Just one is all it takes. For the impressionable politician, a random drugs testing programme in one school soon turns into a habit-forming demand to try it across the country.
As we reported yesterday the pilot scheme at the Abbey School in Faversham, Kent, is being extended to all secondary schools in the county.
Peter Walker, the school's newly retired headteacher, is enthusiastic about the effect of random testing and is acting as ambassador for the scheme. He believes it should be rolled out across the UK.
When the Abbey school scheme started critics were concerned that children who refused a test would be assumed to be guilty and mistakes could wrongly label some as users.
In the event after 600 random drug tests were carried out on the school's 11 to 18-year olds last year, only one tested positive for cannabis.
Mr Walker said: "We had our best set of exam results in the school's history. There's less disruption in the classroom, less incidents in the playground or on the way to school. Children feel that they are far better protected.
"The biggest reason for taking drugs is peer group pressure. It looks like we may well have found a way for children to have a viable way of saying no to their peer group."
But doubters remain - and one of them, Professor Neil McKeganey, of Glasgow University's centre for drug misuse, has the job of evaluating the programme in Kent as a possible prelude to going national.
He will be keeping an open mind but the concerns he raised in an earlier study for the Joseph Rowntree Foundation are still valid.
He argued that random testing could have "unintended adverse consequences", by encouraging youngsters to switch from cannabis, which can be detected for a relatively long time in the body, to drugs that cannot be traced after a few days, such as heroin and ecstasy.
When he wrote that report last year he said a study from the US, where drug-testing programmes are widespread and and have strong political support, revealed they had little impact on drug use among young people.
"In light of this it would seem preferable to avoid the ad hoc proliferation of random drug-testing programmes until such time as there are clear data on effectiveness," he said.
Now it seems Professor McKeganey will get the chance to judge whether random drug testing really is effective.
But politicians are junkies for this kind of high profile intervention. Earlier this year Ruth Kelly, the then education secretary, praised the scheme. Maybe the Kent pilot is "doomed to succeed" whatever the results?
