Predictors of long-term outcome in schizophrenia

Further clarification of factors predicting the outcome in schizophrenia is needed. The present review examines recent research into some of these predictors, focusing on insight, duration of untreated psychosis, cognition and early treatment response. It also addresses the need for standard outcome measures.

There is good evidence that poor insight predicts poor outcome, although perhaps not simply as a consequence of poor compliance. Further support is provided for a link between duration of untreated psychosis and long-term outcome. The relationship between cognition and outcome is complex, with specific cognitive deficits apparently predicting particular outcome domains. Early treatment response is closely related to long-term outcome. Outcome studies may be flawed by sample selection bias, and a lack of standardized outcome measures.

Several predictors are potentially modifiable, indicating that they should be targets for therapeutic intervention. More carefully designed studies are needed. Recently proposed criteria for remission are helpful, and should facilitate cross-sample comparisons.