Testing the Ultra High Risk (prodromal) criteria for the prediction of psychosis in a clinical sample of young people

Criteria for identifying individuals at imminent risk for onset of a psychotic disorder, that is "prodromal" for psychosis, have recently been described. The current study set out to test the predictive validity of these criteria in a sample of help-seeking young people aged 15-24 years who were referred to, but not necessarily treated at, a psychiatric service. Ultra High Risk (UHR) status was determined at baseline and psychosis status was assessed at 6 month follow up. Baseline psychosocial functioning was also assessed as a possible predictor of psychosis. In the sample of 292 individuals, 119 (40.7%) met UHR criteria. Of these UHR+ people, 12 became psychotic within 6 months and 107 did not. Only one person not meeting UHR criteria developed psychosis in the follow up period. Sensitivity, specificity, positive predictive value and negative predictive value of UHR+ status for prediction of psychosis were, respectively, 0.923 (95% CI 0.621, 1), 0.616 (95% CI 0.556, 0.673), 0.101 (95% CI 0.056, 0.173) and 0.994 (95% CI 0.963, 1). UHR+ individuals were significantly more likely to become psychotic than UHR- individuals (Odds Ratio 19.3, 95% CI 2.5, 150.5). Low functioning at baseline was associated with psychosis onset in the whole sample and in the UHR group. The transition to psychosis rate was much lower than in previous samples. This may be a due to the sample being a more general one, not identified as possibly "prodromal". Other potential causes of this reduction in transition are also explored. © 2006 Elsevier B.V. All rights reserved.