Exploring the relationship between cannabis use and psychosis

Abstract

Evidence for an association between cannabis use and psychosis continues to accumulate. A number of population-based longitudinal studies have found that cannabis use is associated with the later development of psychotic symptoms or psychotic diagnoses [1–6]. Studies of subjects with schizophrenia show associations between cannabis use and increased levels of positive symptoms [7,8] as well as increased number of relapses [9,10]. Additional lines of evidence support a causal role for cannabis in the development of psychosis. In experimental studies administration of cannabinoids can induce transient psychotic experiences, whilst molecular studies suggest that cannabis can increase meso-limbic dopaminergic transmission as well as inhibit glutamatergic release, in keeping with current theories of schizophrenia aetiology [11–13]. The problem, of course, is that it is very difficult to establish causality from observational studies [14]. The most likely alternative explanation for the association between cannabis use and psychosis is that personality traits or other confounders predispose individuals both to using cannabis and to developing psychotic symptoms. An association reported between tobacco use and reduced risk of schizophrenia argues against confounding by factor such as personality traits influencing non-specific substance use, but cannot rule it out [15]. There is increasing interest in the possibility that psychotic phenomena exist as a continuum that extends into the general population [16,17]. Stefanis et al. [18] examine the relationship between cannabis use and psychotic symptom dimensions in a cross-sectional survey of a population-based birth cohort. They report associations between cannabis use and the presence of subclinical levels of both positive and negative dimensions of psychosis. The association with negative symptoms persists after adjustment for depressive symptoms, suggesting that this association is not simply due to the rating scales inadvertently measuring depressive symptoms. A similar finding was reported in another non-clinical sample that used the same rating scales [19]. In contrast, people with schizophrenia who use cannabis report fewer negative symptoms [8,20,21]. This has been one argument in support of the self-medication hypothesis of cannabis use [20], though in fact there is little evidence that cannabis use alleviates symptoms of psychosis. A possible explanation for these apparently contradictory results is confounding by premorbid levels of functioning, such that individuals with the propensity for developing a more severe illness characterised by marked negative symptomatology may be less likely to seek out and partake in substance misuse [19]. This could lead to an apparent association between cannabis use and reduced levels of negative symptoms in clinical samples if this effect were stronger than any effect of cannabis on increasing negative symptoms. Stefanis et al. also report a stronger association between cannabis use and positive symptoms in subjects first using cannabis before age 15; results similar to those reported from the New Zealand Dunedin cohort [4]. These findings are of great interest as they suggest that early adolescence, when the brain is still undergoing maturational processes such as myelination, may be a critical period in terms of vulnerability to the effects of exposure to cannabis. If this is indeed the case, this raises important public health concerns given the trends over time for more frequent use and earlier age of first use of cannabis during adolescence. The Stefanis et al. study is not able to address the issue of whether cannabis increases the risk of psychosis or psychotic symptoms. However, it points to a promising method of studying the aetiology of psychotic symptoms in the community. Future longitudinal research of psychotic symptoms in the community could lead to an improved understanding of the relationship between cannabis and psychosis and whether there are critical periods of brain development when cannabis is particularly harmful.