Marijuana exacerbates psychotic symptoms and worsens outcomes in patients already diagnosed with schizophrenia or other psychotic disorders. Several large observational studies also strongly suggest that using marijuana — particularly in the early teenage years — can increase risk of developing psychosis.

An often-cited study of more than 50,000 young Swedish soldiers, for example, found that those who had smoked marijuana at least once were more than twice as likely to develop schizophrenia as those who had not smoked marijuana. The heaviest users (who said they had used the drug more than 50 times) were six times as likely to develop schizophrenia as the nonsmokers.

Until recently, the consensus view was that this reflected selection bias: Individuals who were already vulnerable to developing psychosis or in the early stages (the prodrome) might be more likely to smoke marijuana to quell voices and disturbing thoughts. But further analyses of the Swedish study, and other observational studies, have found that marijuana use increases the risk of psychosis, even after adjusting for possible confounding factors.

Although cause and effect are hard to prove, evidence is accumulating that early or heavy marijuana use might not only trigger psychosis in people who are already vulnerable, but might also cause psychosis in some people who might not otherwise have developed it.

Certainly genetic profile mediates the effect of marijuana. People born with a variation of the gene COMT are more vulnerable to developing psychosis, for example. Because there is as yet no reliable way for clinicians to identify vulnerable young people in advance, however, it is safest to restrict use of medical marijuana to adults.

http://en.wikipedia.org/wiki/Catechol-O-methyl_transferase