Outcomes for 236 patients from a 2‐year early intervention in psychosis service

OBJECTIVE: To examine: i) changes in key outcome measures over time in treatment in a representative first-episode psychosis treatment cohort and ii) baseline predictors of service disengagement. METHOD: Baseline characteristics of 236 patients were examined for associations with outcomes over time using generalized estimating equation models. The data on disengagement were analysed using logistic regression. RESULTS: After controlling for admission scores, patients showed consistently improved outcomes while in treatment on functional recovery (unemployment, P < 0.01; HoNOS, P < 0.001; the Quality of Life Scale, P < 0.001; GAF, P < 0.05) but not symptomatology (as assessed by the PANSS and substance abuse). The 64 (33%) who disengaged were more likely to be unemployed (P < 0.01) and have higher HoNOS (P < 0.01) and GAF (P < 0.05) scores at baseline. CONCLUSION: This evaluation has shown significant improvements in psychosocial functioning but not psychopathology during treatment at an Early Intervention for Psychosis Service. Despite attempts to retain patients, there is a high rate of treatment discontinuation.