By Jerome Burne
One of the alarming and intriguing things about the cholesterol lowering drugs statins is the vigour and ferocity with which supporters defend them. It’s alarming because it makes it almost impossible for both doctors and patients to get accurate information about their risks and benefits. Intriguing because it is so unscientific.
In fact, even though these drugs have been around for 24 years and prescriptions are still being written in their billions, we still don’t know for sure how well they work or how safe they are.
Researchers who suggest that they may not be as safe as routinely claimed have been accused of killing patients by frightening them into stopping taking the drugs given to cut their risk of heart disease. And it is not just researchers who are attacked; the integrity of a top journal has recently been questioned too.
At the end of last month Professor Peter Weissberg medical director of the British Heart Foundation made the remarkable allegation that critical articles about statins had been published in The BMJ (British Medical Journal) two years earlier had not really been intended to report on scientific research but simply with the aim of selling more copies.
Arch statin defender infuriated
He claimed that the journal was not interested in objective reporting. ‘The best way for them to sell copies, he told a newspaper ‘is by creating a bit of controversy,’
Regular readers will be aware that there has been a long running debate about the effectiveness and safety of these drugs. On the one side is the majority of the medical establishment which claims that nearly 30 large trials have shown that these drugs save thousands every year from dying of heart disease and that the side-effects such as muscle pain, a certain mental fogginess or cataracts are usually mild and only affect a small percentage of patients.
On the other are doctors and researchers who claim that the trials run by drug companies are unreliable, that large numbers of people have to take statins for one to benefit and that the side-effects are far more serious and common than admitted. A raised risk of diabetes, detected recently, is a particular concern.
This debate first received widespread media attention two years ago when The BMJ published two features on the higher rate of side effects. This infuriated arch statin defender Professor Sir Rory Collins of Oxford University who pointed out both papers contained an error of a couple of percentage points and demanded they be withdrawn. The BMJ published a correction but didn’t withdraw them when a committee concluded it wasn’t necessary.
An additional 2000 heart attacks and strokes
Sir Rory then went on a media offensive and claimed in print and on the airwaves that thousands would die if such research kept ‘creating misleading levels of uncertainty’. This was widely seen as highly over-dramatic way of correcting an error.
The pro-statin camp seemed to receive a fillip two weeks ago when a new study, also published in the BMJ, found that the widespread coverage of the raised side effect risk two years reported two ago – had coincided with over 200,000 patients stopping taking statins for a while.
The researchers at the London School of Hygiene and Tropical Medicine calculated that this could have resulted in an additional 2000 heart attacks and strokes over the next ten years.
The paper made no attempt to consider what benefits might have followed from stopping statins and so avoiding side effects. In effect it effectively assumed that statins are perfectly safe and effective – precisely what was being disputed. The study report also ignored the irony that the controversy Professor Weissman was complaining about was entirely due to the media blitz about the BMJ articles conducted by fellow statin supporter Sir Rory Collins.
Mixing up correlation with causation
If you didn’t know any of this background you might think that the London School study represented something of a vindication for the pro statin lobby in general and for Sir Rory Collins’ warning in particular. Sir Rory certainly appeared to see it in this light. When it came out he was reported as calling on the BMJ to: ‘stop hiding behind the spurious argument that debate is warranted’ and to ‘take responsibility for the harm it has caused.’
The claim that calling for debate over the safety of statins was spurious and unwarranted would be unexceptional in a faith based system but this is supposed to be about the science. While Sir Rory is scrupulous and unforgiving about an error of a few percentage points by his opponents, he appears happy to casually brush aside one of the pillars of evidence based medicine in defence of his own position. The pillar is the principle that correlation does not equal causation. In other words when two things appear together – such as firemen and fires – you can’t assume one caused the other.
This is precisely the objection that Sir Rory and others have used to question the evidence from observational trials of statins which find that patients report many more side-effects than the low rate that is officially claimed. They claim that only randomised controlled trials give you certainty about causation.
Yet the London School study is also an observational study. It looked at the period when the media reports on statin side effects were most frequent. Then it looked at the rate patients discontinued statins in the period. But Sir Rory assumes that one caused the other. As Gary Switzer, an expert in health care journalism wrote last week in an editorial on this latest study in the BMJ: ‘The authors provide no patient survey data to support the belief that people stopped because of news reports.’
Stopping statins less risky than taking them?
Presumably the hope behind the publication of the London School paper was that it would strongly discourage newspapers from writing about statin risks on the grounds that thousands would die as a result. But the main effect has been to raise yet more doubts about the benefits and risks of these drugs.
To begin with there was the formula that the London School used to calculate the number of deaths that would result from statin withdrawal. According to leading statin sceptic and Cheshire GP Dr Malcom Kendrick, the way heart attack risks are commonly calculated is highly unreliable.
‘Using an updated formula the number of heart attacks and stroke you’d expect over ten years from 200,000 people stopping statins is actually about 500,’he says. Now one of the things about balancing risks and benefits is that when the benefits go down the risks of side effects stays the same. What this means is the benefit from stopping goes up.
Kendrick illustrates what this means as far as your chances of developing one of statins’ most serious side effects – diabetes. A reliable estimate of the diabetes risk is that for every 150 people on statins one person will develop it.’ he says. ‘Using that rate to calculate the benefits from stopping statins reveals that there could be 4000 fewer case of diabetes if 200,000 people stopped taking them.’ Suddenly the claims that warning about the dangers of statins is killing people seems highly misleading
Statin supporters never actually tested their claims
But the irresponsibility of the statin establishment’s use of strong-arm tactics to close down any debate on side-effects doesn’t stop there. Two years ago Professor Collins claimed that reported side-effects such as joint pain and memory loss affected only a few percent of patients – not closer to 20 per cent as the articles The BMJ had just published were suggesting.
But as Dr Fiona Godlee, the editor of the BMJ has pointed out, there are serious problems with the evidence that Professor Collins relies on to dismiss this higher rate. Neither Sir Rory and his team nor any independent researchers has ever conducted any analysis of the rate of such side effects.
That’s because all the data from the trials on statins are not available for independent scrutiny, Dr Godlee said in response to the London School study. Statins have been subject to nearly 30 major trials run by drug companies and the full results – all the records of how patients responded, which is what you need for a proper assessment of side-effects – are held by an institution called the Cholesterol Treatment Trialists (CTT) in Oxford headed by Professor Collins. No independent researcher has ever seen any of this data.
‘This situation should shock people,’ said Dr Godlee. ‘It continues to shock me.’ Last year she was informed by CTT that not only had no outsider seen the data but that the Trialists themselves had never analysed the data for side-effects either. The CTT’s explanation was that this was not part of the original agreement that they had made with the pharmaceutical companies to hold all their data in 1995.
This makes it hard to understand how Professor Collins and the big heart disease charities were able to be so confident that statin side effects were not such a serious issue or that the benefit of taking the drugs always outweighed the risks.
This attitude among senior medical figures – that research and data that challenges an official position should not be engaged with but dismissed -is clearly not in the interests of patients. It could have serious repercussions very soon.
A new and more powerful cholesterol lowering drug has recently been given the green light by NICE. What will happen when studies come out indicating that side-effects are more serious than originally thought? Will that too be dismissed as ‘a spurious argument that debate is warranted?’