This means that serious psychiatric diseases that might not have shown up until kids were in their teens or twenties - or might never had developed at all - are starting in children as young as 12 who smoke marijuana.
The link between using pot and developing serious mental illness is strongest in the youngest smokers, 12- to 15-year-olds, or kids even younger, said Dr. Matthew Large in an interview with Reuters Health.
"We have to (tell) people who have marijuana in their pockets not to give it to younger people," said Large, who headed up the research at the University of New South Wales in Sydney, Australia.
Large and his colleagues looked at thousands of patients with psychotic disorders like schizophrenia. People with psychotic disorders lose touch with reality - usually starting in adolescence or young adulthood.
The authors of the new study found that in the subjects who had been pot smokers, the psychotic symptoms began nearly 3 years earlier than in those who had not been marijuana users.
The vast majority of young people who use marijuana don't develop psychosis. And so far, no one's been able to prove that smoking marijuana actually causes psychotic disorders like schizophrenia, but the new research adds to "growing evidence" that it does, at least in some people, said Dr. Michael T. Compton at George Washington University Medical Center in Washington, DC, who worked on the study.
A number of studies had already suggested that people develop schizophrenia at a younger age if they've been using pot - even if they weren't heavy users - but not all researchers agreed.
To get a better sense of the evidence, Large and Compton and their colleagues systematically combined and analysed data from 83 studies involving more than 22,000 people with psychotic disorders like schizophrenia: 8,167 who used marijuana or other substances, and 14,352 who did not.
"Those who used marijuana had an earlier age at onset of the disorder, by (about 32 months) on average, than those who had not used marijuana," Compton said in e-mail to Reuters Health.
Why was there disagreement in the past over whether this effect really exists? The reasons lie in the way the individual studies might have been done, the researchers say in their article, published in the Archives of General Psychiatry.
In their larger analysis, they made sure to account for several factors that authors of the individual studies might not have considered. For example, psychosis develops earlier in boys than in girls - but Large and Compton found that even when they adjusted for that difference, the pot smokers still developed psychotic symptoms earlier.
Also, older people are less likely to be smoking pot than younger patients, so if the individual studies had different proportions of various age groups, the results might have been skewed. Taking that factor into account, the researchers still found that marijuana users developed mental illness at a younger age.
They also tried to see whether their findings could be explained by the year the research was done (because pot may have become more potent over time), whether the research had been done according to good scientific principles, and whether the investigators had defined the start of mental illness according to the date it was diagnosed instead of the date the symptoms started (the date symptoms started is more accurate). But they kept coming back to the same result: people with psychotic disorders who smoked marijuana had symptoms of mental illness at a younger age than those who didn't use pot.
The researchers did not look at whether family history of psychosis played a role in determining who was most vulnerable to marijuana's apparent early-triggering effect.
Are there some kids who are more at risk of psychosis from smoking pot than others? "I'm sure there are," Large said. Kids with a family history of psychosis, or those with some psychotic symptoms but not full-blown schizophrenia are at particular risk, he said. But also at risk are kids who are struggling, not doing well at school or living in bad situations at home.
One study found that compared to people who didn't use marijuana, people over 18 who did use it had twice the risk of mental illness, but kids under 15 had five times the risk, Large said.
There are two main messages to take away from this work, Large said. One is that there's probably something in marijuana that triggers schizophrenia. What that is, isn't clear yet. "Schizophrenia is still a mystery," he said. "Psychotic illnesses are horrible for the people who have them, and terrible for their families too."
The second message is far more important, he believes. Public health campaigns on the dangers of marijuana are focusing on older users, Large said, and overlooking the pre-teens and young adolescents who get their pot from older peers and even older siblings at home.
"Even if the onset of psychosis were inevitable (for a particular individual)," Large's team writes, "an extra 2 or 3 years of psychosis-free functioning could allow many patients to achieve the important developmental milestones" of adolescence. That extra time could allow a young person to finish school and gain other skills that might reduce the lifelong disability that so often accompanies mental illnesses.
Whereas most studies like his are presented at medical conferences, Large said he's not going to bother travelling around to announce his results to other doctors. Instead, he said, he wants to talk about his results in public forums.
"I'm not a marketing expert," he said, "but we have to find a way to tell young kids to hold off." We might not be able to convince them to never use pot, he added, but they need to wait until they're older - a message, he acknowledged, that will be tricky to deliver.
Has your child developed a mental illness from smoking pot?