The long running spat between senior statin advocate Professor Sir Rory Collins and the British Medical Journal has come to a very satisfactory conclusion. His demand – that two papers challenging his claims about the safety and effectiveness of statins – be withdrawn, has been rejected by a committee specially set up to consider it.
Instead the report, out today, makes several low key but sharp criticism of the way Sir Rory has been dealing with his data. (For more background details on what has been going on see here – Eminence medicine defends status quo and here – Statin debate the ultimate two minute guide)
Sir Rory had been particularly incensed that both the BMJ articles had challenged his claim that statins were effective in healthy patients and caused very few side effects. His specific complaint concerned a minor error in reporting the percentage of side effects contained in another article. This was rapidly corrected
The two authors he was attacking – Dr Aseem Malhotra and Dr John Abramson – had their own concerns about Sir Rory’s work. Notably that his conclusions were based on placebo controlled trials of the drugs run by the drugs companies that kept the data they had collected hidden.
The secret statin data
Since the published results of drug company trials are notoriously unreliable, it’s now widely agreed that independent researchers should be able to look over the data that had been collected on each patient.
The research centre that Sir Rory heads – the Cholesterol Treatment Trialists (CTT) in Oxford – holds the biggest collection of statin data in the world, but no one outside the organisation has been able to study it for 20 years – CTT the house of statin secrets. The committee’s report made it clear that this was no longer acceptable.
‘It is very clear,’ it said ‘that statin trial data is not available for assessment,’ going on to note that this ‘may contribute to uncertainty about the risks and benefits’ of the drugs. That was a damming point since the basis for Sir Rory’s demand for a retraction was that questioning his results from randomised controlled trials was killing people by stopping them taking the drugs. In the committee’s view the issue was still uncertain.
The report picked up on another example of Sir Rory’s eccentric approach to evidence and to normal scientific debate. It noted that despite being asked to write an article rebutting Malhotra and Abramson’s claims and to participate in the rapid responses to the articles, all his correspondence arrived marked ‘not for publication’. ‘This was unlikely’ the report drily observed ‘to promote open scientific dialogue.’
Evidence based medicine not delievering
But there is a much bigger issue here than Sir Rory’s ‘grand old man’ style. While he behaved as if the RCT was an unimpeachable source of information, what has emerged from this battle is that the evidence based medicine project is in intensive care if not already expired. Statins are the most widely prescribed drugs ever and quite possibly the subject of more RCTs than any other.
Yet here we are 20 years on and there is widespread agreement that we still don’t really know how effective they are at preventing heart attacks in healthy people – the group who get by far the most statin prescriptions – or what the true side-effect rate is. Part of the problem is the drug companies’ well-known habit of fiddling of statistics, hiding of unfavourable results, selecting trial subjects most likely to produce favourable results and so on.
The unreliability of RCT’s has recently been acknowledged by one of evidence based medicine’s most enthusiastic supporters – Dr Ben Goldacre. “…We lack reliable information from randomised controlled trial on common symptomatic side effects of statins,’ he wrote in a BMJ editorial last month.
This is something of a volte-face for Dr Goldacre since it was only a few months earlier that a research paper with his name on it concluded that statins didn’t really have any side effects at all; reports saying they did were largely due to a sort of reverse placebo effect – people expected pains and other problems so sure enough they felt them.
A tarnished gold standard
In fact Dr Goldacre has become something of a flip-flopper. It was only a year ago that he wrote a good book – Bad Pharma – that explained in great and forensic detail why results of commercial drug trials couldn’t be trusted because of drug company fiddling.
Yet his statin side effect trial was based on entirely on results from commercial trials. So Dr Goldacre who spent years jeering at any and all non-drug treatments for their lack of RCTs, now admits that even several dozen RCTs can’t decide on the effectiveness or safety of a single drug. It’s a gold standard that is looking fatally tarnished. The debate over how best to fix it wil be covered in another post.
RCT’s unreliablity is particularly unfortunate for patients on statins since waiting in the wings is a new generation of super statins that lower cholesterol even more powerfully and are expected to cost around 20.000 dollars a year – Beware new cholesterol drugs coming. How are we going to decide if they are worth it?
Meanwhile we may be on the verge of having a firm decision about one of statins’ long terms effects – type 2 diabetes. Nearly a thousand women have so far begun legal proceedings against the pharmaceutical giant Pfizer, claiming that their best-selling cholesterol lowering statin Lipitor caused them to develop diabetes – Thousand woman lipitor lawsuit
If they win, and there is considerable evidence to suggest there is a link, then NICE’s decision to recommend the drugs for another four million people or so, along with Sir Rory’s iron clad confidence that there are no side-effects worth worrying about, will look careless, arrogant and dangerous.