DOES CANNABIS USE WORSEN PSYCHOTIC SYMPTOM PRESENTATION?

Abstract

Background: While the relationship between cannabis and psychosis is well established, there is a lack of studies into whether cannabis use is associated with a particular pattern of symptoms at psychosis onset. Moreover, there is much evidence that psychotic experiences are common in the healthy population, and again their relationship with exposure to cannabis has been scarcely studied. We hypothesized that psychopathology in first-episode psychosis patients (FEP), and psychotic experiences in controls, would be qualitatively and quantitatively affected by pattern of cannabis use. Methods: The Operational CRITeria (OPCRIT) system, the CAPE (Community Assessment of Psychic Experiences), and the CEQ (Cannabis Experience Questionnaire) were used for assessing clinical/subclinical symptoms and pattern of cannabis use in 1200 FEP and 1500 controls across 15 European sites. Three different methods were tested to aggregate OPCRIT items in dimensions: principal component analysis (PCF), confirmatory factor analysis (CFA) in structural equation modeling, and item response theory (IRT). Number of components to extract in PCF was determined by Monte Carlo simulation, using R. In all factor analyses, composite scores for each symptom dimension were predicted by regression, and the main effect of cannabis on such dimensions was tested after analysis of variance. Results: The sample was good (MSA = 0.84) and data suitable for factor analyses (Bartlett sphericity test, χ2(1830) = 20 252, P < .0001), with a best-fitting model including 5 latent symptom dimensions: mania, disorganization, depression, positive, and negative symptoms. PCF further deconstructed positive dimension into: (1) first rank passivity delusions, including thought insertion and related phenomena; (2) first rank hallucinations; and (3) paranoid delusions. Traditional and alternative measurement methods showed that individuals with high frequency of cannabis use had more positive symptoms, and, among them, cannabis had a strong impact on a cluster of phenomena of passivity, such as thought insertion/broadcast/withdrawal/echo, and passivity/bizarre delusions, F(1, 1129) = 11.25, P < .0001, heavy use accounting for higher scores. In controls, positive psychic experiences were found associated with high frequency of cannabis use, F(1, 1366) = 11.81, P < .0001]. Conclusion: Our findings suggest that FEP patients are heterogeneous in their psychopathology and cannabis use may have differential influence on positive symptom presentation. Furthermore, heavy cannabis use may increase subclinical psychotic experiences, as part of a phenomenological continuity in the general population. Specific clusters of positive symptoms, that is, though insertion related phenomena, may arise from distinct mechanisms in cannabis-associated psychosis, suggesting that such factors would be more sensitive to detect associations with biological pathways in future studies.