An Investigation of Cannabis Use and Childhood Trauma in Relation to Psychotic Symptoms and Their Outcomes: a Population-Based Follow-up Study of Irish Adolescents

Background: Adolescent cannabis use has been shown to increase risk of later psychosis and childhood trauma is associated with both substance misuse and risk for psychosis. Psychotic symptoms in adolescents have been linked with later common mental disorders as well as psychotic outcomes, however personality disorders have not been investigated. There is limited information on rates of psychiatric disorder among adolescents and adults with psychotic symptoms in community samples, and no information on rates of personality disorders in this group. This study aimed to investigate prevalence (in adolescence and young adulthood), incidence and persistence rates of psychotic symptoms in a young Irish population, and whether early versus later onset of cannabis use and childhood trauma increase risk of psychotic symptoms, separately and/or by an interaction effect. Adult psychiatric outcomes and comorbidities of psychotic symptoms were also investigated. Method: This was a longitudinal prospective population-based study. Secondary schools from North Dublin were randomly selected for inclusion. Children were screened using questionnaire (CDI, SDQ). An at-risk group scoring above threshold on these instruments (n=140) were invited for interview together with 193 matched controls. 212 adolescents aged between 12 and 16 years and their parents were interviewed using the Schedule for Affective Disorders and Schizophrenia for School-Age Children (K-SADS) semi-structured instrument. Childhood trauma was defined as child sexual abuse, child physical abuse or having witnessed domestic violence. Eight years after the initial study, all 212 participants were invited for interview. Follow-up assessment measures included the SCID I and II (schizoid, 2 8 schizotypal, paranoid, borderline, antisocial personality disorders), SIPS screener, ASR and ABCL. Results: Weighted prevalence of psychotic symptoms was 4.2% in adolescence and 7.8% in young adulthood. Participants who had used cannabis and/or experienced childhood trauma were more likely to experience adolescent psychotic symptoms in main effects analyses. In a multivariate analysis, cannabis use was independently associated with adolescent psychotic symptoms, but childhood trauma conferred no additional risk. Presence of both childhood trauma and early cannabis use increased the risk for adolescent psychotic symptoms beyond that posed by either risk factor alone, indicating a greater than additive interaction. Psychotic symptoms in adolescence were associated with depressive and conduct disorders, but not anxiety disorders. No young person had a psychotic disorder in adolescence. Follow-up rate was 80% (n=169). 7.8% had current and 10.1% lifetime psychotic symptoms in adulthood. 56.0% had a lifetime Axis I disorder, 19.6% in the past month. Childhood trauma was associated with adult psychotic symptoms, however there was no independent association in a multivariate analysis including cannabis use and baseline psychotic symptoms. Early adolescent onset cannabis use increased risk of psychotic symptoms in early adulthood in univariate and multivariate analyses (controlling for childhood trauma, current cannabis use and baseline psychotic symptoms), later onset use did not. Adolescent psychotic symptoms predicted mood disorders and personality disorders, but not anxiety disorders at adult follow-up. Adult psychotic symptoms were associated with current anxiety disorders, mood disorders, and personality disorders. Results suggested an interaction between childhood trauma and late adolescent/adult onset of cannabis use in increasing risk of adult psychotic symptoms. Conclusion: Prevalence, incidence and persistence rates of psychotic symptoms in this sample are similar to international figures. The finding of an interaction between childhood trauma and adolescent cannabis use has been replicated in larger samples, but needs further investigation as results have 29 been inconsistent. This may have implications for the identification of individuals at high risk of experiencing psychotic symptoms; measures to discourage cannabis use in abused adolescents may help to prevent the development of psychosis in this vulnerable group. Current psychiatric disorders were more likely in young people with current psychotic symptoms and in those who had reported adolescent psychotic symptoms. Further large population-based studies following up adolescents with psychotic symptoms to investigate both Axis I and Axis II psychiatric outcomes are needed to clarify their prognostic significance. 30