The relationship between cannabis use and psychosis has been extensively studied, with evidence suggesting an association between cannabis use and an increased risk of psychosis or an earlier onset of psychotic symptoms. However, the causal nature of this relationship remains debated. While cannabis can induce transient psychotic-like experiences in healthy individuals, the impact of genetic predisposition on this relationship is not fully understood. This study aimed to address this gap by examining the association between cannabis use and self-reported psychotic experiences in a large population cohort, the UK Biobank. The research questions focused on whether cannabis use is differentially associated with various types of psychotic experiences, whether cannabis users' experiences differ in terms of onset, distress, and help-seeking behavior, and finally, whether genetic predisposition to schizophrenia moderates the association between cannabis use and psychotic experiences. Understanding these relationships is crucial for developing effective preventative and interventional strategies. The clinical relevance of psychotic experiences, even those not severe enough for a formal diagnosis, is significant as they often predict poor outcomes, including the development of psychotic disorders. A meta-analysis has confirmed a dose-dependent relationship between cannabis use and various psychosis-related outcomes, including self-reported psychotic experiences, further emphasizing the need for this research. While cannabis has historically been classified as a hallucinogen, this is now considered controversial, and the relationship between cannabis use and different types of psychotic experiences (hallucinations versus delusions) warrants further investigation. Genetic studies suggest a role for the endocannabinoid system in psychotic experiences, with the CNR2 locus showing a strong association with distressing psychotic experiences. Furthermore, genetic predisposition to schizophrenia might moderate the susceptibility to cannabis-related harm, such as the association between cannabis use and reduced cortical thickness (a risk factor for psychotic experiences). This study uniquely investigated the hypothesis that genetic predisposition to schizophrenia increases vulnerability to cannabis-induced psychotic experiences.

Existing epidemiological studies show a strong link between cannabis use and psychosis, with some suggesting an accelerated age of onset of psychosis. However, the causal relationship remains debated. Cannabis is known to induce acute psychotic-like experiences, often mild and transient, in healthy individuals. The genetic mediation of this association is largely accepted, although environmental factors also play a role. Population-based surveys highlight that self-reported psychotic experiences are far more common than diagnosed psychotic disorders, and even mild experiences can predict poor mental health outcomes. Prior research points to a dose-dependent relationship between cannabis use and various psychosis-related outcomes. While cannabis's effects on perception are noted, the classification as a hallucinogen is debated, with case reports of cannabis-induced hallucinations being rare. Delusions appear more commonly associated with cannabis use. Previous work has also indicated that regular marijuana use is associated with increased odds of both hallucinations and paranoia, although the latter shows a stronger association. Genetic research implicates the endocannabinoid system in psychotic experiences, with studies highlighting the CNR2 locus. The interaction of genetics and environment is crucial, with the association between cannabis and decreased cortical thickness (a risk factor for psychotic experiences) being most pronounced in those with high polygenic risk for schizophrenia. However, prior research has not directly tested the hypothesis that genetic predisposition to schizophrenia increases vulnerability to cannabis-induced psychotic experiences.

This cross-sectional study utilized data from the UK Biobank, a large prospective cohort study comprising approximately 500,000 British individuals aged 40-69 years at recruitment. A subset of 109,308 unrelated White British participants who completed a mental health questionnaire and lacked a diagnosis of any psychotic disorder (ICD-10 codes F20-F29) were included in the analysis. Self-reported cannabis use was determined using questions on cannabis ever-use and frequency of use. Self-reported psychotic experiences included auditory and visual hallucinations, persecutory delusions, and delusions of reference, consistent with previous UK Biobank studies. Additional questions assessed age of onset, distress, and help-seeking behavior related to psychotic experiences. A polygenic risk score (PRS) for schizophrenia was calculated for each participant using data from a recent genome-wide association study (GWAS). The PRS was computed by summing the weighted effects of multiple single nucleotide polymorphisms (SNPs) associated with schizophrenia. Variants with p<0.05 were included after several quality control steps. The analysis used logistic regression to examine the associations between cannabis use (both ever-use and frequency) and each type of psychotic experience. Covariates included demographic factors (age, sex, education, income, employment status, deprivation indices), lifestyle factors (smoking, alcohol use), and the top ten genotype principal components to account for population stratification. Additive interactions between cannabis use (both ever-use and frequency) and the schizophrenia PRS were assessed using linear regression, including covariate-by-exposure and covariate-by-PRS interactions to account for confounding. Multiple testing corrections were applied using the Benjamini-Hochberg method at a false discovery rate (FDR) of 10%. To further interpret interactions, participants were stratified into quintiles based on their schizophrenia PRS, and associations between cannabis ever-use and psychotic experiences were examined within each quintile.

The study revealed a strong, dose-dependent relationship between self-reported cannabis use frequency and all four types of self-reported psychotic experiences. The prevalence of any psychotic experience increased from 4.1% in never-users to 7.0% in ever-users, and further to 9.6% in daily users. The adjusted odds ratio (AOR) for any psychotic experience increased by 20% per unit increase in cannabis use frequency. Persecutory delusions showed the strongest association with cannabis use. A sensitivity analysis showed a stronger association between cannabis ever-use and psychotic experiences among females than males. Cannabis users reported earlier-onset psychotic experiences (AOR = 1.90 for early-onset versus 1.52 for adult-onset) and more distressing experiences (AOR = 1.62) than non-users, but no significant difference in help-seeking behavior. Crucially, the association between cannabis use and psychotic experiences was significantly modulated by the schizophrenia PRS. Interactions were observed between cannabis use and the schizophrenia PRS for auditory hallucinations, delusions of reference, and overall psychotic experiences. This interaction was particularly strong for delusions of reference where there was no significant association with cannabis unless individuals were in the top two quintiles of genetic risk. Stratifying participants by schizophrenia PRS quintiles demonstrated that cannabis ever-use was associated with a greater increase in the odds of psychotic experiences among those with the highest PRS compared to those with the lowest PRS. This pattern was observed for auditory and visual hallucinations and was particularly prominent for delusions of reference.

This study provides robust evidence supporting the hypothesis that cannabis use increases the risk of psychotic experiences, particularly persecutory delusions, in a dose-dependent manner. The finding that cannabis users reported earlier-onset and more distressing experiences is clinically relevant. The significant interaction between cannabis use and schizophrenia PRS highlights the increased vulnerability of genetically predisposed individuals to cannabis-induced psychotic experiences. This emphasizes the complex interplay of genetic and environmental factors in the development of psychosis. The stronger association observed in females warrants further investigation into sex-specific factors influencing the relationship between cannabis use and psychosis. While the study's findings do not definitively establish causality, they strongly suggest a predictive risk factor role for cannabis use. The utilization of a polygenic risk score allows for a more nuanced understanding of individual risk. The study underscores the utility of large-scale biobanks in elucidating gene-environment interactions and informing personalized approaches to harm reduction. These findings have considerable translational potential for developing tailored interventions to mitigate the risk of cannabis-induced psychosis, particularly in individuals with a high genetic predisposition to schizophrenia.

This large-scale study demonstrates a robust association between cannabis use and psychotic experiences, particularly persecutory delusions, with a stronger effect in individuals with high genetic risk for schizophrenia. Cannabis users experienced earlier-onset and more distressing psychotic episodes. These findings highlight the need for personalized harm reduction strategies, considering both cannabis use and genetic predisposition. Future research could explore the specific mechanisms underlying this gene-environment interaction and develop more targeted interventions.

This study is cross-sectional, limiting causal inferences. Reliance on self-reported data may introduce recall bias. The study focused on a specific population (White British individuals from the UK Biobank), limiting generalizability. While the schizophrenia PRS captures a significant portion of genetic risk, it does not account for all genetic factors influencing psychosis.