Random drug testing in schools may be pushing students away from marijuana and on to harder substances that are more difficult to detect, research has found.
Using data from more than 150,000 high school students (aged 14-18) at more than 1,400 US schools, researchers found that institutions that ran random drug tests saw marijuana use decrease but the use of other drugs rise significantly.
About 28 per cent of American high school students are now subject to drug testing. In recent years, the introduction of drug testing has been suggested by governments in countries including the UK, Australia and Sweden.
"I think school leaders should realise that drug testing does not have a simple association with student drug use," said Yvonne Terry-McElrath, from the University of Michigan's Institute for Social Research, one of the authors of the study.
"It is likely that a drug-testing programme would associate with some degree of decrease in marijuana use but also with some degree of increase in use of other illicit drugs," she said.
At schools where all students were subject to random testing, the research found that the probability that a teenager had used marijuana in the past 30 days fell from just over 25 per cent to less than 21 per cent. But the chance of their having used other drugs rose from just under 14 per cent to over 16 per cent.
Researchers said students may know that traces of marijuana stay in the body for longer than most other drugs, so they may choose substances that are less likely to be detected or that are not included in the tests.
In the UK, former prime minister Tony Blair called for the introduction of random testing in 2004, but education officials have not pursued the idea, trialled in a pilot programme.
The trial was led by Peter Walker, who introduced the first state-school drug tests at the Abbey School in Kent, where he was head at the time. Mr Walker dismissed the latest study and called for random testing to be introduced. He criticised the Department for Education for leaving his guidelines on testing in schools "on the shelf", saying that opposition to testing is based on a fear of bad publicity.
Drug testing is widely used in the private school sector. A 1999 report by the Headmasters' and Headmistresses' Conference, which represents more than 250 schools, estimated that 75 per cent of boarding schools used drug testing, but not at random.
Tony Little, headmaster of Eton College, told TES: "We don't use random tests because it undermines trust in the community, in the same way we don't use CCTV to spy on students." But he said that testing for students under suspicion of using drugs, instead of moving directly to punitive measures, had encouraged other students to report their concerns about classmates.