Early course of schizophrenia in a representative Dutch incidence cohort

Purpose: To describe the early course of psychotic disorders in general and to examine whether certain variables can predict the early course of schizophrenic disorders (DSM-IV: schizophrenia, schizophreniform or schizoaffective disorder), Subjects and method: Follow-up and re-diagnosis of a highly representative Dutch incidence cohort (N= 181), thirty months after first contact with a physician for a psychotic disorder. Poor course was defined as a continuous psychotic illness or a score of less than 39 on the Global Assessment of Functioning scale. Results: The follow-up rate was 92%. 125 Subjects were diagnosed with a schizophrenic disorder. Poor course was present in 70 of these subjects (56%). Univariable analysis showed that male sex, heavy cannabis use during the follow-up period (sometimes or often more than one joint a day) and long duration of dysfunctioning before psychosis onset (> 1 month) were predictors of poor course, while age at onset, ethnicity, socioeconomic status and duration of untreated psychosis (trend, p=0.08) were not. The effect of cannabis was confounded by sex. Multivariable analysis showed that male sex was the sole significant and independent predictor of poor course and explained 13% of the variation. The odds ratio for males, adjusted for duration of pre-psychotic dysfunctioning and cannabis use during the follow-up period, was 3.0 (95% CI, 1.0-8.9). Strengths and limitations: This is the first study to examine the influence of cannabis in an epidemiological, highly representative sample. A limitation was the sample size. Conclusion: Male sex is an independent risk factor for an unfavorable early course in schizophrenia. (c) 2007 Elsevier B.V. All rights reserved