Seventy-two patients fulfilling the DSM-III criteria for schizophrenia and schizophreniform psychosis were admitted to a multicentre, double-blind controlled study to evaluate the efficacy and safety of remoxipride in comparison to haloperidol. The mean daily dose of remoxipride at the end of treatment was 353 mg and of haloperidol, 11 mg. Patients were assessed each week on the Brief Psychiatric Rating Scale (BPRS), the Clinical Global Impression (CGI) and the symptoms checklist. No significant differences in efficacy were found between the two treatments. The median total BPRS score in the remoxipride group was 25 at start of active treatment and 17 at the last valid rating (n = 31). For the haloperidol group the corresponding figures were 24 and 15 (n = 29). According to the CGI, 40% of remoxipride patients and 50% of haloperidol patients were much or very much improved. Treatment-emergent extrapyramidal symptoms, such as akathisia and rigidity, occurred significantly more frequently, and were more severe during treatment with haloperidol than with remoxipride (p = 0.012 and 0.024, respectively). Haloperidol-treated patients reported significantly more drowsiness and increased sleep during treatment (p = 0.026 and 0.012, respectively). No statistically significant differences were seen in endocrine or autonomic symptoms. Remoxipride seemed to be as effective as haloperidol, had a lower frequency of side effects, and was used safely in the dose range 150-600 mg/day.