The object of this study is to assess 1) the relationship between plasma antipsychotic drug concentration, serum prolactin levels and the clinical efficacy of risperidone, 2) the relationship between the CYP2D6 polymorphisms and metabolizing of risperidone and 3) the role of 9-hydroxyrisperidone in elevating prolactin levels. One-hundred and eighteen Chinese schizophrenia patients (40 males, 78 females, age 15-60 years) were given risperidone at dosages ranging from 2-8 mg/day for 8 weeks. Clinical efficacy was determined using the Brief Psychiatric Rating Scores (BPRS). Serum prolactin levels were assayed before and after the 8 week treatment and plasma risperidone and 9-hydroxyrisperidone levels were also measured at the end of the 8-week treatment. The results showed there was no significant correlation between the concentration of active moiety and clinical response. Risperidone treatment significantly increased serum prolactin levels. Furthermore, changes of prolactin levels were not correlated with the clinical response. For the risperidone/ 9-hydroxyrisperidone ratio, there was a statistically significant difference among the CYP2D6*1/*1, *1/*10, *10/*10 genotypes (Kruskal-Wallis test, p = 0.012). No significant differences were found in the concentration of 9-hydroxyrisperidone and active moiety among the genotypes. In addition, the concentration of 9-hydroxyrisperidone was not significantly correlated with the increase of serum prolactin. In conclusion, our study has, for the first time, produced evidence that in Chinese schizophrenic patients, the metabolism of risperidone is dependent on CYP2D6. Neither changes in serum prolactin levels nor plasma concentration of active moiety were significantly correlated with clinical efficacy of risperidone. 9-hydroxyrisperidone may not play a predominant role in elevating serum prolactin level.