The public debate concerning ‘scare stories’ about statins is an interesting case study for the politics of peer review. It’s an important reminder that these seemingly technical issues of scholarly communication can have important public consequences. The case seems to be framed in the media as calling into question the ‘gold standard’ of peer review. Is this fair? I suspect it is but that the media coverage is inevitably somewhat simplistic.
It’s interesting to hear the rhetoric deployed to defend peer review and how obscurely, if at all, the institutional context figures into this. It makes ‘peer review’ into a timeless and placeless activity, with the inherent characteristics of the practice working to ensure the promised outcome. But if the context is only invoked by those attacking peer review, sometimes convincingly but often in a way I find rather worrying, it makes the ensuing debate weirdly one sided. It’s important to be able to defend the values of peer review while recognising that a shifting context should be grounds for recalibrating an existing practice.
The authors of two papers published by the British Medical Journal have publicly retracted statements they made about the frequency of side effects experienced by people taking statins, following a charge by an Oxford professor that the information was wrong and could endanger lives.
Prof Sir Rory Collins told the Guardian in March that a paper and a subsequent article in the BMJ were inaccurate and misleading. They had claimed that 18%-20% of people on the cholesterol-reducing drugs suffered adverse events. Collins called on the BMJ to withdraw them and complained that the authors were creating unease and uncertainty in British patients prescribed statins in large numbers to protect them against heart attacks and stroke.
“It is a serious disservice to British and international medicine,” Collins told the Guardian at the time, claiming that the alarm caused was probably killing more people than had been harmed as a result of the paper on the MMR vaccine by Andrew Wakefield. “I would think the papers on statins are far worse in terms of the harm they have done.”
The paper, by John Abramson and colleagues, questioned the decision to extend statins to thousands of people at low risk of heart attacks and strokes, saying that the drugs had not been proven to save lives in that group. They also claimed that an observational study had shown that 18%-20% suffer side effects from statins. An article by cardiologist Aseem Malhotra published the same week repeated the figure. Both authors have now withdrawn that statement.
In an editorial published in the BMJ, author Dr Fiona Godlee said the error was due to a misreading of the data from the study and was not picked up by the peer review process. “The BMJ and the authors of both these articles have now been made aware that this figure is incorrect, and corrections have been published withdrawing these statements,” she writes.
How would a publish-then-filter model be treated by the media? Would it undermine the authority of scientific publication? Would this definitely be a problem given that this authority is already being undermined by cases such as the BMJ one above? This seems an obvious transition to make, albeit with great caution, under conditions of fast capitalism. Unless the over-production of academic publication goes away (it won’t) I can’t see how the current peer review system is sustainable. My suspicion is that the time spent on peer review has declined precipitously given the broader shifts in the occupational demands placed upon academics. Publish-then-filter shifts the evaluative burden to a different sector of the working lives of academics.