Marijuana continues to have the reputation among the general public as being benign, non–habit-forming, and incapable of inducing true
addiction.
39, 48 For most users this may be so. Experimentation with marijuana has become an adolescent rite of passage, with the prevalence of use peaking in the late teens and early 20s, then decreasing significantly as youths settle into the adult business of establishing careers and families. With a lifetime dependence risk of 9% in marijuana users vs 32% for nicotine, 23% for heroin, 17% for cocaine, and 15% for
alcohol,25 the addiction risk with marijuana is not as high as that for other drugs of abuse. Unlike cocaine dependence, which develops explosively after first use, marijuana dependence comes on
insidiously.49 Marijuana use typically starts at a younger age than cocaine use (18 vs 20 years of age). The risk for new-onset dependence is essentially zero after the age of 25 years, whereas cocaine dependence continues to accrue until the age of 45 years. Likewise, the average age at first alcohol use is the same as for marijuana, but alcohol users will keep on making the transition from social use to dependence for decades after first
use.
49
One in 11 users—1 in 6 for those starting in their early teens—is hardly an inconsequential percentage,
however.50 Like all addictive drugs, marijuana exerts its influence through the midbrain reward center, triggering dopamine release in the prefrontal
cortex.51 Although its existence was questioned until recently, a withdrawal syndrome is increasingly appreciated, characterized by irritability, anxiety, anorexia and weight loss, restlessness, disturbed sleep, and
craving.
52
DuPont7 writes that “marijuana makes users stupid and lazy,” citing an extreme amotivational syndrome characterized by listlessness and apathy in heavy smokers, not just when using the drug but all the time. The befuddled, endearingly dissolute stereotype, parodied in “stoner” movies like Cheech and Chong's
Up in Smoke, is not what happens to most occasional users who experience only temporary mild perceptual changes accompanying a general sense of well-being and ease with the world. The disputed amotivational syndrome of heavy use resembles the negative symptom complex of
53, 54
Using hospitalization as a proxy for serious psychiatric illness, Schubart et
al55 identified a dose-response relationship, with incidental users having 1.6 times the chance of hospitalization and heavy users 6.2 times the risk. “The association of cannabis use with psychiatric inpatient treatment is a clear indication of the association of cannabis use with mental illness,” they wrote. More specifically and more ominously, those with a psychotic predisposition may respond to marijuana with more marked perceptual changes into which they have little insight, accompanied by elevations in hostility and
paranoia.56Schizophrenia has been posited as a hypercannabinoid condition because schizophrenic patients have significantly elevated cerebrospinal fluid levels of anandamide, the most important endogenous
cannabinoid.57 Cannabis use has been implicated as a potential cause, aggravator, or masker of major psychiatric symptoms, including psychotic, depressive, and anxiety disorders, particularly in young
people.30, 58, 59 In underscoring the potential for psychosis, a longitudinal study of more than 50,000 Swedish conscripts has been influential. During a 27-year follow-up period, the more cannabis individuals had used in adolescence, the more likely they were to develop schizophrenia, with those who had used cannabis on more than 50 occasions nearly 7 times more likely to manifest the disease than those who had never used
cannabis.
60
This association between cannabis and psychosis notwithstanding, the question of whether cannabis causes psychosis remains unresolved, even as evidence mounts that its use worsens the course of psychotic illness. In an Australian cohort, Degenhardt et
al61 tested 4 hypotheses regarding the association between cannabis use and schizophrenia, including that cannabis use (1) may cause schizophrenia in some patients, (2) may precipitate psychosis in vulnerable individuals, (3) may exacerbate symptoms of schizophrenia, or (4) may be more likely in individuals with schizophrenia. They noted that during the last 3 decades of the 20th century, cannabis use had significantly increased in Australia without a corresponding increase in schizophrenia prevalence, an observation that gravitated against a simple cause-and-effect relationship between the two. However, they also found that cannabis use precipitated the onset of the disease in the vulnerable and exacerbated the course of the illness in those who already had
In a 2007 meta-analysis pooling 35 longitudinal, population-based studies, Moore et
al59 found an elevated odds ratio (OR) of 1.41 (95% confidence interval [CI], 1.20-1.65) for psychosis in individuals who had ever used cannabis. They also demonstrated a dose-response effect, with the OR increasing to 2.09 (95% CI, 1.54-2.84) for more frequent users, defined—depending on the study—as daily, weekly, or more than 50 times in their lives. A Dutch
study62 shows how this association plays out in actual numbers. For 3 years, van Os et al followed up 3964 psychosis-free individuals, 312 of whom used cannabis. During the observation period, 8 of the 312 (2.2%) developed psychotic symptoms, with 7 of the 8 (88%) having severe enough symptoms to justify receiving a full-fledged diagnosis. Of the 3652 nonusers, 30 (0.8%) developed symptoms, with only 3 of the 30 (10%) meeting criteria for a psychotic disorder. The risk was small in both groups but impressively elevated in users vs
For individuals already diagnosed as having a schizophrenic spectrum disorder, ongoing cannabis use predicts a rockier course. Comparing 24 abusing and 69 nonabusing schizophrenic patients who were otherwise clinically indistinguishable, Linszen et
al63 found 42% of abusers vs only 17% of nonabusers experiencing psychotic relapse during the year-long study period
(P=.03). Moreover, when they compared heavy users (>1 marijuana cigarette per day) with mild users (≤1 cigarette per day), they found an even more robust correlation, with 61% of the heavy users vs 18% of the mild users experiencing relapse
(P=.002). The longer the period of cannabis use, the higher the risk of relapse. In a 10-year follow-up of 229 patients after first hospitalization for schizophrenia, Foti et
al64 demonstrated that the 10% to 18% who continued to use cannabis throughout the study period had a more severe course as measured by the intensity of positive psychotic symptoms. The association was bidirectional: cannabis smokers had worse psychosis, and the more intensely psychotic individuals were more likely to smoke
van Os et al hypothesize that cannabis may exert its negative influence through causing dysregulation in the endogenous cannabinoid system that (among many other interactions) modulates dopamine and other neurotransmitter systems within the brain. They posit a “preexisting vulnerability to dysregulation” that accounts for why some individuals and not others respond to cannabis with
psychosis.62 Using contemporary epigenetic terminology, Henquet et
al65 attribute the greater psychosis risk in certain cannabis users to a synergy between gene (inborn susceptibility) and environment (exogenous trigger). Moreover, increasing evidence implicates a vulnerable developmental period—peripuberty—when cannabis use is more likely to cause
