Objective: This study compares the efficacy of risperidone and olanzapine to that of first-generation antipsychotics (FGAs) in patients with schizophrenia, who failed to show a response to initial trials of FGAs. Method: This study was an 8-week treatment, randomized, rater-blind, active-control study with 3 treatment arms. 48 patients, who showed inadequate response to 1 FGA, were enrolled and randomized into risperidone, olanzapine, or FGA (haloperidol or trifluoperazine) groups. They were blindly assessed with the Positive and Negative Syndrome Scale (PANSS), the Clinical Global Impression Scale-Severity, and the Extrapyramidal Symptom Rating Scale (ESRS) at baseline and biweekly. Results: All 3 groups demonstrated a significant decrease in the PANSS total, positive, and general scores from baseline to endpoint (p-values range from 0.003 to 0.021). There were no significant differences among the 3 groups in score changes. The olanzapine group had significant score reductions than the risperidone and FGAs groups in terms of the ESRS subjective total score and did not experience a significant increase in the dose of anticholinergics. The FGA group demonstrated that extrapyramidal syndrome (EPS) worsened under an increased dosage of anti-EPS drugs. Olanzapine was associated with significant body weight gain (2.69 +/- 4.0 kg, p = 0.026), but there were no significant group differences on weight gain. Conclusions: Haloperidol or trifluoperazine demonstrated similar efficacy as risperidone or olanzapine for patients with schizophrenia who had failed their first trial with a FGA. Related double-blind, fixed dose studies with a larger sample size are needed to confirm the results of our study.