromate present whatsoever in the treatment (it was a 30C dilution). The editor of Chest didn’t seem to think that there was anything very odd about this, but he did publish a response from me: Treating Critically Ill Patients With Sugar Pills, Chest, 131 , 645, 2007 [Get pdf ].

“It is one thing to tolerate homeopathy as a harmless 19th century eccentricity for its placebo effect in minor self-limiting conditions like colds. It is quite another to have it recommended for seriously ill patients. That is downright dangerous.”

This was accompanied by an unrepentant response from Frass.

The Frass paper has now received some close attention on the Respectful Insolence blog. Someone posting under the name ‘getzal’ has done a nice analysis which shows that the control group must have contained patients who were were more seriously ill than the homeopathically-treated group.

Print Friendly

Tags: Academia · Anti-science · Dangerous advice · homeopathy · science · statistics

4 responses so far ↓

  • 1 badly shaved monkey // Nov 26, 2007 at 10:53

    and Dana Ullman made a big play of this paper here;

    http://commentisfree.guardian.co.uk/denis_maceoin/2007/11/your_ignorance_is_showing.html#comment-947829

    just before the thread got timed-out.

    Fortunately there was time for other posters to skewer his flagrant misrepresentation of that paper. However, there were numerous other loose ends left dangling that were shaping nicely into nooses for him, but the timeclock intervened. I wonder whether he might show up here for more of the same.

    [Returns to knitting by the scaffold waiting for the sound of the tumbrel.]

  • 2 niall // Nov 26, 2007 at 12:35

    Yes, saw the CIF also. Looking at DC’s letter to the journal, and the original author’s response, I was struck by the juxtaposition of the end of their letter:

    “We are always open to tolerant, academic-driven scientific cooperation. Let us make a common effort to reveal the riddle for the sake of millions of patients by supporting the Institute for Homeopathic Research, at Interuniversity College. Financial help is appreciated:
    Account No. 92,178,414, bank code 60,000, IBAN: AT95 6000
    0000 92178414, BIC: OPSKATWW.”

    with the Conflict of Interest statement after their signatures:

    “The authors have reported to the ACCP that no significant
    conflicts of interest exist with any companies/organizations whose
    products or services may be discussed in this article.”

    :-)

  • 3 zathrus // Nov 26, 2007 at 15:07

    It’s a shame that thread turned into a pumpkin when it did.

    One claim that was made by a poster midway through that thread was that MacEoin lectures to medical students on CAM – anyone know the truth on this? Because, if true, it scares the hell out of me.

  • 4 jdc325 // Nov 26, 2007 at 18:14

    I thought monk7b did well in the comments and scooby71 brought up some interesting “competing interests”.

    Re “MacEoin [lecturing] to medical students on CAM” – it’s on his wiki page that he lectures on the ‘sociology and politics of CAM’. So not giving actual (pseudo)medical training. Still scary, though.

You must log in to post a comment.