Incidence of schizophrenia among Moroccan immigrants to the Netherlands.

With interest we read the report of a psychosis incidence study in the Dutch town of Utrecht (Zandi et al., 2010). The authors hypothesized that previous reports of an increased incidence of schizophrenia among Moroccan–Dutch people in the Netherlands were due to cross-cultural bias. They examined first-onset cases using a “culturally sensitive” version of the Comprehensive Assessment of Symptoms and History (CASH; Andreasen et al., 1992) and arrived at the conclusion that the first-contact incidence of schizophrenia among the Moroccan– Dutchwas no longer significantly increasedwhen this culturally sensitive instrument was applied. The purpose of this letter is to examine whether the authors have made a strong case. The study procedures were as follows. Patients suspected of a first psychotic episode during the period May 1st 2002–May 1st 2004 were reported to a central office and interviewed twice: (1) academic psychiatrists (or residents) administered the standard Dutch version of the CASH and made a DSM-IV diagnosis; (2) the authors applied the culturally sensitive version of this instrument (CASH-CS), discussed all findings during a diagnostic meeting and made another DSM-IV diagnosis. The Relative Risks (RRs; Moroccan–Dutch versus Dutch nationals) based on the standard CASH turned out to be much higher than those based on the CASH-CS. Using information from the CASH-CS the RR of schizophrenic disorders (DSM-IV: schizophrenia, schizophrenifom disorder or schizo-affective disorder) dropped from 7.8 (95% CI 4.0–15.2) to 1.5 (0.5–4.3). The authors also observed that they could not replicate the high RR of 9.3 (95% CI 3.7–23.4) for secondgeneration Moroccans in The Hague (Selten et al., 2001), because not a single second-generation Moroccan was reported to the central office of the researchers. We wish to make the following comments. Firstly, the previous epidemiological studies found an increased incidence or prevalence of schizophrenia among Moroccan–Dutch males, not among Moroccan–Dutch females (Brook & de Graaf, 1985; Selten and Sijben, 1994; Schrier et al., 2001; Selten et al., 2001; Veling et al., 2006). Since Zandi et al. reported a RR of schizophrenic disorders for Moroccan–Dutch males of 2.4 (95% CI 0.8–7.7), the 95% confidence interval of which includes the usually reported RR of about 4 to 5 for Moroccan–Dutch males, their finding is not significantly