Cannabis and psychosis

Publisher Website
Google Scholar
Abstract
Epidemiological evidence Contemporary interest in this topic began with a longitudinal study of Swedish conscripts reported by Andreasson and his colleagues.1 Their findings have been replicated and extended in a series of longitudinal studies2–6 all of which have found increased rates of psychosis or psychotic symptoms in people using cannabis (table). Furthermore, these findings of longitudinal, case-control studies have been augmented by a series of cross-sectional studies of large populations7 and high risk populations.8–11 These studies produce the following suggestive evidence that supports the conclusion that the link between the use of cannabis and increased risks of psychosis is likely to be causal. View this table: In this window In a new window Summary of prospective studies of cannabis use and psychotic symptoms Demonstrator for the legalisation of cannabis Credit: PHILLIPPE HAYS/REX What is psychosis? Psychosis is used in this research as a generic description of severe mental illness characterised by the presence of delusions, hallucinations, and other associated cognitive and behavioural impairments that interfere with the ability to meet the ordinary demands of life. It is measured either by using standardised diagnostic criteria for psychotic conditions such as schizophrenia or by using validated scales that rank the level of psychotic symptoms from none to severe. Association—All studies found that the use of cannabis is associated with increased risks of psychosis or psychotic symptoms. The table shows the associations between use of cannabis and psychosis across existing longitudinal studies; odds ratios range from 1.77 to 10.9, with a median of 2.23-2.3. Dose response—Although most studies have compared cannabis users and non-users, several studies have shown that the increasing use of cannabis is associated with an increasing risk of psychosis,1 4 6 with odds ratios or relative risk for the groups with highest use groups increasing to 6.0,1 1.77,4 and 6.81.6 Assessment of outcome—The associations between cannabis use and psychotic symptoms have proved robust to different methods of assessing outcomes. Associations have been found using various outcome measures including clinical diagnoses of psychoses,1 diagnostic classifications based on self report data,2 3 and symptom scores.2–6 Confounding—The validity of studies of cannabis is threatened by the possibility of residual uncontrolled confounding factors. More recent studies2–6 have controlled for a wide range of factors that could confound the relation between cannabis and psychosis—for example, genotype, sex, age, psychosis before using cannabis, education, personality, IQ, affiliation with deviant peers, conduct and attention disorders, other substance use, social functioning, previous mental health, parental age, parental divorce, changes in parents, parental attachment, parental offending and substance use, socioeconomic factors, physical and sexual abuse, and childhood trauma. Reverse causality—A further threat to the validity of claims of a link between cannabis and psychosis comes from the possibility of a reverse causal association in which the development of psychotic symptoms encourages the use of cannabis, perhaps as self treatment. To control for reverse causality, prospective studies have assessed the use of cannabis before the onset of psychotic symptoms.2–6 In addition, a recent study used structural equation modelling to examine the causal linkages between cannabis and psychotic symptoms.4 This study concluded that although use of cannabis was associated with increased rates of psychotic symptoms, the development of psychotic symptoms was associated with a decrease in rates of use of cannabis. Effect modification—Further evidence suggests that use of cannabis is linked with development of psychotic symptoms in people who are susceptible to developing psychotic symptoms, including those with a past diagnosis of a psychotic disorder,6 those reporting psychotic or paranoid symptoms at baseline,5 and those with a family history of psychotic disorder.8 A recent behavioural genetic study found that this link is stronger in those who have the Val/Val variant of the catechol-O-methyltransferase (COMT) gene.3 This gene has a role in regulating dopamine concentrations and has been implicated in the development of schizophrenia (see below). In summary, epidemiological research using longitudinal designs has produced suggestive evidence of a causal link between the use of cannabis and the development of psychosis or psychotic symptoms. This link has been shown to be robust and resilient; however, questions may still be raised regarding the measurement of psychotic symptoms, control for confounding factors, and the possibility of reverse causality. Priorities for future research include improving techniques for covariate control, and assessing the precise nature of the symptoms or disorders that may be associated with cannabis use.