Predictors of 10-year outcome of first-episode psychosis

Background. Predictors of outcome for psychosis are poorly understood. Duration of untreated psychosis (DUP) appears to predict short-term outcome although its medium- to long-term role remains unclear. Neurodevelopmental indices such as pre-morbid function and/or neurological soft signs may predict longer-term outcome. We aimed to assess the impact of a range of clinical and demographic variables on long-term outcome of a geographically defined, epidemiological first-episode psychosis cohort. Method. A 10-year follow-up was undertaken of a consecutively presenting sample of 109 cases of first-episode psychosis aged 16-50 years. Baseline assessments included positive, negative and depression symptoms, DUP, neurological soft signs and pre-morbid functioning. Multi-dimensional outcomes were assessed blind to baseline data. Results. All participants were traced at a mean of 10.5 years post-index admission: 11 had died, 10 from non-natural causes. Of the surviving cases, 70% were comprehensively re-assessed by inter-view. Summary data on the remainder were collected from their family practitioner and chart review. Poor 10-year outcomes were predicted independently by poor pre-morbid functioning, baseline negative symptoms and longer DUP. The same measures, plus neurological soft signs, appeared to predict outcomes in a DSM-IV schizophrenia/schizo-affective subgroup. Conclusions. Poor pre-morbid functioning, baseline symptoms, DUP and neurological soft signs at onset independently predict poor long-term outcome in first-episode psychosis.