It’s only when you line these jokers up side by side that you realise what a vast and unwinnable fight we face.There was the miracle pixie dust which made a man’s fingertip grow back, although fingertips do just grow back by themselves. The Telegraph’s ever-reliable health correspondent reported that red wine prevents breast cancer – with the flimsiest of nutritionist-style evidence – just two months after writing that alcohol causes breast cancer (the latter is more correct).
We saw the Sunday Express claiming on its front page that an impressive government adviser called Dr Roger Coghill had performed scientific research, and found that the Bridgend suicide cases all lived closer to a mobile phone mast than average: this was an issue of great public health significance, but when I contacted the researcher, he wasn’t a doctor, he wasn’t really a government adviser, he couldn’t tell me what he meant by “average”, and he had, in a twist of almost incomprehensible ridiculousness, “lost” the data.
He wasn’t alone. Esure and Mischief PR refused to hand over their data on vermin and bins for inspection, although it had been reported credulously in every national newspaper. I got a leaked copy, and it was rubbish. Citigate PR refused to hand over the data on their Carbon monoxide and council flats story until I raised a fuss.
In a world where rigorous evidence from scientific research languishes unpublicised, the media continued to churn out bogus wacky science stories. Britain’s happiest places were mapped by “scientists”, although the differences were just chance findings; there were innumerable “surveys” from unrepresentative populations; and the right wing press claimed that “Lord Nelson and Captain Cook’s ship logs question climate change theories,” although they did nothing of the sort, as the researchers themselves helpfully explained. We saw how the BBC misrepresented the statistics on parents’ choices about keeping a Down’s Syndrome pregnancy, producing their a publicity avalanche on the back of an incorrect story, and learnt along the way about confounding variables, baseline changes, and more.
In the world of evidence based social policy we saw how the government quietly dropped death as an outcome indicator for their drugs policy, the fascinating inconsistencies in food additive judgment calls, and more. We also watched with delight as right-wing thinktank Reform produced a report on the crisis in maths in which they got their maths wrong.
The pointless formulae stories continued unabated. The “Fame Formula” media frenzy was triggered by the Guardian itself: it wasn’t just mathematically stupid, it demonstrably failed to model reality. People like to say that actually you need to be really clever to write for a tabloid (“actually”) because it’s the hardest job, although nobody at the Sun spotted that their Cambridge mathematician’s Britney boobline equation (“0×70x(20×5+32)/75”) gave an answer of zero, not 123.2.
It was an interesting year for the drug companies, with most of our fun revolving around selective non-publication of unflattering data. The SSRIs fared especially badly, with repeated studies showing that evidence of non-superiority over placebo was left unpublished, as was evidence of potential harm. We saw how the drug company Lilly have published strikingly similar data on duloxetine – a new-ish antidepressant drug – twice over, in two entirely separate scientific papers.
We saw how the people running the ENHANCE trial were really rather slow to publish their results, and altered their chosen endpoint after the experiment was finished. The same thing was happening with cancer trials, where researchers showed that only one in 5 cancer trials actually gets published at all (and only 5.9% of industry-sponsored trials, but in those 5.9%, golly did they do well: 75.0% gave positive results…).
Regulation has unforgivably failed to deal with these simple problems, but in a spectacular episode of collective point-missing, at the same time we saw how ethics committees have now made trials so administratively cumbersome that only multinational corporations can perform them.
Other repeat offenders continued to churn out good comedy. The Dore “miracle cure” for dyslexia invented by a paint entrepreneur called Wynford Dore was puffed throughout the media (including Radio 4’s investigative consumer slot You and Yours), until it turned out they’d gone bust, leaving some very distressed customers, at which point journalists suddenly decided they agreed with me about the dubious evidence.
The comedy factory of the Durham County Council fish oil “trial” struggles on. In March they announced – in defiance of everything they had said on the subject for several years now – that there was in fact no trial on childrens’ performance, and they had never intended to release results. In September they released the results. They had analysed their data with such unnecessarily laughable incompetence that the results can only reliably be interpreted as a false positive.
The media continued to mischievously misrepresent the evidence on MMR, ten years on, and lest we forget, vitamin pill entrepreneur Matthias Rath dropped his 15 month libel case against me and the Guardian. We saw quacks in universities and a TV nutritionist who wound up in court after her client wound up in intensive care.
Most importantly I was allowed to sneak onto the news pages of a national newspaper carrying explanations of absolute and relative risks, numbers needed to treat, publication bias, confounding variables, the counterintuitive maths on screening programmes, genius research into the placebo effect and irrationality, corrections for multiple comparisons, selection bias, cumulative meta-analyses, clinical trial methodology and more. For that I salute and adore you all.