28 January 2010

Faithbook

Interesting social media news from God's ReTweeter himself last weekend, as Pope Benedict urged priests to evangelise via Twitter, Facebook and blogging, on his YouTube channel: "The world of digital communication, with its almost limitless expressive capacity, makes us appreciate all the more Saint Paul's exclamation: 'Woe to me if I do not preach the Gospel'"

I know what you're thinking: "the Pope has his own YouTube channel?" Actually, the first thing that entered my head was: "Well, that will certainly make it easier for them to groom young boys for sex."

But in the spirit of enquiry, I thought I'd see how far this project has come, so I checked out the Pope's Twitter account for examples of his own 140-character 'microsermons'. Well, Benedict's right on the money - he's got his Twitter account up and running, and has already racked up 172 followers. Except he hasn't actually posted any tweets yet - not one. I think this is the most literal example of someone not practising what he preaches. And as head of a major religion that claims over 1 billion adherents (including the lapsed), 172 is a little disappointing.

He also isn't following anyone, which I suppose is fair enough until God gets a Twitter account. Being generous, I guess he fears that, were he to start tweeting and his entire flock to start following, lapsed or otherwise, he would permanently crash the Twitter servers, given how shaky Twitter was yesterday during the launch of the new Apple iHype device.

He does have a Facebook page, but, again, it only has two postings, the last one more than 18 months old. Now far be it from me to suggest that, just because he doesn't tweet himself, Facebook regularly, or have a blog, that he clearly doesn't actually get the the whole social media revolution. But it was only a year ago that the Pontiff was actually speaking out against the use of such media: in January 2009 he warned that “obsessive” use of mobile phones or computers “may isolate individuals from real social interaction while also disrupting the patterns of rest, silence and reflection that are necessary for healthy human development".

Believe me, Benedict, the old tweeting can get a bit compulsive, so maybe it's just as well you have drawn back from the brink before submitting to temptation. But if you do, I'll be waiting for your thoughts, as follower number 173.

24 January 2010

Thinking positive

At the risk of sounding a little like a scratched record*, I had a bit of an epiphany while mulling over thoughts of homeopathy and evidence of efficacy. Homeopaths make a living by thriving on those who feel conventional medicine has, in some way, failed to cure their illness. And given there are 60 million of us in the UK alone, it only takes the tiniest of tiny minorities to supply enough anecdotes of miraculous cures to supply an entire industry. One can cite all the studies in the world about placebo, RCTs and the body's natural immunities, but personal testimony can be powerfully affecting to those considering an alternative to prescription medicine.

Mainstream science tries to counter this attack-by-anecdote through a number of arguments, most commonly by citing 'regression to the mean' - or "you would have got better anyway" as most would understand it. Impossible to prove at the individual level, it also seems unsatisfactory in dealing with the more rigorous homeopathy cure stories: where patient has pain, takes drug and feels worse and then takes homeopathic cure and leaps Lazarus-like to his feet, on top of the world. This is 'regression to the mean' + side effects of drugs - a simple enough explanation, but hardly a glowing testament to the wonders of modern medicine. But there is something else we should also consider: supposing you aren't sick at all?

I was reading about the second of the great cholera epidemics to sweep London, in 1854, when fatalities at the London Homeopathic Hospital were almost one third of those found at nearby hospitals practising conventional medicine. This is sometimes cited as evidence for homeopathy's noble and efficacious tradition - and at first sight appears impressive. Until you consider that "conventional" medicine at the time was unaware of germ theory, did not employ anti-septics and considered blood-letting as a mainstream cure for many conditions. In the usually dirty and crowded conditions, it would seem that doing nothing cleanly presented a better chance of survival than any of the leading medical interventions of the day.

This prompted me to reconsider the much trumpeted homeopathic success stories in the light of the modern version of harmful interventions: the false positive. Not even the most brilliant tests modern medicine can muster are 100% accurate. Every year a tiny proportion of people taking any number of tests will test falsely positive for a condition they don't actually have. And of course the less accurate the test, the higher the number of false positives, especially if the condition is more marginal - tests less accurate, drugs less well developed, longer term understanding of their side effects less documented. If even 0.1% of testees are false positives, that will give us a fair chunk of people every year taking drugs for a condition they don't have. And when they turn to the magic water of homeopathy, they get a miraculous "cure" and tell all their friends.

In many ways conventional medicine is a victim of its own success. The powerful demonstrations of its power and range, from the elimination of smallpox to the creation of calpol, makes people blase about its limitations, and the risks at the margins. No treatment is without risk (even homeopathy, if taken as a cure for a serious condition), and if people believe it is, they will continue to be vulnerable to quackery when they become disappointed. Public education about science and medicine needs to make people aware of this, to manage expectation and help people understand its vulnerabilities, as well as its triumphs.

* = for the under 30s a 'record' is the old word for a collection of MP3 files. A disc made from vinyl, it carried up to about 12 tracks embedded within the grooves of both its surfaces whose sounds were reproduced via a needle through amplification equipment, making it very vulnerable to damage, often in the form of getting stuck in a repeat loop of several seconds.

Fighting water with water

As regular readers of hofflimits will know, I've been watching with amusement the current spat between the purveyors of homeopathy and the recent 10:23 campaign to highlight its barminess (www.1023.org.uk). If any of you are on Twitter, I'd recommend joining in (#ten23) as it can be very fun in small doses, pun intended. However, it does get a bit wearying after a while seeing the level of "debate " it can sink to - Godwin's law doesn't even come into it. Mainly, I think, because both sides are arguing from two totally different definitions of what constitutes evidence, which holds the key not only to the whole debate, but to the very heart of why people are attracted to moonshine theories.

As I have said elsewhere, looking at all the available evidence, homeopathy performs no better than placebo. Already by saying that, I risk incurring the wrath of believers who will cite a number of studies that supposedly do prove its efficacy. Until you discount those studies that are unreliable because of methodological flaws, and then proofs start looking a little thin. So homeopaths will come back with personal testaments of all the people they know who have benefited from homeopathic "treatment" - and here's where we hit the crux of the matter: Sceptics will never win the argument until they debate on the same terms - the battle of PR.

Say I read an account of someone who passionately recounts the wonderful effect a homeopathic treatment has had on his life. It's an anecdote - file it. But supposing you were to read a book of 100 such anecdotes, all true and all argued with conviction. By the time you get to number 76, many people would be wilting under the pressure of such "evidence". The human brain is hardwired to be fooled by poor risk assessment skills when dealing with large, impersonal numbers in the face of direct experience. Risk of anything is always judged as higher if you know someone who has been personally affected; it's been a useful survival skill, and perfectly reasonable when you consider that for 99% of hominid existence, most individuals would not know more than 30 people.

In their heart of hearts homeopaths know they will never win the argument on scientific grounds, unless the laws of chemistry and physics have changed (and the fact our planet continues to spin would indicate this isn't so). Their best shot is emotional appeal based upon weight of anecdote, something scientists naturally shy away from. But I would argue that half a billion years of human evolution can't be wrong - and one of the biggest parts of the UK economy is testament to this approach: Public Relations. Scientists need to embrace it, to take the argument to the homeopaths.

Returning to my mythical book of 100 Homeopathic Cure Stories, scientists need to come back with a database of 100,000 stories about how a particular drug saved a life. Start with one drug - probably any favoured target of the flat earthers - and bring weight of anecdote to bear on its efficacy argued from an emotional, personal perspective. Keep the evidence handy for those who want to check it, but remember it's not the number of peer-reviewed articles that will count, but the number of names you can cite.

Ridiculous? Probably. Impossible? Well, difficult certainly. But it's not actually the stories or names themselves, it's the act of compilation that will be the story. The world's biggest book that will knock a few placebo tales into a cocked hat. If they rise to the bait, then you've forced them to undermine their own position by dismissing anecdote as being non-scientific. And the science itself leaves them nowhere to go.

23 January 2010

A tale of two cities.

One of the interesting things about economics is that just because it is "economically rational" to do something, it doesn't mean it is actually sensible, morally good, or even in the best interest of the majority. Phenomena like the Tragedy Of The Commons and Externalities demonstrate that, under capitalism, self interest can create Bad Things as much as it can be the dynamo of wealth creation.

So I thought it was interesting to consider two different approaches to the ebb and flow of market forces. I offer this merely as an observation, not necessarily proof than one is a better outcome than the other:

2005 - Danone shares rise by 20% on the back of rumours of a takeover by PepsiCo
2006 - Kraft don't rule out non-hostile bid for Danone, provoking further takeover speculation
2006 - French government introduce laws to protect "strategic industries" such as Danone, to popular acclaim.

2008 - UK government lends RBS £20bn to save it from collapse
2009 - UK government extends its stake to 68% of RBS
2010 - RBS lends Kraft £7bn to take over Cadbury-Schweppes to popular indignation

On the one hand, it confirms that there is no room for sentiment in British banking. On the other, I suppose it means, as a UK taxpayer, I now own part of Cadburys. Next time I'm in Birmingham I'll try to pop in to claim a curly-wurly before they start filling them with cheese.

21 January 2010

A data remember

The digital world went a little moist with anticipation this week, as Sir Tim Berners-Lee launched the new government website data.gov.uk, opening up reams of HMG's information to the wider world. Everyone seems terribly enthusiastic about the event and, in particular, how user friendly it is. The vision is for a nation of mash-ups - developers beavering away to build applications that draw on this data to tell us, well, lots of important things that we can't yet anticipate.

The idea seems to have come about with frightening ease, according to Sir Tim in The Guardian:

"Gordon Brown said to me, 'How should the UK make the best use of the Internet?' and I replied that the government should just put all of its data on it," Berners-Lee recalled. "And he said 'OK, let's do it'."

Job done. Imagine how tempting it would have been to say, with the Prime Minister just waiting to carry out whatever you say, "Free porn for the over 75s". Or "A webcam in every home".

While I welcome the principle of open government and free access to public information, I can't help thinking this is symptomatic of the main problem with the Internet itself - providing a waterfall when all you wanted was a cupful. A quick search under "schools" will yield more data sets than you can possibly have thought existed: Post-16 participation in training in Wales; "Core Accessibility Indicators", sorted by school type; Cross Local Authority border movement of school pupils resident in England. In fact, my over-riding thought is not "hooray for openness" but "blimey, doesn't the government collect a lot of information".

Actually what set alarm bells ringing in my mind was the prospect of unlimited data sets for limited minds. In particular, the offhand remark by Sir Tim that it offered the chance "potentially to discover hidden patterns that may not be obvious from the raw information." Generally speaking, that is the sort of thing best left to statisticians, very clever people who understand things like data points, randomness, regression to the mean, biases, clustering, statistical significance and the Bonferroni Correction. I fear we are opening up the candy store to the idiot kleptomaniac children who will have capacity to waste an inordinate amount of everyone's time because they don't know anything about how stats work.

I may not know much about how statistics works myself, but I do know that data dredging will allow all the conspiracy theorist nuts, quacks and fanatics to convince themselves they have the evidence that backs up their claims about UFOs, telephone masts and fluoride in the water. I also know that, as a general principle, data collected for one purpose that is used to demonstrate another is often flawed - it's the oldest trick in the book if you want to fake evidence. Say you collect data to show a correlation between school attendance and smoking that gives nothing, but you notice an unusually high number of blond haired children showing up in the results. It's a short step from there to "proving" that blond children are more likely to take up smoking.

I don't predict this will give us any labour-saving apps anytime soon, since its use in creating hacked applications will pass by 99.5% of the population. But I do predict that it will be used by chancers wanting to get misleading, dangerous or malicious stories into the news with the weight of "evidence" behind them. The Daily Mail's health agenda is about to go nuclear.

16 January 2010

It's not a lie if you believe it....

At the risk of inviting some very boring spam, this week I've been getting inside the debate about homeopathy, following the new year campaign by the 10:23 organisation (www.1023.org). This is a collective organised by the Merseyside Skeptics Society, and fronted by Simon Singh, to raise awareness of the truth about homeopathy, and to discourage those retailers who should know better from selling it as a remedy for anything. As well as the occasional printed media outing, a lot of chatter across the web and Twitter has been generated, and not a few wounded homeopaths wondering why the big nasty doctors have started picking on them.

I don't propose to get into the detail of the debate, which has been conducted more eloquently by others, but the centre of the debate is one of efficacy and, in particular, the Placebo Effect. The demands of science are for randomised controlled trials whose results can be openly scrutinised and whose methods can be examined. Under such harsh light, homeopathy doesn't come off too cleverly, instead finding better favour as a list of "my friend's auntie took a homeopathic remedy and it cured her sciatica" referrals. But as they say, the plural of anecdote is not data, and bona fide trials demonstrate homeopathy performs no better than placebo. To which I would say - that's not a bad outcome, considering how mad the theory is.

Placebo is not just smarties to be given to hypochondriacs but actually a powerful and profoundly mystifying phenomenon. The Placebo Effect - the successful cure of a patient using fake medicine - can be quantified into degrees of effectiveness: certain colour pills work better than others, water injections work better than fake pills, injections and pills work better after a "consultation" than self-medication. The human mind's ability to convince itself an intervention is having a healing effect and then to enable that cure is astonishing - and far more interesting than quack potion theories. If only it could be relied on under all circumstances, NHS costs would plummet, and if I were a homeopathic practitioner, I'd take the result "no better than placebo" as a compliment.

Which brings us to Boots the Chemist, Britain's leading retailer of prescription drugs. Boots has been a focus of this campaign because it continues to sell homeopathic "medicine" despite freely acknowledging the lack of evidence for efficacy. Their defence is the products' popularity with punters, ignoring the apparent chicken-and-egg relationship that might create that popularity, as an organisation in whom vulnerable ill people place a lot of trust. Boots have been looking for a way to wriggle out of this scrutiny - being accused of selling treatments they know have not been proven to work. How do they continue to rake in the profits without damaging their reputation as an upholder of pharmacological best practice?

Here's how: they claim that, in order for homeopathic products' placebo properties to be effective, the consumers must believe in their efficacy (as long as they don't read the label too carefully, or do any basic research). And what better way to reinforce that belief than by having the largest high street drug retailer in the UK selling the products? If Boots sell it, it must be all right, and the placebo effect starts at the moment the punter queues up to pay. They could claim they would be threatening the homeopathic delivery of effective placebo treatment by not selling the products.

One thing preventing such a defence from succeeding, of course, would be the homeopaths themselves. Their shrieking insistence on acceptance and legitimacy by mainstream medicine and the wider general public would never let them admit any cures they do achieve arise from placebo - which could be the basis of a legitimate practice. But, again, they themselves are in a bind, because if they admit it's all just placebo, the Placebo Effect of their treatments would be diminished. So they are forced to strive for mainstream credibility, in order to keep up what successes they presently enjoy. It's like some hippy version of the Heisenberg Uncertainty Principle. Or, rather, The Emperor's New Clothes.

12 January 2010

Bottle of source

This week I finally received an answer to a question I posed to Basildon Council back in October, after a few gentle reminders, ending with an email to the Leader of the Council. It was a little whimsical, so I can understand their reluctance to answer, as it related to something I overheard being broadcast in their office reception while I was awaiting an appointment. A large video display was showing a series of public information films to entertain the great unwashed of Basildon about things like healthy living, council services etc, including an entertaining little movie about the dangers of binge drinking. It included the following memorable line:

Binge drinking can lead to: Alcohol Poisoning, High Blood Pressure, Liver Cancer and Sexually Transmitted Diseases.

To quote Sesame Street, one of these things is not like the others, which I thought a little disingenuous - so I called them on it. If I spend the rest of my days in a pub, boozing my life away (and the opportunity would be tempting sometimes), I may well die prematurely of high blood pressure and liver cancer, and suffer several bouts of alcohol poisoning. But unless they change the recipe for beer, I will not catch an STD, no matter how much I drink. It is like saying binge drinking may lead me to die in a car crash, because of the risks associated with drink-driving.

Basildon Council said they were highlighting a correlation and explained their motivation:

"From a communications point of view, utilising the impact alcohol can have on contracting sexually transmitted diseases is an impactive way of portraying the message to young people"

Got that? Untying the syntax of that sentence, I think they mean telling teenagers not to drink because it will rot their genitals is more effective than warning against a long-term risk, such as liver cancer. A valid point - teenagers think they will live forever, and can't see beyond the end of the next dole cheque, never mind 30 years hence.

However, the film did not seem to be aimed at young people, certainly not judging by its stars and its audience in that office (including me). It featured the type of "forgotten binge drinker" the government is keen that we unforget: middle aged delinquents like me who neck bottles of wine at the weekend. People who would no more attempt some casual sex after a couple of snifters than would try to drive a car, operate heavy machinery or host a Radio 2 show. If you start to draw in surrogate outcomes for binge drinking into what purports to be an information film, how far do you wander into areas where self-control plays a part?

How about if we said binge drinking leads to divorce, domestic violence, public nakedness and drowning? Hell, go for the jugular: binge drinking will kill you. If you strain the extrapolation and, most importantly, remove the agent from the action, you strain the credibility of your argument. To me there is a simple rule of thumb: if you want to be trusted more, claim less. It's actually an age-old sales technique - if people think they are getting more than they were promised, they trust you more and are more likely to purchase from you again. Now, where's that bottle of gin?...