Background.Despite evidence on the short-term benefits of early intervention (EI) service for psychosis, long-term outcome studies are limited by inconsistent results. This study examined the 10-year outcomes of patients with first-episode psychosis who received 2-year territory-wide EI service compared to those who received standard care (SC) in Hong Kong using an historical control design.Method.Consecutive patients who received the EI service between 1 July 2001 and 30 June 2002, and with diagnosis of schizophrenia-spectrum disorders, were identified and matched with patients who received SC first presented to the public psychiatric service from 1 July 2000 to 30 June 2001. In total, 148 matched pairs of patients were identified. Cross-sectional information on symptomatology and functioning was obtained through semi-structured interview; longitudinal information on hospitalization, functioning, suicide attempts, mortality and relapse over 10 years was obtained from clinical database. There were 70.3% (N = 104) of SC and 74.3% (N = 110) of EI patients interviewed.Results.Results suggested that EI patients had reduced suicide rate (χ²₍₁₎ = 4.35, p = 0.037), fewer number [odds ratio (OR) 1.56, χ² = 15.64, p < 0.0001] and shorter duration of hospitalization (OR 1.29, χ² = 4.06, p = 0.04), longer employment periods (OR −0.28, χ² = 14.64, p < 0.0001) and fewer suicide attempts (χ² = 11.47, df = 1, p = 0.001) over 10 years. At 10 years, no difference was found in psychotic symptoms, symptomatic remission and functional recovery.Conclusions.The short-term benefits of the EI service on number of hospitalizations and employment was sustained after service termination, but the differences narrowed down. This suggests the need to evaluate the optimal duration of the EI service.