The difference between science and magic

Posted on November 21st, 2010 in Culture, Science, Skepticism, storytelling | 1 Comment »

Clarke’s Third Law: Any sufficiently advanced technology is indistinguishable from magic.
– Arthur C Clarke

Harry Potter bursts back on to our screens again this week, with the final book split into two no-doubt lumberingly ponderous full-length motion picture “events”. But, we must take solace in the fact that a) it will soon all be over and b) remember that magic has been a huge part of the fabric of narrative since stories began and try not to be too grumpy about the success of this “franchise”. If the appeal of stories is their ability to take us to places and show us things which ordinary life denies us, then it’s easy to see why magic should play such a strong part in especially early stories. Most fairy stories have a magical element – in fact the word “fairy” implies magic.

As we grow up, we leave such things behind, for the most part. The broad appeal of those Harry Potter books and films (and the amusing existence of editions of Harry Potter novels with “adult” covers and a £2.00 price premium) is a notable exception, but especially among male readers, although fairy stories are left behind, fantasy and especially science fiction stories remain popular.

I have observed before that the key feature of stories, the primary quality which distinguishes story from not-a-story is cause and effect. Without cause-and-effect, all you have is a succession of images. (Ironically, if you present such a succession of images to an audience, they are apt to invent the missing cause-and-effect, justifying the action they see in terms of A causing B, such is the storytelling hard-wiring in the human brain.)

Even in stories which don’t present themselves as fantastical, we can often see cause-and-effect being applied in a very limited way. Consider the phenomenon, often noted in pulp, pop or escapist fare, of “goodat”. If characters suddenly announce some convenient talent, skill, expertise, relationship or ability hitherto unmentioned and unsuspected, the audience is likely to feel cheated. “Oh, didn’t I mention I can speak Japanese / crack safes / recite pi to 150 decimal places / fly unaided / control birds with the power of my mind?” It doesn’t matter if these powers are magical, merely require talent and/or practice or are borderline. Conversely, however, if it is set-up that the character in question is “goodat” languages / escapology / mathematics / psychokinesis / animal sympathy, then the audience will probably accept almost any manifestation of this ability, no matter how fantastic. The archetypal version of this would be – as I think Eddie Izzard noted – the ability of someone who is goodat computers to hack into any system in a matter of minutes, just by waggling their fingers lightly over the keyboard and announcing “I’m in”.

So far, so fair enough. Storytellers take short cuts but even bad ones stop short of outright cheating. But there’s a difference between someone, for example, using their total mastery of Spanish to eventually make communication possible with a person who speaks only Catalan – and on the other hand, somebody uploading a Macintosh virus to an alien computer. I suggest that the latter is magic and that the distinction lies in the cause and effect.

Let’s pause for a moment and consider three ways in which a fictional person may be made to disappear before our eyes – Star Trek transporters, Star Trek phasers and Harry Potter vanishings. In Star Trek as you may know, a device known as a transporter is able to whisk people from one location to another. In effect, this device disintegrates the body at one end, transmits only the pattern (transmitting data is much quicker and easier than moving mass I suppose) to the other end, where it is reassembled. The practical difficulties of achieving this are not to be underestimated. A human being is composed of around 10^25 atoms, the exact location and type and state of which all have to be recorded. That’s a heckuva big file. This data is then transmitted to a precise location where no specialised machinery exists, and then the original body is reconstructed out of – what exactly? The TV shows, films and books are generally vague on this point.

But you could (and I’ve no doubt others have) construct vaguely plausible theories about what is going on here. The transporter pad creates a sort of cone or column around the person to be “transported” and everything within that column is processed. 23rd century computers have to “lock on” to remote locations. Maybe something like a tractor beam (not that that exists yet either) is used to process atmospheric atoms within a similar column at the destination in order to make the mass required to reconstruct the transportee. It’s a hugely big problem, but as presented in Star Trek, a hugely big solution is required to make it all work – and sometimes it doesn’t. It can go wrong, with horrible consequences. So this is borderline, but I would say, provided one makes allowances for three hundred-odd years of technological advancement, there is cause-and-effect here, and crucially, the cause (although we can’t examine it all in detail) seems sufficient for the effect.

Now compare those instances with death-by-phaser. Here’s a YouTube video. Watch the phaser deaths that occur around the 1’05” mark.

A phaser beam hits someone and they simply disappear. On a very basic level, there’s cause-and-effect here. Pull trigger on phaser, phaser beam hits redshirt, redshirt disappears. But what do we suppose the phaser beam is doing? It’s not just a narrow pencil-thin column of heat, it doesn’t drill a neat hole through its unfortunate victim. It’s not affecting the air between the barrel and the target, so it’s not radiating out a destructive cone – that would also affect surrounding matter, or if you were too close, not get the whole target. No, the phaser, or the beam itself, just “knows” what is target and what isn’t, what is to disappear in a flash and what is to be left unaffected. Not only is there no plausible mechanism by which this information could be imparted, the presentation doesn’t even begin to hint at a possible solution. Transporters are almost certainly impossible, but at least the series waves its metaphorical hands in the direction of a proper cause. Here the given cause is far to slight for the actual effect.

So, to me, death-by-phaser is far more like a wand being waved and a word being spoken and a Harry Potter character vanishing into thin air. No mechanism exists to define the scope of this action. No care must be taken to isolate only that which you wish to vanish. The waving of a wand and the saying of a word are in no way sufficient to account for the immensity and the complexity of the action which results – which is why Harry Potter is magic and not science-fiction. But magic gets smuggled in under our noses all the time. “I’m in” is a form of magic, as is the bus jump in Speed as is Radar’s ability to anticipate Colonel Blake and so on. The rule is – if the cause is sufficient for the effect, then what you have is science, but if the cause is insufficient for the effect, then what you have is magic.

But this applies to the real world of knowledge, science and medicine as much as it does to storytelling. Remember that humans look for cause-and-effect even where none is present. So, in earlier times, we ascribed the sudden and unpredictable shaking of the earth to the wrath of peevish gods. But since we have no idea where these gods are, what we have done to annoy them, or how they physically interact with the earth to cause such violence, this is magical thinking. Modern plate tectonics provides a much more thorough model (although not yet quite complete) with all the cause-and-effect you could wish for.

Another area where this test can be applied – one familiar to regular readers – is that of homeopathy. Both homeopathic and pharmaceutical treatments for a given condition, joint pain say, have the same superficial cause-and-effect appeal. You take pill. Pill makes pain better. But ask the chemist how their pill works and the answer will include fantastic quantities of detail, each element enhancing the cause-and-effect in the account. Non-steroidal anti-inflammatories inhibit cyclooxygenases, leading to a decrease in prostaglandin production. Opioids bind to and activate opioid receptors in the body’s tissues, and so on and so on. As with plate tectonics, the account may not yet be complete, but gaps are acknowledged, and effort continues to plug them. Pharmacology is science.

With your homeopathic pill on the other hand, the explanation is far sketchier. Like cures like, you may be told. How? It just does. Dilution increase potency. How? It just does. Does that mean half a pill will be twice as powerful as a whole pill? No, you have to dilute and the succuss (shake). What does that do? We don’t know. A cursory examination of the cause-and-effect supposedly driving homeopathy is enough to tell anyone that a woefully insufficient cause exists for the claimed results. Thus, we can confidently state that homeopathy is magic.

And, of course, the other thing we know about magic is that it isn’t real.

Talking to my GP father about homeopathy #5

Posted on May 10th, 2010 in Skepticism | No Comments »

Part four is here.

TOM

I don’t think there is any evidence that a significant number of people are being harmed by choosing to go to alternative practitioners. There is undoubtedly a considerable morbidity and mortality arising from the side effects of drugs and the mistakes that happen in the conventional medicine system. Is the harm caused by inappropriate use of alternative medicine on the same scale?

Interesting. This is basically the Toyota defence isn’t it? “Until a lot more people are killed by our product, we don’t see the need to perform a recall.” Not really what I want my healthcare to be based on.

Your challenge to produce harm produced by alternative medicine “on the same scale” as that caused by evidence-based medicine does not make clear whether you want the same absolute number of corpses or the same level of morbidity per person treated. In either case, it is unlikely that I will be able to provide such evidence. In the first, less reasonable, case, despite my fears that alternative medicine fairy stories of miracle cures are distressingly seductive, it is obviously the case that almost everyone in Britain will be treated by conventional medicine at some point, but only a portion will seek alternative treatments. So even if they kill at the same rate, alternative medicine will be way behind. Even in the second, more reasonable, case, I doubt I will be able to meet this target since if someone is really sick and foolish enough to seek useless alternative therapies, it is likely that once they are at death’s door, they, or a relative, or someone will have the sense to take them to a proper doctor, and we can only hope they will be in time.

But just as we wouldn’t do without conventional medicine on the basis that a significant percentage of people who go into hospital will be killed simply because they went into hospital rather than because of the condition which brought them there, the question we should ask of homeopathic and other alternative remedies is not “what harm do they cause compared to other things in the world with the potential to do harm?” but “are they a net force for good?” or in other words “are we better off with or without them?” These calculations are not always easy to perform. Some people argue that the use of pesticides on fruits and vegetables contributes to deaths from cancer, and this is likely true. It is estimated that around 20 extra cancer deaths occur in the United States each year due to chemical pesticides. While this is a very small number (around 300 Americans die each year drowning in the bath by way of comparison), surely even 20 is too many. We should obviously get rid of these horrible carcinogens polluting our mealtimes. Alas, the immediate upshot of cutting back pesticide use would be to decrease farmers’ yields, which in turn would raise the price of their produce, which would in turn reduce the national consumption of fruits and vegetables, which would in turn increase the rate of cancers in the USA, probably adding 26,000 to the total (less the 20 who would no longer die because of exposure to pesticides). (Sources for these numbers on request).

So, let us ask what the harm caused by homeopathy is. I (and some of the commenters to this blog) have given you some examples of the horrendous consequences that can follow trusting that homeopathy will be efficacious in treating serious conditions. You say that you hope most people will have the sense to avoid magical treatments when their life is in danger, yet you seem quite sanguine at the prospect of people who are quite determined to erode this life-preserving common sense. Furthermore, the harm which homeopathy can cause is well-documented, by Simon Singh, through the campaign behind the “mass placebocide” or through this website dedicated to answering this very question. Note that Simon Singh makes the same connection that I do between homeopathy and mistrust of vaccinations. The body count due to recent refusals to vaccinate is easy to check.

Having agreed that harm from homeopathy is both possible and actual, we have to ask who it helps. Broadly, there are three kinds of people who might seek this kind of therapy.
• People suffering from serious conditions which require medical treatment. I am reassured that the Royal Homeopathic Hospital would refer these people, but still marvel at the cognitive dissonance required. Clearly, any people in this group run the serious risk of suffering quite unnecessary harm. For this group, the existence of alternative modalities is nothing but a negative. It will cause them to delay treatment.
• People suffering from self-limiting conditions which require little or no treatment in any case. These patients will not be directly harmed, except in the wallet. They will be spending money needlessly since their condition will improve on its own. This is a minor negative, but not in terms of health outcomes. It’s their money, you may argue, and they entitled to waste it however they please. Unfortunately, if they develop the belief that the alternative modality of their choice is effective in treating their minor self-limiting condition, then they may be more inclined to believe the more exotic claims made for this treatment and subsequently find themselves members of the first group. Thus, for this group also, the existence of alternative modalities is nothing but a negative.
• Finally, we come to the group which you have chosen to emphasise. Those suffering from chronic, non-life threatening conditions, for which no evidence-based treatment is offered or available. In a competitive marketplace, they are likely to get more sympathy from a quack than from a real doctor, and they may benefit in the short term from the (very powerful) placebo effect. They may even be permanently cured if they condition had no physical reality in any place (e.g. phantom limb sufferers or people who claim that mobile phone masts give them migraines). While I agree that some people in this group are made to feel better, I do not believe that this benefit is worth the cost, nor do I agree that the Royal Homeopathic Hospital (and other less scrupulous purveyors of fairy stories) is the best possible means of providing this kind of comfort. Relaxation techniques, improved diet, better trained GPs who understand the importance of providing emotional support to their patients and a better patient understanding of how psychosomatic illnesses affect us are all available and likely to be efficacious. You can probably add to that list.

So, no, I don’t believe that the harm done by homeopathy is on the scale of the harm done by conventional medicine. But conventional medicine does tremendous amounts of good which is absolutely unavailable anywhere else, whereas homeopathy does quite a lot of insidious harm, for only a little bit of good, almost all of which is available elsewhere.

Overall what makes me cross is big companies misleading people about important issues such as health. It isn’t just homeopaths who do this. Idiotic media celebrities like Gillian McKeith do the same thing. So do some drug companies. But lying to the general public about health is unlikely to have a positive effect on society, and institutions like the Royal Homeopathic Hospital, whether or not they cause harm themselves, make the lies of the unscrupulous so much more convincing. Campaigns like 1023 aim to provide more accurate information as a corrective to this – to help ensure that people do know that they need to be in hospital with severe chest or abdominal pain.

One of the recommendations made in the recent government Evidence Check was that homeopathy should have to be approved by NICE before being made available on the NHS. Do you think it would get NICE approval? On what grounds? If it wouldn’t, why should it be made available at no cost to patients when other more efficacious drugs are only available privately?


JOHN

I think the argument along the lines of ‘how many more people have to die before you admit I am right?’ is a bit over the top.

I think there are two areas where we are out of synch with each other. They concern the concepts of ‘health beliefs’ and ‘risk management’. I also have a problem with your relentless positivism. Science and logic and common sense are very valuable but there are other powerful forces that govern human beliefs and actions and they need to be understood rather than dismissed as simply foolish.

Risk management. Doctors deal with this every day. We are constantly having to decide whether to advise a patient to take a particular drug which has potential for good and for harm. This can be quantified to some extent. We can say, trials show that if you take aspirin your risk of a heart attack will be cut by X percent over ten years. There is also a risk that you will have a cerebral haemorrhage but the risk is much less so the odds are in favour of taking it. Often, when presented with this information, people say: but I’d rather not take any risk!  But this is impossible. It seems to be a hard concept to grasp. The other problem is that you can’t guarantee that taking the drug will have any benefit for that individual. The effect is only measurable on the population as a whole.

Health Beliefs. You argue that ‘the placebo effect’ can be more safely produced by doctors taking more time, being kind and sympathetic, being trained in psychology etc. But some people’s health beliefs are very physically based. They have very fixed ideas that only by swallowing a pill or having needles or massage or whatever can their bodily pains be relieved. Trying to argue people out of this is usually a waste of time. Often it’s based on personal experience or family traditions.

But I still think that for the overwhelming majority of people will be guided by conventional medicine when there is a really effective treatment available. I have yet to meet a patient who said, no, I’ll just have homeopathy etc. when it was a matter of life or death, or when I said, try this, we can really help you.

Maybe one can’t justify alternative medicine being paid for out of taxation or insurance. But I don’t think it will stop those people who believe in it from getting it privately if they have to. I don’t believe it should be banned by law because a few rather idiosyncratic people may wrongly choose it instead of conventional medicine when conventional medicine is in a position to prolong their lives or relieve their suffering.

That would be an abuse of their human rights. better to ban smoking if you must ban something. Or alcohol? Though, that has been tried.

And now I really have had enough of this subject. By all means have the last word. But after that,  can we talk about something else? How about euthanasia?


TOM

This seems a good place to end the conversation, as we are moreorless in agreement! I agree with everything you say about risk management and almost all of what you say about health beliefs.

I agree that for some people a physical intervention such as a pill, or a series of carefully placed needles, is required to trigger the placebo effect. You can’t argue someone in or out of the placebo effect – it isn’t a product of conscious decision-making.

I also agree that attempting to ban alternative medicine would not make it go away, any more than prohibition stopped people from drinking. Even regulating advertising of alternative medicine only encourages those who sell it to seek editorial promotion instead, which is more convincing than advertisements in any case.

So what do I want? Well, essentially just what you want – for everyone in the world to understand the difference between the claims made by evidence-based medicine and the claims made by magic-based medicine and so make informed health decisions. It’s just that you see very few people who get this wrong coming through your surgery and I read lots of blogs and listen to lots of podcasts which detail case after case, so we have formed very different views as to the scale of the problem. No doubt each of us has a somewhat skewed perspective.

The one area where I would be tempted to disagree would be on the subject of my “relentless positivism”. I am well aware that forces other than science and logic govern human beliefs and actions. However, I submit that to better understand these forces in an objective manner, the only option is to study these very forces with the tools of science and logic. But that’s not really what is under discussion here.

I imagine we’d boringly agree about euthanasia. Human dignity, relieving suffering and patients’ wishes all sometimes trump “first do no harm” but great care must be taken in exercising this option. Is that roughly your view too?

Talking to my GP father about Homeopathy #4

Posted on April 30th, 2010 in Skepticism | 4 Comments »

Part three is here.

TOM

Muddy waters? Not at all!

You say that you don’t see complementary/placebo medicine as competing with scientific medicine. However, not all placebo practitioners see it that way. Many peddlers of placebo cures directly attack what they sometimes term “allopathic” medicine as causing harm, arranged to make money for “big pharma” or who-knows-what other ills. Here’s a quick sampling of web pages and opinions which it took me about two minutes to find.

“Antibiotics kill bad as well as healthy bacteria. This results in weakening of immune system. Homeopathic medicines strengthen the immune system by building resistance to sickness. They do not disturb or hamper digestive system.”
http://www.otherhealth.com/research-scientific-validity-homeopathy/9877-modern-medicine-allopathy-homeopathy.html

“Allopathy’s strength lies in intervention. When you need a shot of penicillin, you need a shot of penicillin. When your body needs insulin, no therapy other than insulin injections will do. If surgery is required, then surgery is the solution. Allopathy’s weakness lies in the simple fact that removing the symptoms of a disease does not necessarily remove the cause of the disease. Its greatest weakness is that many of its cures are killing people. One reason many people have left conventional medicine for healthier alternatives is that death is not an acceptable side effect.”
http://www.mnwelldir.org/docs/therapies/what_is.htm

“[The World Health Organisation] wants to remain ignorant about AIDS, because it is under the control of the big Multinational Companies manufacturing the socalled HIVdetection kits and the highly toxic drugs like AZT, which products they must sell by any means to make the big bucks.”
http://www.wonder-cures.com/art8.htm

Now, it’s easy for you and me to dismiss these as the ravings of cranks, and indeed you then say “you think most people are aware that they [evidence-based and alternative medicine] are different in conception”. Presumably you think this is important. If people are aware that antibiotics are a sensible (although not a guaranteed) treatment for pneumonia but that homeopathic remedies are unsuitable, then they will steer clear of them – despite the fact that they are readily available.

How does it help this vital distinction between evidence-based and placebo medicine to be maintained if placebo medicine shares shelf-space with evidence-based cures in Boots? If NHS doctors provide imaginary treatments on request? If evidence-based medicine is offered side-by-side with placebo medicine, as birthing pools are provided side-by-side with epidurals in maternity wards? How can the lines not be blurred in patients’ minds? And do you not share my concern about the possibility of harm once these lines are successfully blurred? This is where the muddy waters really are, I believe.

Yes, there is no market for herbs on broken legs, but there is a big market for homeopathy which is no more magical, but which presents a more credible face to the western world. Homeopathic pills look like medicines (which partly accounts for their effectiveness, although studies show that saline injections – a more impressive intervention – can be even more effective in pain relief for example) and so they appeal to a moderately medically-literate audience. But that appeal depends on blurring the distinction which you identified. And it works – the market for homeopathy in the UK is estimated at £30m annually (£4m on the NHS). For complementary medicine as a whole it is £1.6bn. That’s a lot of money to spend on magic.

I agree, that if it were the case that all promoters of homeopathic medicine (and other placebo interventions) were eager to direct their patients towards evidence-based interventions for more serious or urgent situations (regardless of the inevitable cognitive dissonance that this requires), then they would do considerably less harm. But your personal experience of this kind of co-operation happening in China or on the NHS doesn’t change the fact that it is not the norm. When researchers visit high street homeopathic vendors, they are prescribed useless sugar pills as a malaria prophylaxis and told not to bother with evidence-based immunisation. When patients visit websites for information about homeopathy, they may read information which presents evidence-based medicine as a cure which is worse than the disease. When practitioners of placebo medicine are criticised in the press, they often attempt to silence their critics by legal means, as happened recently to Simon Singh, heedless of the fact that open scientific debate is essential for determining best options for patient care.

The upshot of all of these behaviours could be and is that patients die who might well have lived had they not been fed this misinformation. The further consequence is that it becomes easier for evidence-free attacks on mainstream medicine, such as the anti-vaccination campaign, to take hold in the public consciousness. This seems to me like far too high a price to pay for the temporary alleviation of chronic pain in some patients, or the illusion of relief from self-limiting conditions for others.

We know that pneumonia kills, but that it can be effectively treated by antibiotics. We know that a lot of people are unaware of the fact that homeopathy is a useless treatment for pneumonia, because homeopathic treatments are readily available, and if no-one thought they would be effective, market forces would ensure that – like herbs for broken legs – they would be nowhere to be found. People who believe that homeopathic remedies will treat their pneumonia will likely die. Homeopaths have a strong vested interest in maintaining this belief. Homeopathic pills in Boots and the presence of an institution such as the Royal London Homeopathic Hospital are very effective in helping to maintain this belief.

Your initial question to me was “why are you so cross about this?” My question to you is “why aren’t you?”


JOHN

The reason I am not cross about it is that I don’t think there is any evidence that a significant number of people are being harmed by choosing to go to alternative practitioners. Do you know of any?  Its true that these advertisements appear to be very misleading, but I think people have enough sense to know  that they need to be in hospital with severe chest or abdominal pain and that they need a surgeon for a broken leg.

They go to the alternative practitioners only when they have discovered that they have the kind of symptoms which are painful or distressing but not due to any disorders that conventional medicine can help except by support and the doctor’s continuing interest  and concern. Sadly,  that may be more forthcoming from a homeopath.

And, as I have said there is something about these  miracle cures that has a very strong appeal to human nature. I’m not sure that one can or should legislate against it unless you can show that harm has resulted on a significant scale. What is significant? Well, there is undoubtedly a considerable morbidity and mortality arising from the side effects of drugs and the mistakes that happen in the conventional medicine system.   Is the harm caused by inappropriate use of alternative medicine on the same scale?

Talking to my GP father about homeopathy #3

Posted on April 19th, 2010 in Skepticism | 2 Comments »

Part two is here

TOM

I’ll keep this as brief as I can.

Parts of your reply misrepresent my position slightly. I’d like to clear up any misconceptions.

I don’t equate evidence with certainty. Absolutely, we can be less sure than we would ideally like that a given intervention will be effective in a given case, but we always have an evidence base on which to make such judgments – even if the best we can say is “this is a wildly experimental treatment which stands as good a chance of curing your cancer as it does of killing you on the spot, but we’ve run out of other things to try – what do you say?” If this is a lottery, it’s one where we can often buy an awful lot of tickets.

Thus, I continue to insist that everything is split into two categories. A physician can recommend a treatment because there is evidence to suggest that it will be effective or can ignore the evidence which shows that a treatment is no better than placebo and prescribe it anyway. A physician can honestly tell a patient that the evidence is inconclusive, or dishonestly claim results which are not supported by data of any kind.

The other slippery issue is that of false claims. You approve of the wording on the Homeopathic Hospital’s website regarding Iscador, but I’m not sure you are looking at this website the way a patient might. Here’s a page from a website all about Iscador.

Here we read that “one of the primary functions of Iscador® is that it stimulates parts of the immune system that can slow the growth of cancer cells”. This, as we know, is rubbish. On the same website, there is a link to a single study, published in a Complementary Medicine Journal which purports to show a significant benefit to cancer sufferers using this preparation. However, as we have already seen, a review of all the available literature shows that overall there is no good evidence that Iscador has this effect. This is the usual pattern. Evidence-based (or science-based) medicine which reviews all the available evidence, and people with a pill to sell (herbal or pharmaceutical) who pick only the studies they like and ignore the rest.

Now consider the effect on a patient who may have seen these other claims made for Iscador’s magical cancer-healing powers, coming across Iscador on a National Health Service website. It isn’t the details of the mode of action that stick in the mind – it’s the key message MISTLETOE CURES CANCER. And this message is reinforced, not contradicted, by the product’s availability through an apparently prestigious and trustworthy source. So while the NHS might stop short of actually saying “mistletoe cures cancer”, it ends up delivering that message just the same.

It is at least partly for this reason that the Ten23 campaign is targeting Boots. When homeopathic preparations are seen side-by-side on the shelf with active pharmaceuticals in the country’s largest and most-trusted pharmacy, it is almost inevitable that patients will get the wrong impression (this is reinforced by dire warnings on the sides of the bottles). So, it becomes impossible to clearly send a message that homeopathy will not cure your cancer, is not a suitable defence against malaria, will not cure your child of eczema and so on, and the result is that people believe the hype, and sometimes that belief turns out to be deadly.

My feeling is that placebo cures possibly do have a role in treating chronic conditions such as back pain, may even have a role in treating minor self-limiting conditions although this is more likely to be simply a waste of everyone’s time,* but are a danger as soon as they are let near anything remotely life-threatening. With a powerful lobby that can’t bring itself to say “for god’s sake get proper medicine if you feel really poorly,” the only answer is to try and destroy the credibility of these interventions as much as possible, or stand by as greedy corporations and misguided practitioners continue heedless of the harm they cause.

So it isn’t that homeopathy on the NHS is itself prescribed in a cavalier or life-threatening way, rather that this activity is an enabler, enhancing the credibility of others with fewer ethics or less regard for evidence, and also fuelling the fires of hysterical media coverage of things like Andrew Wakefield’s infamous MMR paper. Is public demand for medicalised quick-fixes a good enough reason to accept this corrosion of public standards of evidence? I don’t think it is.

You asked the excellent question – can you have medicine without quackery? I don’t know if you can. But I do know that mainstream medicine has no business flattering the quacks.

One final question for you. Many churches pray for people with life-threatening illnesses and many people feel happy that people are praying for them. What if this were offered as a service on the NHS? Is government-sponsored prayer also something you would endorse?

* Caveat – when I have a sniffle, I’m perfectly well aware that no matter what I do it will probably last 3-4 days, but I prefer Lemsip to a homebrewed hot lemon drink because the fact that Lemsip tastes a bit medicine-y makes me feel better. The placebo effect is also present with pharmaceutical interventions of course.


JOHN

I don’t see complementary medicine (or if you prefer ‘placebo medicine’) as competing with scientific medicine. I think most people are aware that they are different in conception and that for some major illnesses or accidents only science will deliver. Very few people put herbs on broken legs these days. There would be no demand for it on the NHS.

Whether there would be a demand for prayers for healing on the NHS I don’t know but it might catch on and save a lot of money currently spent on visits to A and E.

When I was in China last year, I was interested to see that some of the hospitals and clinics we visited  had Western scientific medicine  and TCM ( traditional Chinese Medicine) departments running side by side. The patients chose which one to go to, but if the doctor they went to initially thought they had got it wrong they were cross-referred to the other wing.  So the two approaches can co-operate.

That’s all I  have to offer for now. I hope I have muddied the waters a bit so you can get to work clearing them again.

Talking to my GP father about homeopathy #2

Posted on March 23rd, 2010 in Skepticism | 3 Comments »

Part one is here…

TOM

Let’s start by enlarging the list of things we both agree on.

I agree that placebo pills cause no direct harm whereas absolutely any active chemical introduced into the body has at least the potential for harm, even those which have long been associated with health and well-being; there is, for example, some evidence that routinely taking vitamin supplements when not indicated may have negative health outcomes.

I also agree that homeopathy and other alternative modalities make some people feel better but regret that the only way to safeguard the efficacy of the intervention is to mislead patients as to what is going on. It is important that the patient believes that the pill, needle, shamanic ritual or whatever will be efficacious. When prescribing even (or especially) a very risky treatment like radiotherapy, a physician can explain the potential costs and benefits, based on a large body of evidence, in order that the patient can make an informed choice. Prescribing placebo treatments as if they are effective removes that informed choice, but telling the patient “this won’t work” is a self-fulfilling prophecy, so homeopaths – whether or not they are themselves deluded – are deceiving their patients.

Of course, if patients want it and are made to feel better, why should we worry about removing informed choice? I do have an answer to this question, but I don’t want to get ahead of myself. Let me finish responding to your points first.

You write that you think that homeopathy is an appropriate use of public funds to treat chronic conditions for which no effective evidence-based treatment exists, and it may be that the London Homeopathic Hospital does restrict itself to interventions in those cases. It may also be that all of the Homeopaths working there are medically qualified (I wasn’t aware of this – do you have a reference?).

But it is not true, despite the name, that within the tranquil environment of the LHH, homeopathy is the only evidence-free service available. A quick look at their website reveals that they are also happy to provide Reiki, Acupuncture, Reflexology and Iscador – a new one on me – which we learn is “a preparation of mistletoe, which enhances immune system responses.” Bearing in mind that this is on a page titled Complementary Cancer Care Programme, doesn’t this worry you at all? Do you really want your patients believing that mistletoe will cure their cancer? Who benefits from sustaining this belief, except those who have “Iscador” to sell?

Only one passage in your very thoughtful email stood out as something I would really take issue with – this one: “Every form of medical practice has its own philosophy, including evidence based medicine, which, by the way is not always as reproducible in an ordinary community setting as it is under strict experimental conditions.  Even EBM in most cases only offers a probability of a cure or improvement: a significant proportion of those treated will not benefit at all. Some will be harmed.” Let me begin with your last point. Yes, every active treatment has a risk, but as I’ve said EBM enables patients and physicians to make informed choices. It is possible to do great harm with the best intentions, which is way we must hold evidence in such high regard. Are you familiar with the on-going libel case which the British Chiropractic Association recently brought against the journalist Simon Singh? This case illustrates one of my chief reasons to want to see homeopathy and other sham treatments no longer flattered by government support.

You warn me that lab trials are not always replicated in the real world. But my definition of “evidence” is not so narrow as to only include clinical trials (although I still regard a large, prospective, double-blind, randomised trial as the gold standard for determining whether and to what extent a given intervention will be effective). We can certainly obtain data from retrospective analyses, and from seeing what actually happens in the real world. Sometimes these results are surprising, but EBM means we can’t ignore results which sound wrong to us or which go against long-cherished notions.

For example, the United States Preventative Services Task Force recently altered its recommendations regarding breast cancer screening. In their report was the rather surprising and controversial finding that self-examination appears to have no effect whatsoever on breast cancer death rates, but that it can create a lot of unnecessary fear and alarm. The emphasis (at least in the States) is now on breast awareness – has anything in your breasts changed recently? – rather than teaching women to perform rigorous and regular self-examinations. Many people object to this advice on the basis that they have a belief that breast self-exams are important, or that they have a story about someone who found a lump during such an exam, and whose life may have been saved thanks to the operation which swiftly followed. This doesn’t alter the evidence which is that such exams make no difference to health outcomes, and we take lives in our hands if we allow our preconceptions and biases to defeat the evidence.

Where we really part company is when you describe different forms of medical practice as having their own philosophies. It sounds to me as if you are claiming that each of these philosophies is equally good, and the only real difference is in the philosophical starting point, EBM being no better and no worse than homeopathy. While that might be a useful diplomatic position to take, I simply can’t accept it, any more than I would accept Apartheid (or any one of a number of other atrocities) as a mere difference in political philosophy.

While it is true that there is a bewildering array of supposedly therapeutic modalities available, they all slot cleanly into one of two categories – based on evidence or not based on evidence. Anything not based on evidence is based to some extent on subjectivity and personal choice. A practitioner who accepts Reiki but rejects Iridology has no objective basis for making this choice. They simply prefer Reiki and allow confirmation bias to do the rest.

When the system is working (and god knows it doesn’t always) evidence-based medicine prevents even the most enthusiastic champion of a particular intervention from being able to claim that it works when it doesn’t. We know that Dimebon is ineffective for treating Alzheimer’s and so it will not be prescribed. We know that mistletoe will not cure cancer, but at the Royal London Homeopathic Hospital, they will apparently provide it to anyone who asks, and at the taxpayer’s expense.

This means that proponents of non-evidence based modalities have no choice but to duck the issue, or misleadingly cherry-pick from the data in order to create a false picture – which is exactly what the select committee experienced when it asked notable homeopaths to give evidence. Who benefits from this corruption of the scientific method? Isn’t this exactly the “denying the validity of science itself” which you hoped wouldn’t happen? Why does homeopathy get a pass? Because it does no harm? I’m not so sure that’s true.

You began by saying that you wanted to include in our discussion the whole context of treatment, and here again I agree enthusiastically. So, while we both know that a sugar pill in itself is guaranteed harmless, we also know that we are talking about a complicated cultural phenomenon which is not limited to a person taking a pill, and therefore nor is homeopathy’s reach confined to people suffering from chronic back pain. Nor is it always the case that homeopaths advise people who need conventional treatment to go and get it. In 2006, a survey performed by Sense about Science which was followed up by Newsnight found that every high street homeopath visited by a researcher about to travel to a malaria-infested country was given useless homeopathic quinine and none were told to visit a GP or a travel clinic or even to buy a mosquito net.

How are “responsible” homeopaths supposed to respond to this? The Royal Homeopathic Hospital did its best to smack down this irresponsible behaviour, but what are they really saying? “We all know the sugar pills don’t really work, so don’t for goodness sake go giving them to someone who actually needs medicine.” What would the True Believers make of that!? By beginning from a position which isn’t supported by evidence, “responsible” homeopaths are now in an impossible bind. They either have to defend the magic to the hilt, or admit that it’s all a load of imaginative nonsense. Not surprisingly, they hem and haw and waffle and evade, they dissemble in front of parliamentary select committees, and essentially they let the harm continue, such as in the horrifying case of the Australian couple who attempted to treat their baby daughter’s eczema with homeopathy until eventually she died after months of excruciating pain (they are now in prison).

So even if we are happy to divide quackery into good (prescribed by responsible physicians in tandem with evidence-based treatments where appropriate) and bad (prescribed willy-nilly by people with little or no medical training, and cheerfully offered as a substitute for evidence-based treatments, or non-medicalised interventions) – the question must still be asked can you have one without the other? This line of argument will eventually bring us back to vaccines, but I feel I should stop here, having written quite a lot already, and get another response from you.


JOHN

Can we aim for something a little more concise? I don’t have time for a discussion in this depth!

I am very pleased that you are taking such an interest in medicine and medical ethics. I note that you are a very much a postitivist in your approach and have no time for post-modernist ideas of different beliefs having equal validity! Well I don’t know that I have in extreme cases where it is clear that science has the best solution (e.g. antibiotic for pneumonia or angioplasty for blocked artery).

However, there are many grey areas where all is not as certain as one would like. As I have already mentioned, the much venerated ‘gold standard’ trial doesn’t simply distinguish effective from ineffective interventions. When the effect is really clear you dont need a trial. What we usually get is a statistical result showing that the probability of an effect is greater than could be due to chance alone. This means that some patients will do well but others won’t. The NNT (number needed to treat) may be 4 or 5 for a good drug or 10 or more for many others. So we can’t promise our patients that even a gold standard treatment will do them any good (or no harm). It’s a game of roulette. No one can tell what will happen to any individual.

It is unfortunately not the case that everything slots neatly into two categories, evidence-based or not. Evidence can be strong or weak. Evidence is continually being modified or even contradicted. It doesn’t apply to all individuals. So I tell my patients who ask me if complementary treatments are any good, that some patients find them helpful but others don’t. Same difference.

I agree that it would be dishonest to say that ‘a sugar pill’ will have a biochemical effect. But you can say that, some practioners believe that the way this pill was made can make a difference to the ability of some people to cope with an illness that conventional medicine has not been able to help very much.

I agree, of course, that no one should make false claims about these remedies. The information for cancer patients from the Homeopathic Hospital seems to me to make it very clear that these treatments are to help you cope with the illness and its definitive treatment rather than to act alone. I dont know about the eczema girl who died, it is often difficult to tell from these accounts exactly what happened. But no responsible doctor would have advised the parents to ignore conventional medicine. I don’t think this would happen at the Homeopathic Hospital. I went on a visit to the place and met the medical director Dr Sara Eames who was very clear about this. She also told us that all the medical staff have medical qualifications.

Can you have good quackery without bad? Can you have medicine without quackery? I don’t think you can have medicine based on evidence alone. There isn’t enough evidence for one thing. And people are often have health beliefs that are so firm that no amount of statistical information will influence them. The placebo effect, if you want to call it that will often make something happen. If you prefer you can call it the effects of thoughts and feelings being processed by different brain areas resulting in changes in pain perception, sensation, reduced muscular stiffness, improved mood and reduced anxiety etc. etc.

Enough for now!

Talking to my GP father about homeopathy #1

Posted on March 22nd, 2010 in Skepticism | 1 Comment »

I recently invited my father to an exchange of views about homeopathy – me from the point of view of an ethusiastic skeptic, eagerly slashing down falsehood and flummery wherever I find it; he from the point of view of a physician with decades of experience actually helping people to get better. Here is the first of our exchanges…

TOM

As I understand it, you think that NHS should continue to provide homeopathic remedies under some circumstance, and you are surprised at the strength of my feeling that they should not. Through this exchange of emails, I hope to better understand your position (since you speak from experience of treating patients, which I can’t) and to get you to better understand mine.

I imagine there is quite a lot of common ground between us, so before we get to the fun of the debate, here are some statements I think we can both agree with.

–          Homeopathy doesn’t work. Reviewing the evidence base as a whole reveals that homeopathy works no better than placebo. While a number of studies do exist which apparently show a more significant effect, these are always smaller studies often with methodological flaws. The better the study, the smaller the effect. This finding is confirmed by the Cochrane Collaboration and the recent Select Committee report.

–          Homeopathy couldn’t possibly work. The notions that “like cures like” and that “dilution increases potency” are pre-scientific magical thinking. Many homeopathic remedies are so diluted that it is literally true that not a single molecule of the original substance can possibly remain. How the water used for the dilution “remembers” the substance (and none of the other substances it had previously been in contact with) is a mystery which homeopaths avoid tackling.

–          Evidence-based medicine is, in general, a good thing. Especially within a cash-strapped NHS, treatments should be provided to patients on the basis of the best evidence available, not on a patient’s demand, nor a physician’s whim, nor in order prop up a discredited theory.

Assuming that we agree on these three points, it seems that you wish to make an exception to the evidence-based rule for homeopathy. Is that right? If so, here are the questions that next spring to mind.

  1. Under what circumstances do you think it is appropriate for the NHS to provide homeopathic remedies? What are the possible negative consequences of doing this under these circumstances (if any)?
  2. Do you feel the same way about chiropractic, reflexology, acupuncture, iridology, reiki (to name a few)? Are some of these interventions more worthwhile than others in your opinion? Are they all equally beneficial placebos or are some more worthy of public funding than others?
  3. Are you concerned about the drop-off in vaccinations which has taken place in the UK and the USA recently? Do you see any connection between this and people’s fondness for alternative medicine?

JOHN

I think the first thing I’d like to say is that when considering the possible beneficial or harmful effects of any form of  treatment, I like to include not just the intrinsic chemical power of the substance prescribed, but the whole context of treatment.

The so-called placebo effect also includes the  treatment environment, the way the doctor approaches the patient, her kindness and consideration,  her ability to listen and to accept people’s distress. The faith of both doctor and patient in the theory, the method, the procedures, and in the personal relationship are all important too.  Every form of medical practice has its own philosophy, incuding evidence based medicine, which, by the way is not always as reproducible in an an ordinary community setting as it is under strict experimental conditions.  Even EBM  in most cases only offers a probability of a cure or improvement: a significant proprtion of those treated will not benefit at all. Some will be harmed.

You and I don’t believe that homeopathic remedies can have any chemical effect as such. But this is not chemotherapy. These patients are not being treated with agents that can kill cancer cells or eliminate antibodies. The patients who attend homeopathic practitioners and the homeopathic hospitals are suffering from chronic diseases for which conventional chemical medicine or surgery can do no more. The treatment is not and should not be a substitute for ‘proper’ medicine.  At the London Homeopathic hospital, where all the clinical staff are medically qualified, they examine people in the usual way and advise anyone who needs conventional treatment to go and get it.  The homeopathic treatment process  (including the empty tablets) has the effect of making patients feel better and feel cared for. It gives them some hope. The placebo effect can be incredibly powerful and can certainly  relieve pain and other symptoms such as nausea, giddiness, anorexia etc.

Are the remedies expensive?  No, compared with modern drugs, often unscrupulously promoted by pharmaceutical companies and overprescribed by doctors, they are cheap. Unlike most conventional drugs there is no risk of  serious adverse effects. The London Homeopathic Hospital has just had a fairly expensive refit. It looks beautiful, calm, tranquil and peaceful. It would be a pleasure to go there for treatment.  OK they don’t cure any cancers or kill any bacteria. But a large part of medical practice is about helping people to cope with  chronic illnesses that can’t (yet) be cured or with distressing symptoms that are not even understood.

There has always been a need for some sort of magic and mystery in medicine, going back to the ancient shamans. This has not gone away, despite the acievements of scientific medicine. If some patients are helped to feel better by what we see as pseudoscience, why should we be outraged? It doesn’t mean that we are denying the validity of science itself.

I think the money is well spent!   And, by the way, I have no  personal interest to declare.

Finally in answer to your last three points:

1) I have argued above that the remedies should be seen in the context of the treatment as a whole and the kind of ill-health for which homeopathic treatment is suited.

2) I think a similar case can be made for the other complementary therapies but I dont think any of these are publicly funded at present. I think some treatment ‘stories’  will  help some patients  but not others and there is no reason why those whom conventional medicine still cannot help adequately should not explore them. It’s important, if you have a chronic illness, to have a plan, to keep trying.

3) I think the drop-off in immunisations has been largely due to Dr Wakefield, a thoroughly conventional medical researcher who got carried away and did a lot of harm. I don’t think homeopathy has anything to do with refusing vaccines of proven effectiveness.