A devastating and definitive 20-year study “demolishes the claim that cannabis is harmless”, said the Daily Mail last week. Which is rather worrying, since 13.5% of adults aged between 16 and 24 took cannabis last year. The study quoted is actually a narrative review of the evidence in the journal Addiction by Professor Wayne Hall from the University of Queensland. Hall says that the euphoria and increased sociability induced by taking cannabis are mostly produced by delta-9-tetrahydrocannabinol (THC). Over the past 30 years, the THC content of cannabis has risen from less than 2% to 8.5% in the United States, and probably in other developed countries. Even so, it is extremely unlikely that overdosing on cannabis can be fatal. Rat studies suggest you would have to take more than the heaviest user could manage in a day to be at risk.

Hall says that driving while “cannabis-impaired” can double your risk of accidents (consuming alcohol increases the likelihood of crashing between six and 15 times). He finds that one in 10 regular users develops dependence, increasing to one in six if they started in adolescence. This rate, however, is much lower than in those taking heroin (23%) or regularly drinking alcohol (15%).

Regular (almost daily) cannabis use in adolescence doubles the risk of cognitive impairment. One study showed an eight-point drop in IQ with heavy use between 13 (before the drug was used) and 38 years of age. A link with psychosis is most likely in people with a genetic susceptibility to it. In this group, the risk of developing psychosis could rise from 10 to 20%. Hall’s best estimate from the research is that in the general population, it would take 4,700 young men aged between 20 and 24 to be persuaded to remain cannabis-free in order to prevent one of them getting schizophrenia.

So, is this research review enough to make us more wary of cannabis? Should you stop smoking it, or start sniffing your teenager’s clothes after a house party?

The solution

Cannabis use is falling among 16- to 24-year-olds – it was at its highest in 1998, when a quarter of this age group used it. Hall’s review is well written, but can’t solve the underlying problem of much cannabis research – that people who smoke the drug may be fundamentally different from those who don’t. It is therefore hard to say what is cause and effect. The review is also not clear on whether it has included all relevant studies – and the observational studies the evidence is based on are not robust, because they can only suggest associations. But the associations might suggest that cannabis is not harmless, only less harmful than other things people take to become sociable.