My attempt to write as though I were headlines editor for The Sun notwithstanding, I have had an interesting evening.
I saw Ben Goldacre give a talk at the Babbage Lecture Theatre. By following his column on and off for years, and more recently his blog, I’ve come to respect him a good deal. Moreover, several discerning friends of mine saw him at the RI and recommended him highly, so I did not want to miss him.
The Babbage was packed. There were folks sitting on the stairs because all the seats were taken. Goldacre was introduced to the audience by David Spiegelhalter, who commented that the last talk he’d been to at the Babbage hadn’t been so well attended, and that the speaker on that occasion had been the Archbishop of Canterbury. Perhaps that puts Goldacre in the same class as The Beatles, at least locally.
In my notes, which follow, I have striven to condense Goldacre’s talk into a concise enough form for me to type it into my PDA. Here’s the inevitable disclaimer: I have paraphrased Goldacre, and I though I have tried not to misrepresent his views, it is possible that I might have erred. Please bear that in mind: these are my raw notes, not a word-for-word transcript of Ben’s talk.
Ben Goldacre – Babbage Theatre, University of Cambridge. 21/10/2008
How does the media promote public misunderstanding of science?
- Coverage.
– Wacky stories. Commisioned via mailing lists. Goldacre volunteered once, but then declined. E.g. “Jessica Alba has the perfect wiggle”; “All men will have big willies” (this especially silly, but received wide coverage); “Formula for the most depressing day of the year” (this one sponsored by Sky Travel); many other ‘formulas’. Newspaper editors wrestle with their science writers over this. PR companies strongarm the editors; editors strongarm the journos. But sometimes its direct: each time a product gets mentioned in a national paper, the sponsor sends the commisionee a cheque – even if the coverage was negative about the original story (e.g. If Ben covers it)!
–Miracle cures & sinister hidden scares. Short hist. Of medicine: rubbish b4 1930s. Great progress from ’30s to ’70s: lots of new drugs that worked, ICUs, CPR, inside-body scanners, artificial lungs, etc. All these ‘miracle cures’ developed during this period. But also during this period, many hidden dangers discovered too: tobacco’s carcinogenic effects, for instance. Since then, rate of revolutionary change has slowed. But media still stuck in 30s-70s mindset: miracles on one hand & hidden dangers on other. Also obsessed w/commodification of health. Yet when genuine studies cover this sort of thing, media doesn’t report, because the stories haven’t been pushed on them. In fact, it turns out one of the biggest “miracles” in modern medicine is the placebo effect, but because this doesn’t fit the media’s template of miracle substances, etc, it doesn’t get covered. Another problem is that science is covered as a series of breakthroughs rather than a long, slow process. This means that newspapers more likely to report ‘maverick’ science stories, which are more likely to be wrong.
–Risk data. Papers always use Relative Risk Increases – “you’re now a third less likely to get cancer”. A third less than what?
–Made up figures. “Cannabis is now 25 times stronger” (Independent 2007). Nonsense: we have measurements, and the claim doesn’t fit the data. This is really about people inventing data to underpin an agenda, such as an anti-drug agenda. Another example: MRSA “growing on the doorknob of the department of health.” Reported by Sun as “The UK’s top MRSA expert” – actually had no microbiology training, and couldn’t provide his samples. Turned out to be making the ‘experiments’ in his Garden shed. Yet Evening Standard, who had reported the story, nonetheless defended the story.
–How it is covered. E.g.: generalists not specialists chosen to write pieces; ‘authority figures’ preferred over evidence. Science as gossip rather than evidence-based assertions. But media’s choices of authority figures often dubious (e.g. Gillian McKeith, PhD – or to give her her full title, Gillian McKeith). What’s wrong with, say, authority figures in nutrition? (1) They portray food as the ultimate arbiter of health (it isn’t: the Multiple Risk Factor Intervention Project demonstrates this). (Incidentally, I’m going to offer t-shirts that say, “I think you’ll find it’s a bit more complicated than that.” & run competitions where people take pictures of themselves wearing them next to people wearing t-shirts with other political slogans. Hm, have I been talking for a long time? (Audience: yes.) Should I stop? (A: no.) Right then, so that’s the introduction finished! (A: laughter & applause.) (2) Extrapolating from lab or observational results to clinical ones. E.g. Patrick Holford: “AZT, the first prescribable anti-HIV drug, is potentially harmful, and proving less effective than Vitamin C.” This fails my AIDS test: say something silly about AIDS, and you’re an idiot. Turns out this was based not on a randomised clinical trial, but rather on a small test-tube trial.
—MMR/autism hoax. The big one. (Shows photo of Andrew Wakefield w/Dixie Chick.) Notice: Andrew Wakefield is at an opulent party in this photo. He’s privileged. Incidentally, given I support the Dixie Chicks’ anti-war, liberal stance, I’m disappointed they backed the hoax. Whenever liberal politics approach science, they tend to make conspiracy theories instead of informed critiques. Back to Wakefield. He made £500k consulting against MMR, & this disappointing ad hominem issue is why the story has been dropped from the media. But the truth is the money probably didn’t affect much. Newspapers should have investigated story better & that should have been the end of it. In fact, Leo Blair (Tony & Cherie Blair’s son) was a bigger factor in the story than Wakefield: Blairs wouldn’t divulge whether or not they had given him the vaccine, claiming it was a privacy issue (despite Cherie having described the shag that produced Leo, in her autobiography!). Leo Blair issue covered by generalists. Much unpublished (i.e. not peer-reviewed) research was quoted by papers, despite the fact that two accredited research papers were published on the same topic at the same time. These research papers comprehensively refuted the MMR-autism link.
There’s no sign this sort of thing is going to stop any time soon, which means I’ll never be out of a job!
Q&A
Q: What practical steps can you take to improve coverage of sci. In media?
A: Write to them or work for them. Ensure your press releases are clear & accurate; don’t use Relative Risk Indicators – use Natural Frequency Indicators. Blog about these issues. Make sure your homepage gives a concise account of what you research is – an intelligent 9-yr old should be able to follow it.
Q: How often do you get sued?
A: Only once so far, by a European vitamin salesman who took out ads in S.Africa advertising his tablets as a better cure for AIDS than anti-retrovirals. Other people have issued legal threats, e.g. Patrick Holford. But they haven’t been able to back them up.
Q: Are professers for the public understanding of science good value? Should we have more of them?
A: Well, some are, but they aren’t always. There’s a need for good communication about science, and professional scientists – and professional communications generalists – often aren’t sufficient. But I’d prefer to see scientists encouraged to spend a part of their time on promoting public understanding of their work, and supported in actually doing this effectively and not just in stupid childish ways, as all too often happens.
After the lecture, people queued up to speak to Goldacre and have him sign copies of his book. I don’t have a copy of his book, but I asked him to sign a piece of paper instead, which I could paste into the book when I got a copy, and posed to him a question along these lines: “In your professional opinion, would it be better to disabuse my mother, who uses homoeopathic remedies and Patrick Holford nutritional supplements, of her conviction that they are efficacious, or would it be better to let her reap the benefits of the placebo effect?” He replied that I should give her a copy of his book.
I pointed out that this suggested he had plumped for the first option. His response was that “There’s some evidence that the placebo effect survives even if you know that’s all it is.”
I reflected upon this on the way home. It sounds like the best of both worlds, doesn’t it? To know something doesn’t have any mechanism for helping you, but to take it anyway and to know that it might help you regardless. That’s having your cake and eating it.
Of course, giving her a copy of the book would also mean another sale for Goldacre. Following, like a good sceptic, Cicero’s question Cui bono?, it seemed worth investigating Goldacre’s last assertion to me a little further than simply taking it on trust. A quick web search revealed that Goldacre’s blog gets some of the highest results for searches about these topics. This post of his gives an example of the research he mentioned. But that example is from 1965, and it’s only one study, and it had a small sample size (15 patients). I’d want more evidence before I’d be entirely convinced that the placebo effect survives the knowledge that the medicament in fact possesses no specific medicinal qualities. Consequently, I am looking for more evidence. If I get it, I will buy Goldacre’s book for my mother. If not, well, I might buy it anyway, because it promises to be entertaining, and because it fits well with what I might as well call my creed, even though it’s a line from a Rage Against The Machine song:
If ignorance is bliss,
Then knock the smile off my face.
Ah, if you want to know more about placebo research, we wrote about Goldacre’s 2-parter on the placebo on Radio 4.
Some of this, may or may not be useful to you.
Nice synopsis of the talk.
Thanks for the link! I’ll be reading your write-up shortly
The pertinent part of the write-up seems to be the section that deals with a pilot study of open-label placebos in the treatment of ADHD. Fortunately, I have Athens access, so I was able to read the article in full. Besides the conclusion, the most interesting part of the article for me was the choice of scripts, which I give here.
Before I go any further, I should mention that I think that the study uses a poor choice of subject disease, because the medicalisation of attention-deficit is problematic[1][2]. Aside from this, the study is a reasonably convincing demonstration of the placebo effect (at least, upon first reading).
However, it’s a less clear-cut demonstration that a patient can know they’re taking a placebo and still benefit from it. This might seem like a strange observation to make, given that I’ve provided the scripts, above, in which the practitioners explained to the patients that the placebo contained . But that phrase is not the same as, , which would have eliminated the possibility that the patients benefited because they did not understand that the pill was free of active ingredients. After all, to a layperson, medications and drugs are somewhat magical, and may be considered to be supposed to be effective even if they are not ; alternatively, and perhaps more plausibly, the patients may have supposed the pills to contain active ingredients that were not considered to be medications or drugs: for instance, substances designed to slow the absorption or utilisation of the main dose by the body – a mechanism, as it is often described by vitamin pill companies (e.g. [3]).
Furthermore, as the form of words was different to that in the 1965 study (), the results of the two studies cannot be directly compared, as the issue of the patients’ potentially different interpretations of the different wordings is significant.
I saw Goldacre on Sunday after Robin Ince’s variety show at the Hammersmith Apollo, and asked him whether, besides the two papers I’d found (see above), he knew of any more articles about non-blind, patient-informed placebo trials. He said he didn’t.
[...] months later, I saw Ben Goldacre give a talk in Cambridge. At the end of his talk, he mentioned he would be on the line-up at an event in [...]