This study compared the long-term (12 months) effectiveness of risperidone (RP) with that of conventional neuroleptics (CNs) in a population with chronic schizophrenia who had shown suboptimal response to CNs. A randomized, open, parallel, multicenter design was used. One hundred eighty-four subjects meeting DSM-IV criteria for schizophrenia were randomly assigned to receive either RP or a CN, and 165 of them completed the follow-up. Outcome measures were taken at 3, 6, and 12 months and included the Positive and Negative Syndrome Scale (PANSS) and the Extrapyramidal Symptom Rating Scale. Within this 12-month follow-up, RP was found to be superior to CNs in terms of both the average change in score from baseline on the PANSS (p = 0.006) and the proportion of good responders (as defined by a 20% decrease in total PANSS scores;p = 0.03). For positive symptoms, the effectiveness of the RP treatment tended to increase over time. At 12 months, the percentage of good responders in the RP group was twice as large as that in the CN group (30% vs. 15%;p = 0.03). The superiority of RP over CNs was constant over the three dose categories. In both the RP and the CN groups, the maximum decrease in psychopathology was achieved with the lowest dose range. A worsening of akathisia was less frequent in subjects receiving RP than in those receiving CNs (p = 0.02). In conclusion, this study showed that, compared with CNs, RP is beneficial in the treatment of patients with chronic schizophrenia and that some of these benefits may appear only after longer-term treatment.