IMPORTANCE Long-acting, injectable, second-generation antipsychotic medication has tremendous potential to bring clinical stability to persons with schizophrenia. However, long-acting medications are rarely used following a first episode of schizophrenia. OBJECTIVE To compare the clinical efficacy of the long-acting injectable formulation of risperidone with the oral formulation in the early course of schizophrenia. DESIGN, SETTING, AND PARTICIPANTS A randomized clinical trial performed at a university-based research clinic, between 2005 and 2012. Eighty-six patients with recent onset of schizophrenia were randomized to receive long-acting injectable risperidone or oral risperidone. Half of each group was simultaneously randomized to receive cognitive remediation to improve cognitive functioning or healthy-behaviors training to improve lifestyle habits and well-being. An intent-to-treat analysis was performed between October 4, 2012, and November 12, 2014. INTERVENTIONS A 12-month trial comparing the long-acting injectable vs oral risperidone and cognitive remediation vs healthy-behaviors training. MAIN OUTCOMES AND MEASURES Psychotic relapse and control of breakthrough psychotic symptoms. RESULTS Of the 86 patients randomized, 3 refused treatment in the long-acting injectable risperidone group. The psychotic exacerbation and/or relapse rate was lower for the long-acting risperidone group compared with the oral group (5%vs 33%; ρ21 = 11.1; P < .001; relative risk reduction, 84.7%). Long-acting injectable risperidone better controlled mean levels of hallucinations and delusions throughout follow-up (β = .0.30; t68 = .2.6, P = .01). The cognitive remediation and healthy-behaviors training groups did not differ significantly regarding psychotic relapse, psychotic symptom control, or hospitalization rates, and there were no significant interactions between the 2 medications and the 2 psychosocial treatments. Discontinuations owing to inadequate clinical response were more common in the oral group than in the long-acting risperidone group ( ρ21 = 6.1; P = .01). Adherence to oral risperidone did not appear to differ before randomization but was better for the long-acting risperidone group compared with the oral group (t80 = 5.3; P < .001). Medication adherence was associated with prevention of exacerbation and/or relapse ( ρ21 =11.1; P = .003) and control of breakthrough psychotic symptoms (β = 0.2; t79 = 2.1; P = .04). CONCLUSIONS AND RELEVANCE The use of long-acting injectable risperidone after a first episode of schizophrenia has notable advantages for clinical outcomes. The key clinical advantages are apparently owing to the more consistent administration of the long-acting injectable. Such formulations should be offered earlier in the course of illness.