The most detailed and passionate statement of what has gone wrong with evidence based medicine and how it is damaging patients can be found in a new book called Pharmageddon by psychiatrist and campaigner Professor David Healy (published by the University of California Press, £27.95).
I wrote a summary of the book with David Healy, a shorter version of which appeared in the Daily Mail yesterday under his name. What follows is the more detailed original which contains more examples of how the distortion and cover-up happens.
Especially important is his explanation of the way the evidence based medicine principle is used to ensure that patient accounts of side-effects are dismissed as anecdotal and so not proper evidence. Combine that with clinical trials that are designed in various ways to downplay or ignore side-effects and it becomes impossible for doctors and patients to make a realistic assessment of the risks and benefits of any drug.
The published version appeared in the Daily Mail 24/04/2010[ Daily Mail permalink]
We are all swallowing an increasing number of pills, not just to treat disease but to keep us healthy. Even in a time of austerity the UK spend nearly 12 billion pounds on drugs and the total keeps rising. We believe science shows they are safe and effect because they have all been properly tested in clinical trials.
But this is a dangerous delusion. Our system for testing drugs is not fit for purpose and its failures are damaging the health of tens of thousands every year. Pharmaceutical companies spend billions conducting trials to come up with evidence for the benefits of drugs but they have a number of ways of making small benefits look impressive. Then as far as possible they play down or conceal damaging side effects. When patients do report adverse reactions they are described as anecdotal and dismissed as not proper scientific evidence.
“Pharmageddon” is packed with examples of drugs with serious side effects that the pharmaceutical companies have actively hidden from the drug watchdogs as well as patients. Recent cases include the anti-inflammatory painkiller Vioxx that caused heart attacks, the diabetes drug Avandia that also caused heart attacks and antidepressant SSRIs that raise the risk of suicide, especially in children.
Side effects brushed under the carpet
Clinical trials today are what might be called ‘results based medicine’. Their aim is to produce a result that can be used as a marketing tool. The original medical ideal – to find out if the drug was effective and safe – is at death’s door. Without being able to get proper data on what is going on – which you can’t do because either it has been manipulated, or hidden or never properly gathered in the first place – medicine is adrift on the high seas at the mercy of pirates.
It wasn’t supposed to be like this. Over fifty years ago following the thalidomide disaster – when a sleeping pill given to pregnant woman damaged children in the womb – a new scientific system of properly testing drugs before they were allowed onto the market was set up.
A key principle was that before any drug could be prescribed to patients trials should prove it was safe and more effective than an inert “sugar pill”, a placebo. What’s gone wrong is that at every stage of the process that was supposed to protect the patient, the commercial interests of the pharmaceutical industry have trumped the original sound scientific principles.
Drugs have been turned into just another commodity, supported by brilliant marketing of the benefits while problems and dangerous side-effects are brushed under the carpet. This is what has been called Pharmageddon.
In theory this shouldn’t be possible. Before a drug can be widely prescribed to patients it has to go through an intensive period of testing, first on animals then on a small number of patients to see if it is more effective than a placebo. Finally two larger trials, usually involving several hundred people treated for between three to six months. All the information from these is then passed on to the drug regulators in each country – the FDA in American and the MHRA in the UK – who then check it before issuing a licence.
Doctors threatened and vilified
So how have the drug companies managed to get round this system designed to protect patients? Hiding the truth about drugs can be done in a number of ways. The simplest is just not to publish unfavourable results. Another way is to publish negative trials but to make them look good. There are plenty of cases on record of senior consultants being paid handsomely to present the findings from trials in a favourable light. Then there is simple intimidation. Doctors who speak out about drug dangers can be threatened and vilified and journals who try to publish their work can be threatened with expensive litigation.
As a psychiatrist who has spent nearly fifteen years uncovering concealed evidence of how SSRI’s can cause patients to commit suicide I’m very familiar with all these techniques . I also have seen many examples of the ability of the drug companies and the drug regulators keep asserting there is no problem in the face of the most damning evidence.
In a system that was genuinely committed to patient safety, reports of side-effects would be carefully investigated. But when I uncovered evidence that the companies making the SSRI drugs for depression were hiding studies that showed a raised suicide risk, no one wanted to know. The drug watchdog said they were looking into it and did nothing, while medical journals which should have been committed to exposing such clear wrong doing were very cautious about reporting it.
For instance, an article for a science journal would normally be checked before publication by two experts. But when a leading psychiatry journal asked me to write about hiding results, they were so frightened of being sued they sent it out to ten people to review it. The British Medical Journal took over a year to sort all the issues their lawyers were worried about in another article of mine.
Of course publications have to make sure what they publish is accurate but fear of upsetting the drug companies who spend millions advertising in journals makes it far too hard to get this information out into the public domain where it belongs.
Ghost writing to hide poor resuslts
Fear of the drugs companies’ deep pockets even affected a magazine called the Index on Censorship, whose whole reason for existing is to expose just this sort of cover-up. I’d written a feature about the hiding of evidence for a link between suicide AND SSRIS. After many discussions with lawyers, the editor wrote to me saying “this is a hugely important subject and we should be covering it”. But the threat of legal action proved too much and the article never appeared.
Another technique of keeping unfavourable results under wraps is what is known as ghost writing. In theory, the scientist heading a clinical trial analyses the findings and the results are published with his or her name on it. Not so in the distorted version of evidence based medicine created by the drug companies.
We are talking here about the controlled trials which doctors and the DRUG REGULATORS rely on to tell if a drug is more likely to work for you or to cause you harm. However many that appear in top journals SUCH AS the New England Journal of Medicine have actually been written up “ghosts” employed by the pharmaceutical company that has a patent on the drug, and so far more likely to provide the results they want.
However the name that appears on the paper is often that of a senior consultant who is well paid for lending his name and authority to the work. We know this happens because company emails that have come to light discussing who would be suitable.
On the few occasions when other scientists have got to see the original results of the trial they were actually involved in there has been a shocking gap between what the trial had found and how it was written up.
One of these scientists was Dr Aubrey Blumsohn of Sheffield University who had been the a senior investigator on a trial of the osteoporosis drug Actonel. He objected strongly when he found his research has been written up by one of these ghosts. Although the trial was funded by Actonel’s manufacturers, he assumed that the final report would be an accurate report.
Side-effects airbrushed out
However in the ghosted paper tables showing his results had been heavily cropped, losing much of the original data. When this data was included, the claims being made for the drug did not stand up. Blumsohn took his name off the paper, a move that ultimately lost him his job with Sheffield.
SO how do you know that the trials showing that the drug you are taking to cut your risk of broken bones hasn’t been tampered with to make it look more effective than it really is? You don’t. You also don’t know what side-effects have been airbrushed out of the trial results either.
Ghost writing isn’t the only way of concealing signs of harm. Another is by reclassifying patients. All trials have patients who drop out – could be for personal reasons or could be because they couldn’t stand the side effects. But they are classified as being non-compliant; that is they didn’t take their drugs. No need to say what their side effects were.
American court cases, when full data from trials has to be released , have revealed an even more blatant technique. In some of the clinical trials of SSRI drugs when a patient getting the drug commits suicide that can be listed as a suicide that happened in the placebo group.
Switches like these aren’t spotted because the drug regulators who are supposed to check such things don’t usually get to see the raw data. Instead they make decisions licensing a drug based on summaries of what happened that have been complied by the company that is applying for the licence. In the summary evidence of the swap will have vanished.
But there is an even more fundamental reason why doctors and patients can’t get reliable information about side effects from supposedly “scientific” clinical trials. It’s not just that signs of dangerous side effects are downplayed or hidden, the trials aren’t designed to find these problems in the first place.
Patients are effectively guinea pigs
The reason is that the trials used to licence drugs only run for a few months with, at most a few thousand people. That’s not enough time or numbers to detect rare but serious side effects when millions of people take a drug for years.
All drugs are toxic, the important thing is to do proper trials to find out if the benefits outweigh the risks. But if the system for spotting risks is so open to abuse and cover up, how can the regulators may a proper judgement that a drug is worth licensing? If you are put on a drug that has just been released you are effectively a guinea pig waiting to find out what the real risks are.
A clear example of this failure showed up last year in a big review of the effectiveness and safety of cholesterol lowering statin drugs carried out by the prestigious Cochrane Collaboration. It found that half the trials it looked at hadn’t made any attempt to report on adverse events at all. Even such important ones as the possibility, now confirmed, that the drugs make you more likely to develop diabetes.
Just how much the current system allows drug companies to accentuate the positive and eliminate the negative – showed up clearly in the way a heavyweight tranquilizer called Zyprexa(olanzapine in the UK) was promoted. Zyprexa was licensed on the basis of four clinical trials involving several thousand patients.
However the company was able to carve up these results into separate articles that appeared in no fewer than 234 medical publications all saying how effective this drug was. This gave the impression that the evidence for its benefits was overwhelming when in fact it was just the same data being repeated. But this had the desired effect and sales soared.
This blizzard of positive articles was totally silent about the potentially deadly side effects of the drug, even though it was known to raise the levels of glucose and cholesterol in patients’ blood –increasing the risk of diabetes and heart disease – as well as being linked with suicide.
Catch-22 disappears side effects
The way such skewed promotion has a direct and damaging effect on patients showed up clearly a couple of years ago as a result of an investigation by the Institute of Psychiatry in South London. The study found that Zyprexa and similar heavy tranquilizers had been given to 200,000 dementia patients annually and been responsible for 1800 early deaths a year.
What goes on with adverse events in trials and afterwards is similar to what happened to subprime mortgages before the banking collapse. The companies cut the risk up into small bits so that overall it looks safe. So it comes as huge shock when a serious danger does emerge in big independent trials.
What is so totally unscientific about the current way of testing drugs is that the companies own the data the trials have produced. Until it is freely available for independent researchers to examine – allowing other scientists to check your data is a fundamental scientific principle – all these abuses are likely to continue.
But changes in regulation take a long time . What’s needed is something that can make a difference now. That’s why I’ve set up a website which will give patients and doctors a more powerful voice in reporting side effects. As we’ve seen clinical trials are very poor at picking up most of the side effects that are likely to affect patients once the drug is widely used.
But once a drug is on the market a catch 22 swings into action. Clinical trials count as scientific evidence. So companies can say that there is evidence that the drug “is well tolerated”. However patients own reports of side effects, once a drug is being prescribed, only counts as anecdotal evidence, not hard scientific evidence.
The site will be a way round that catch. It will contain a lot of data about side-effects of different drugs and we will be collecting patients’ reports in a way that gives a much better picture of the effect of the drug on their lives.
Imagine you’ve reported a symptom and then both you are your doctor both get letters saying that a thousand other people have reported a similar problem and a hundred doctors have agreed there could be a link between the drug and the symptoms. At that point it’s just not reasonable to keep saying these reports are not scientific and that there is no hard evidence that this is really happening.
If enough people start contributing the effect of sharing side-effect data much more widely the effect could be dramatic. It should start to show what other drugs or treatments people need to treat the side effects of taking a drug in the first place. At the moment we have no idea of the cost of follow up treatments – it could be huge.