What drives the higher incidence of psychosis in London compared to Palermo?

Background: Incidence of psychosis seems to be lower in Italy than in other European countries (Tansella et al. 1991; Lasalvia et al. 2012, Tarricone et al. 2012); however there are no studies comparing the incidence of psychotic disorder in Northern and Southern Europe. Methods: Incidence and socio-demographic data on all psychotic patients presenting for the first time to the mental health services of Palermo were collected over a period of three years. Palermo incidence rates were compared to South London rates obtained from the AESOP study (Kirkbride et al 2006). South London rates were reanalyzed excluding people aged 16-17 years and substance related psychoses. Second generation migrants (people who were born in UK belonging to ethnic minorities) were also excluded to make the sample comparable since migration in Palermo referred to people who were not Italy born. The term migrants was used in the present analysis to indicate non-native born British and Italians respectively. Incidence rates of overall psychosis, schizophrenia, affective psychoses and other non-affective psychoses were compared in Palermo and in South London by indirect standardization (by age and gender and then by age, gender and migration) to take into account the differences in the population structures between sites. Standardized morbidity ratios (SMRs) and their reciprocal of overall psychoses, schizophrenia, other non-affective psychoses and affective psychoses were calculated. Results: During the study period two hundred and four patients affected with a first episode of psychosis (FEP) were ascertained in Palermo. South London cases were 195. Standardized incidence rates of overall psychoses were 16.9 (95% CI 14.7-19.4) per 100,000 per year in Palermo and 36.8 (95% CI 31.8-42.3) in South London. Migrants had an increased risk of developing a psychotic disorder both in Palermo OR: 3.12 (95% CI 1.89-4.93) and in London OR: 2.9 (95% CI 2.15-3.93). After standardizing by age and gender the risk of psychosis was significantly higher in South London compared to Palermo for all psychoses 1/SMR = 2.18 (95% CI 1.98-2.39) and for each diagnostic category. After standardizing also for migration the difference in risk of overall psychoses between Palermo and London decreased: 1/ SMR = 1.39 (95% CI 1.23-1.56) and no differences in risk were found any longer between Palermo and South London for schizophrenia and other non affective psychoses; there was however an increased risk of affective psychoses in South London compared to Palermo 1/SMR = 3.31 (95% CI 2.52-4.21). This result confirms that migration explains the majority of the difference in incidence rates between Southern Italy and London. Discussion: This is the first epidemiological study of psychosis ever carried out in Sicily and one of the few from Southern Europe. The risk of psychoses was higher in South London when compared to Palermo. However no significant differences were found in rates of schizophrenia and other non-affective psychoses after taking into account the different proportion of migrants in the two sites suggesting that migration might explain the majority of the difference in the risk of psychosis. However, it did not explain all the difference as there was still an excess of affective psychoses in South London. Further studies are needed to explore the role of other risk factors (Mediterranean diet, vitamin D, social fragmentation, drug use) in influencing the risk of psychosis