Apart from the very real dangers in diverting people from possibly effective medicine, the obscenity is in the dishonest claims and profiteering. Nelson and Boiron are basically conducting a legalised scam. They produce sugar pills and water and sell them as if they are something else containing medicinal properties. Prison sentences have been handed down for less.

If one insists that doctors have the option of dishing out a placebo when they wish, there is nothing to stop them. There are forms of words to do it without calling it homeopathy or anything at all for that matter, and whereby the patient isn’t being fleeced. The effect is in the ritual not in the sugar or water. And the cost of sugar pills and water should reflect the cost of sugar pills and water, not some arbitrary value of recovery from which the likes of Nelson and Boiron, and a bunch of dispensing charlatans, can profit.

Overall, I find a lot to agree with you on and good luck in the campaign.
But…

“I am sure homeopathy is a placebo (when it does anything) but I’m willing to admit that I don’t think we’ve quite proved that yet.”

Why are you sure when you don’t think it’s been proved yet. In other words, what is the basis of your certainty. That doesn’t sound like a scientific approach to the evidence.

“The harm is, of course, if the cost is higher than the benefit.”

That is the minor harm. In Africa homoeopaths are killing people with homoeopathic alternatives to vaccines.

“homeopathy resists scientific disproof in the same way that the existence of God and orbiting teapots do”

The theistic god is a false hypothesis and the deistic god an unnecessary hypothesis. The orbiting teapot is a just an unnecessary hypothesis and no one actually takes it seriously. So, homoeopathy is more like the deistic god hypothesis than the orbiting teapot hypothesis.

“it’s dangerous to say that a meta-analysis proving that something is no better than placebo is a conclusive scientific proof.”

I agree. Like BSM said. The there is a subset of the placebo effect where it does work. That is in the treatment of psychosomatic illness, depression, and pain.

“Richard Hughes; Vilma Bharatan”

And Edzard Ernst as the best example. A professor of alternative medicine and practicing homoeopath who set out to provide evidence for altmed and found that there was none.

“we waste a hell of a lot of energy on this when, as far as I can tell, it doesn’t matter that much. Most people are sensible…”

But some aren’t and some are exploited. Some die. And I guess Africans don’t count in a UK election.

The key word here is slim. But even the slimmest of evidence makes this scientifically tantalising.” [from the book's website]

This is where you go horribly, horribly wrong IMHO, Dr Brooks. The idea that a few anomalous laboratory results – which admittedly could appear to a poor scientist to provide support for (part of) the homeopaths’ absurd fantasy – constitute something “scientifically tantalising” is breathtakingly eccentric.

“she is a homeopath but also a good scientist … her evidence is leading her towards the idea that the whole dilution thing is bunkum.”

Righhht… In the same way that someone involved in one of the computerised searches for prime numbers saying that their evidence was leading them towards the idea that there is no largest prime would be a “good mathematician” perhaps?

“My last point is this: we waste a hell of a lot of energy on this when, as far as I can tell, it doesn’t matter that much.”

Well I think you’ve unconvincingly downplayed the harms done by high street homeopathy etc. and entirely missed some of the other homeopathy related issues we waste our energy on. For example, I wonder if your mind would be “open” enough to accommodate the prospect of a son or daughter changing their mind about following in your footsteps and going off to UoW to study for a BSc in homeopathy instead? ]:->

Cf. “GROLIES”

http://news.bbc.co.uk/1/hi/3159813.stm

On a more sober note, could you give us citations/links to those placebo meta-analyses? 

Other readers may know I’m a vet and even in my profession I meet a lot of, “Yes, well, maybe the placebo effect is useful, so why be rude to homeopaths.”

I think that citing the placebo effect has become the easy get-out for the mild sceptic of CAM who doesn’t want to think through the logical consequences of the use of unmedicine in animals and people.

I know Ben Goldacre has played up the significance of the placebo effect, but I do wonder whether that is a matter of perspective. Most of the examples of a ‘real’ placebo effect seem to derive from work on conditions with a huge and obvious psychosomatic component or conditions that are completely.  psychological/psychiatric. Do those meta-analyses preclude the existence of a clinically meaningful placebo effect even in those psychosomatic conditions?

Research into the placebo effect is confounded by a tricky problem- with psychosomatic/psychological conditions, a willingness to lie to your therapist/doctor about how you feel is very hard to dissect from an actual internal experience of feeling better. It’s easier when the subjective report can be tested against an objective measure. An interesting and important example from my field is the analysis of outcomes of cruciate ligament surgery. There are several competing techniques and the literature suggests they all result in a sound patient 6mths post-op in 60-80% of cases as assessed by owners and vets. It turns out that if the gait is analysed objectively (weight transmitted to a pressure plate in the floor when the dog walks over it) only 15% of patients return to normal function. In other words 50-70% of the reported benefit of the surgery is at best over-estimated and at worst, just plain wrong. With human psychosomatic/psychological conditions, what we lack are good objective measures that would reveal the degree to which patients and therapists are over-estimating their own improvement.

And good luck against Tredinnick. But at least Tredinnick is an explicitly obvious loon. I think more worrying are the large fraction of those 70 MPs who are clearly incapable of accepting objective scientific fact but actually wield a larger aggregate amount of leverage in favour of CAM behind the political scenes.

We should develop a metric for this similar to the Canard. Let’s call it a Tredinnick. The MP for Bosworth stands as the reference, 1 Tredinnick. If that list contains 5 outright loons, that is 5T. But the other 65 may be on average 0.2T so their effective lobbying weight is 13T and they, en masse, are the more dangerous. 

By the way, 1,000T, or 1 kiloTredinnick can also be called 1 Ullman.  

I liked the discussion on the placebo effect. You mentioned the ethical problem without going very deeply into it. It is something that I’ve had churning in the back of my mind ever since I became a reader of Bad Science blogs. Perhaps because I’m older than the current generation of medics I don’t have any great objection to doctors prescribing what are effectively placebos for a range of conditions for which there aren’t any good evidence-based treatments. With my scientific hat on it angers me to read the nonsense that alternative medicine produces. With other hats on I agree with you that for the most part there is no great problem if the middle classes want to feel edgy. That section was wonderful btw – how interesting it would be to see research on how social class and CAM interact.

But enough on homeopathy. David Tredinnick believes in many and varied pieces of woo and (a greater failing) that public policy should not be guided by science where it conflicts with that woo. Go get him.