Science bad and good: a review

Whether specifically applied to science or not, books that educate about rational thinking, and critique irrational/dishonest thinking and its practitioners, have a very long pedigree. Robert H Thouless' classic Straight and crooked thinking springs to mind, as do Darrell Huff's How to Lie with Statistics, Martin Gardner's Fads and Fallacies in the Name of Science and Science : Good, Bad, and Bogus, and Michael Shermer's Why People Believe Weird Things: Pseudoscience, Superstition, and Other Confusions of Our Time.

There's always room for topical reanalysis, however, and this is the thrust of Ben Goldacre's Bad Science, the spinoff book from his Guardian column of the same name. It concentrates mostly on bad science in relation to UK-based medicine and health: partly because Goldacre is a UK doctor; partly because this is a major arena of conflict between science and what's pejoratively called "woo". The material will be familiar to regular readers, who probably follow the Bad Science website too. But the newspaper pieces are necessarily short, and easily read as a series of loosely connected "whack-a-mole" episodes. The book corrals the moles, so to speak, drawing together and analysing recurring themes so that (in the words of Sir Iain Chalmers, Founder of the Cochrane Library) you can "become a more effective bullshit detector".

A brief tour by chapter: 1) "Matter" (detox methods). 2) Brain Gym. 3) The Progenium X-Y Complex. 4) Homeopathy. 5) The Placebo Effect. 6) The Nonsense du Jour (primarily about claims of nutritionists). 7) Dr Gillian McKeith PhD. 8) 'Pill Solves Complex Social Problem' (the Durham fish oil trials). 9) Professor Patrick Holford. 10) Is Mainstream Medicine Evil? 11) How the Media Promote the Public Misunderstanding of Science. 12) Why Clever People Believe Stupid Things. 13) Bad Stats. 14) Health Scares. 15) The Media's MMR Hoax. 16) And another thing.

The first three chapters start with the fish-in-a-barrel stuff as an appetiser: claims containing factual errors that are trivially debunkable (e.g. "detox" methods that demonstrably make the stuff claimed to be extracted from the body, or Brain Gym's claim that "processed foods don't contain water"). Then Goldacre gets to "the meat", first using the currently controversial homeopathy as focus to introduce some core tools for deciding if a treatment works - blinding, randomisation, and meta-analysis - before moving on to an extensive discussion of the strength of the placebo effect.

The next three chapters move on to the claims of nutritionists, beginning with general problems - such as cherry-picked data, invalid extrapolation to humans from test-tube results, and outright invention - then analysing the claims of Gillian McKeith and (via the Durham fish oil trial) Patrick Holford as modern, contrasting examples (one theatrical, one scientific in style) of a long-standing type of media health guru. Goldacre is not exclusively batting for the mainstream medical side; chapter 10 covers the similar and varied ways the pharmaceutical industry massages data to promote particular drugs.

The final half a dozen chapters attempt to identify causes for the general mess, and Goldacre points to two factors. One is human cognition (as a side effect of cognitive mechanisms dazzlingly successful in rapid processing of our world, the human brain - however clever - simply is wired to be hopeless at analysing statistics and other complex data). The other is the media, where standard story formats such as "formula for the perfect <whatever>", "maverick against the system", "miracle cure" and "hidden scare" almost always misrepresent science. The Daily Mail’s "ongoing mission to divide all the inanimate objects in the world into those that cause or cure cancer" gets a mention, as does the spate of tabloid MRSA stories based on tests by a completely unreliable expert, the late Dr Chris Malyszewicz. Another media story, discussed in detail, is the recent MMR/autism scare, where such factors produced a national media-propagated health-endangering myth that persisted long after the peer debunking of Dr. Andrew Wakefield's minority view. The book finishes with an exhortation for scientists to get involved, and not to get suckered into media-distorted versions of their work.

I admit I'm a regular at the Bad Science forum, and long since sympathetic to Ben Goldacre's view of things; it's hard to see the book as has having much surface appeal to enthusiasts of alternative medicine. However, I strongly recommend it to those readers, as Goldacre's approach is not as antagonistic as might be expected. Unlike the black-and-white pro-science authors of many books of this sort, he's thoroughly open to finding territories in common. For instance, he views "detox" procedures as a manifestation of an ancient, human and positive form of psychological cleansing ritual, that only becomes scammy when pseudoscientific fixtures are bolted on. Likewise, he regards the placebo effect as a fascinating, powerful and positive effect; again, only wrong when used unethically, such as prescribing a placebo for dangerous conditions such as malaria where it as no effect. His strongest criticisms of alternative medical belief systems are in areas such as their general hostility to evidence-based procedure and critical self-appraisal, and the egregious habit of chilling factual criticism by legal threats (expect a future out-take, removed from the book pending now-settled legal action, on exactly this point in relation to Matthias Rath).

Lack of critical self-appraisal applies also, of course, to newspaper and television; Bad Science mentions, for instance, various media refusals to reevaluate their MRSA scare stories even when eminent microbiologists pointed out problems with the methods of Dr Malyszewicz. The media is the main villain of the piece, with its immense power to influence public perception, coupled with its entrenched capacity for failure to 'get' science (perhaps due to persistence of the syndrome of CP Snow's 'Two Cultures' - newspapers sideline their specialist science writers, so front page scientific/medical stories are written by non-scientists). The book's overall flavour is cheerfully acerbic, but shot through with a sympathy for the human condition. People, in Goldacre's view, are emphatically not stupid; but they make better decisions about their health when not wilfully misinformed. Bad Science attacks those misinformers, not the believers.

Bad Science, Ben Goldacre, Fourth Estate, 2008, ISBN-10: 0007240198 ISBN-13: 978-0007240197.

- Ray