Why you should never trust new wonder drugs

The secret they revealed was that public statements about the safety of the drug were a lie; that the company knew Prozac was responsible for a raised risk of suicide and was only slightly more effective than a placebo.

Several years later I was faced with the secrets of another antidepressant - Seroxat.

No one outside the two companies, and few within them, knew what those boxes contained; I saw them because I was an expert witness in a court case.

Unfortunately, such revelations have since become all too common. Documents prised out of companies by American court cases have become the main way we have of discovering the truth about some of our best-selling drugs.

As well as Prozac, there have been another four or five other drugs exposed in this way.

Two years ago, the painkiller Vioxx was found to double the risk of heart problems; company documents revealed that the drugs sales team were instructed to keep this information from doctors.

Just before Christmas, internal documents suggested attempts were made to play down the serious health risks associated with the top-selling anti-psychotic drug Zyprexa.

Last week similar boxes formed the background to a Panorama programme showing how the company making Seroxat distorted evidence that the drug raised the risk of suicide in children.

The effect of these scandals is to show that something has gone badly wrong with the way our medicines are researched and sold.

The scientific literature, the very place doctors would look for a warning, contained barely a hint of problems. What's more, no one seems likely ever to have to answer for what appears to be fraud.

And no one in Britain has any means of finding out why their husband or child might have died.

The company boxes show the progressive demolition of a brilliant system for discovering how well drugs work: the clinical trial.

Developed more than 50 years ago, this involves giving a drug to one group of patients and a dummy pill to another, similar group.

Drugs such as antibiotics, hypertension drugs and the first treatments for cancer proved their worth in trials.

Until the 1980s, these were run in universities. Independent academics did the research and published the results. But for the past 20 years the drug firms themselves have increasingly run trials.

The result has been to undermine the value of the whole system. Now company employees analyse the data and medical journalists are hired to write them up for publication. Then Oxbridge and Ivy League professors are paid several thousand pounds to be credited as the authors of these articles.

This is something I've come across personally. Articles have been "ghostwritten" for me by drug companies, and then - when I refused to put my name to them - the same articles have appeared under someone else's name.

I have also seen company memos asking who would be the most "useful" academics to have as the authors of articles already written.

Once published in top medical journals, these confections will deliver more sales than any number of drug representatives, but all too often they are not accurate reports of the trial results.

In other organisations when evidence of disregard for public safety emerges, heads roll. But there have been no resignations following these drug disasters - barely a flicker of embarrassment.

The distortion and lack of corporate accountability makes me much slower to hand out new drugs. I even feel apprehensive if someone I know has to go to the doctor.

And what would I do if a loved one suffered a serious drug reaction? It's virtually impossible to sue a pharmaceutical company in Britain, partly because of the difficulty of getting any funding for such actions.

There's no point appealing to the government drugs watchdog - the MHRA. It is a small outfit entirely funded by the drug industry.

It has never taken any action against the academics who make fraudulent claims in ghostwritten articles, nor doctors working for the companies who repeat such claims, even when they have been shown to be untrue.

So the only other body to turn to is the General Medical Council, whose job it is to investigate the conduct of doctors - but it has shown no inclination to act.

But there is one simple thing we could all do to reclaim the clinical trials. People who participate as guinea pigs should insist the consent form states that the data will be available to all, and without this any drug tester should refuse to sign.

In doing this, we can change clinical trials back to a tool for scientific research and stop them being just a marketing exercise.

We will be told Britain will pay a heavy cost in lost jobs and breakthrough drugs if the industry is more tightly regulated.

But can it be any higher than the price we are paying in the deaths of husbands from heart attacks and children from suicide?