LAUSR

AIMS Cognitive impairment is a possible indicator of neurodevelopmental impairment, but not all psychotic patients are cognitively compromised. It has been suggested that heavy cannabis use may precipitate psychosis in those who show no such compromise. This study compares two samples of patients with first-episode psychosis and their respective non-psychotic controls, in London (UK) and Palermo (Italy), and examines whether different patterns of cannabis use are reflected in differences in IQ. METHODS The two studies used the same inclusion/exclusion criteria and instruments. The sample comprised 249 subjects from London (106 patients and 143 controls) and 247 subjects from Palermo (120 patients and 127 controls). ANCOVA was performed with IQ as the dependent variable and city and frequency of cannabis use as predictors. This was then repeated with the case group only, by adjusting for relevant confounders. RESULTS We found a greater amount of cannabis use in the London sample, compared to Palermo and patients from London had higher IQ than patients from Palermo, a difference that was more significant than that reflected between controls (Fgroup*city(1, 402) = 7.6, p = 0.006). Once corrected for symptomatology and treatment, patients from London who had never used cannabis were similar to patients from Palermo regarding IQ. Thus the higher IQ of patients from London was mainly due to the subgroup of cannabis-using patients (Fcannabis*city(2,145) = 4.6, p = 0.011). CONCLUSIONS We can speculate that a greater amount of cannabis-use may have contributed by precipitating psychosis in patients with a higher IQ in London but less so in patients from Palermo.