The cannabis-psychosis link
Summary of key studies linking cannabis and psychosis
Many studies have explored the link between cannabis and psychosis (Table). In a systematic review, Moore and colleagues10 surveyed the literature on this topic. They looked at population-based longitudinal studies as well as nested case-control studies that assessed the impact of cannabis use on the later development of psychosis. The “psychosis” outcomes required the diagnosis of a primary psychotic disorder or affective psychosis, or the occurrence of delusions, hallucinations, or thought disorder during the study period. Results from 7 cohort studies showed a 40% increased risk of psychosis in cannabis users compared with nonusers. The data also revealed a dose-response effect—the risk of psychotic symptoms was increased approximately 50% to 200% in those who used cannabis frequently compared with nonusers.
Critics of this hypothesis believe that cohort studies have inherent limitations that prevent any clear conclusions from being drawn. McLaren and colleagues11 evaluated the methodological strength of the existing cohort studies. The definition of psychosis was a recurrent limitation in the studies. Many studies used psychotic symptoms, not diagnoses, as their outcome, which may not be of clinical significance. Moore and colleagues10 also noted this limitation and attempted to correct for it by separately analyzing the 2 studies that required the diagnosis of a primary psychotic disorder. Interestingly, they found an odds ratio of 2.6 for the development of psychotic disorders in those who had ever used cannabis compared with nonusers. Important confounding factors, such as noncannabis drug use, a family history of psychosis, and unmeasured vulnerability to psychosis, were not adequately controlled in these studies.11
Age at onset of psychosis and cannabis use
Certain risk factors have been reported to interact with cannabis use to increase vulnerability to developing psychosis. One suspected important variable is the age at which cannabis use is started. The age effect was first noted in a Swedish conscript cohort study that demonstrated that cannabis use by age 18 led to a 6-fold increase in the risk of schizophrenia later in life.12 It is unclear, however, whether the psychotic symptoms predated the cannabis use.
To clarify this issue, the Dunedin Multidisciplinary Health and Development Study conducted a prospective longitudinal study of adolescent cannabis use, taking into account psychotic symptoms that occurred before cannabis use.13 The data were compiled from a birth cohort that consisted of 1037 individuals born in Dunedin, New Zealand. Information about psychotic symptoms was obtained at age 11, and drug use was assessed by self-reports at ages 15 and 18 and by a standardized interview schedule at age 26. Two psychosis-related outcomes were measured—the presence of symptoms of schizophrenia and the diagnosis of schizophreniform disorder.
The results showed that those who had used cannabis by ages 15 and 18 had more schizophrenia symptoms than controls, a finding that remained significant after controlling for the presence of psychotic symptoms at age 11. However, the increased likelihood of schizophreniform disorder at age 26 was no longer significant after controlling for psychotic symptoms at age 11. Taken together, this suggests that early cannabis use confers higher risk of psychosis.
These findings may be explained as follows: Adolescence represents a sensitive period of neurodevelopment, during which the brain is more vulnerable to the effects of cannabis. Alternatively, the heightened risk may simply be a consequence of greater cumulative cannabis use, since these subjects began using it at a younger age. These theories are not mutually exclusive, and the latter explanation is consistent with the previously mentioned dose-response relationship observed in many studies.
