Cannabis is the most commonly used illicit substance, especially among young persons. The consequences of cannabis use remain unclear and highly controversial. Although use of the drug has been associated with psychologic problems, reduced educational achievement, antisocial behavior, and a range of psychosocial conditions, some believe these associations are not causal. Rather, psychosocial problems could predispose young persons to drug use rather than vice-versa, or drug use and the various other factors may be manifestations of underlying adverse circumstances. Macleod and colleagues reviewed the evidence for cannabis use leading to psychologic and social harm in young persons.

The authors searched several electronic databases, abstracts, and references to identify relevant publications. They also contacted experts to identify all published and unpublished data relating to illicit drug use by persons 25 years or younger to any measure of subsequent social or psychologic harm. The initial search identified more than 200 publications relating to 48 longitudinal observational studies. The studies were assessed independently by several researchers, and consensus was reached on quality. Sixteen studies were judged to be of higher quality, but little consistency was evident among studies in participant eligibility and recruitment, methodology, follow-up, or outcomes measured.

Cannabis use was associated consistently with lower educational achievement. The crude estimates for school dropout were high (six to eight times that of nonusers), but these were reduced to three to five times by adjustment for other factors associated with early termination of schooling. Cannabis use also was associated consistently with reported use of other drugs. The use of cannabis was associated in one study with a 6.8-fold increase in risk for later injection drug use, but this risk was reduced to 3.3 after adjustment for other factors. Another study reported an almost 60-fold increase in subsequent use of other drugs in young persons who used cannabis at least once per week. Studies on the use of other drugs relied on self-report and did not assess problems related to drug use.

The association between cannabis use and psychologic problems was inconsistent and usually depended on self-report of subsequent problems. Adjusted estimates for use of cannabis at age 15 and any mental disorder varied between studies, from 0.97 for both sexes to 2.00 for males and 0.75 for females. Individual studies reported an enhanced risk (adjusted) of 3.1 for development of schizophrenia and 5.6 for depression in women but not in men (estimate: 1.1). The relationship between problem behavior and police contact was similarly inconsistent and greatly affected by adjustment for other associated factors; however, cannabis use at younger ages was associated consistently with subsequent behavior problems.

The authors conclude that cannabis use is associated with psychosocial harm, but the extent, strength, and nature of this association are not clear. They stress that these conclusions cannot be interpreted as indicating that cannabis is harmless. The article and a related editorial discuss the public health and policy implications of these findings.