Jeremy Lawrence: We've known the risks for years. So why is Clarke rethinking now?

Charles Clarke's second thoughts on cannabis are puzzling. The two studies, from Holland and New Zealand - which apparently prompted the Home Secretary's rethink - suggest that regular use of the drug increases the risk of psychosis in vulnerable people.

Charles Clarke's second thoughts on cannabis are puzzling. The two studies, from Holland and New Zealand - which apparently prompted the Home Secretary's rethink - suggest that regular use of the drug increases the risk of psychosis in vulnerable people.

But they merely confirm what was already known about the drug. They increase the weight of evidence against cannabis, but they do not point to a higher order of risk.

It has been known for a century that heavy doses of cannabis induce hallucinations. The drug stimulates the brain to increase the production of the chemical dopamine, an excess of which is the hallmark of schizophrenia.

Any drug that increases the release of dopamine will therefore worsen the symptoms of schizophrenia. What has been unclear is whether cannabis only triggers psychotic attacks in someone who is already ill, or can cause schizophrenia to develop in someone who was previously well.

At least five studies published before the Government downgraded cannabis from Class B to Class C last year had shown people who used cannabis were at higher risk of mental illness. Research in Sweden and the Netherlands revealed that regular consumers were up to six times more likely to have psychotic episodes than non-users.

What appears to have alarmed Mr Clarke is the latest finding by a team of Dutch researchers from Maastricht University, published in the British Medical Journal last December, which showed that frequent users of cannabis who were psychologically vulnerable to its effects were at high risk of suffering a psychotic episode needing hospital treatment.

Up to 10 per cent of adults - about four million people in Britain - are estimated to have a tendency to paranoid thoughts or grandiose delusions, and could be tipped into psychosis by the drug. The researchers led by Professor Jim van Os studied 2,437 young people aged 14 to 24. They found half of those who smoked cannabis regularly and had a pre-existing risk of psychosis developed psychotic symptoms over the four-year period of the study.

The finding prompted British experts to warn that the growth of cannabis smoking among schoolchildren could trigger an epidemic of psychosis.

Professor Robin Murray of the Institute of Psychiatry said: "It may be to do with how early you start. The earlier it is, the greater the risk." Surveys suggest the average age at which people start smoking cannabis has come down since the 1970s. In the UK, two out of five 15-year-olds say they have tried cannabis, more than anywhere else in Europe.

Further evidence confirming the Dutch findings has now come from New Zealand, the second study cited by Mr Clarke, where a survey of 1,000 young people has found that those who used cannabis every day were nearly twice as likely to suffer psychotic symptoms as non-users.

Daily users were between 60 and 80 per cent more likely to experience symptoms such as hearing voices, paranoid thoughts and feelings of isolation. Professor Jim Fergusson of Otago University, who led the study published in the international journal Addiction, told the New Zealand Herald: "These are not huge increases in risk and nor should they be, because cannabis is by no means the only thing that will determine if you suffer these symptoms."

He added that the debate over cannabis had become polarised between those who believed it caused serious harm and those who argued it was harmless and should be legalised. "I think the truth lies somewhere in the middle," he said.

Cannabis is still widely accepted to be less damaging than the Class A drugs heroin and cocaine - and less damaging than the legal drug tobacco.