Jeremy Sherr is a prominent homeopath who claims to treat AIDS patients in Tanzania with homeopathy, based on the usual poor quality anecdotal evidence and wishful thinking. He recently caused a stir in the badscience blogosphere, with a blog entry in which he mused about conducting a trial of homeopathy in AIDS patients. The mooted trial is transparently unethical, as pointed out by a number of sceptical bloggers (notably in the comments on Sherr's blog, and at Gimpy's blog, here and here, and The Lay Scientist, here and here), and as we'll see below. Now, Sherr likes to edit his blogposts and delete comments, but let's have a look at what Sherr had to say about the trial.
I am happy to go for a simple trial initially, treating AIDS patients who are not taking ARVs. There is no shortage of patinets who, although they have been offered ARVs, have chosen not to take them, usually because of the serious and debilitating side effects. There are plenty of statistics on ARV treatment and patients with no treatment at all that we can compare to.
Why is this unethical? I would say for three reasons. Firstly, it is a general principle of medical ethics that patients in a clinical trial should not be denied proven treatments for their condition. Clearly, in this trial AIDS patients would be denied ARVs. Sherr seems to think that this is OK because his patients have decided not to take ARVs themselves, but this is, I think, irrelevant. You would still be running a trial in which the subjects are not receiving the best possible standard of care. The issue of informed consent is again critical here; patients would need to be informed that not taking ARVs could be severely damaging to their health.
Secondly, the trial as mooted will not provide any useful information, because there is no control group. Whatever happens in the trial, it is impossible to say that it happened because of homeopathy, rather than sources of bias in the trial design. Since the trial could not provide any usable information, it would be unethical.
Thirdly, I would say we have enough evidence and knowledge about homeopathy to say that it is not going to cure AIDS. Given that there is no likelihood of a true positve result, it is unethical to involve patients in a clinical trial. Informed consent comes into play again here: patients in the trial ought to be told that the current evidence shows that there's essentially no chance of homeopathy having any beneficial effects beyond placebo.
Sherr says, in the recent blog post in which he calls the waaaaambulance over the criticism he has received, that "Any research I may undertake will be subject to rigorous ethical review of the highest standard". Hopefully that will in fact happen, in which case the mooted trial will surely not go ahead. What is disturbing is that Sherr has stated in the past, referring to research protocols and ethics review, that "You have to find willing partners and get a protocol through an ethics committee, and you need to talk their language. I hope it will work but if not, I will just go and do it on a small scale myself - I am determined to do that". This is the most telling comment, I think: it makes it clear that Sherr is not really interested in medical ethics, except as a hoop he must reluctantly jump through in order to experiment on terminally ill patients. And if he can't get ethics approval, he'll just do it anyway.
So much for Sherr. Disturbingly, however, he seems not to be anomalous in CAM circles in his total lack of any sense of ethics. A review of a book "Complementary and Alternative Medicine: Ethics, the Patient, and the Physician” has just been posted on the Science Based Medicine blog. The reviewer comments that "We do not read a word about how to approach a patient who has suffered damage due to CAM, or how to approach those who have stopped their regular treatment" [emphasis mine]. One would have thought that this would be a key issue for any book purporting to address ethics in CAM.
As one of the commenters to the Science Based Medicine piece astutely points out, CAM is a "deprofessionalization phenomenon". Researchers in the field of CAM seem to have no idea about research ethics, and no idea about the linked issue of how to conduct good research. If a text on ethics in CAM is so careless of these important ethical questions, how can we expect CAM practitioners to be any more careful?
UPDATE: There is now an excellent and comprehensive post on the Sherr saga at Respectful Insolence...