Sir: Poor compliance with antipsychotic medication, which would potentially lead to disease relapse, has been challenging for psychiatrists when treating schizophrenia.1 Long-acting injection of antipsychotics is an appropriate alternative, since better compliance improves treatment outcomes. Mirror-image studies, in which each patient acts as his/her own control, of long-acting injection of conventional antipsychotics have shown significant decreases in numbers of hospitalizations and days of hospitalization. The development of atypical antipsychotics has provided a new treatment paradigm based on their superior tolerability, if not efficacy. Risperidone long-acting injection (RLAI) is the first licensed long-acting injectable atypical antipsychotic agent and has recently been reported cost-effective by reducing total admission number and inpatient days in a communitybased inpatient setting.3 To our knowledge, a national claimbased database has never been used in any mirror-image study for RLAI.Method. The data source used for this 6-month mirror image study was the Psychiatric Inpatients Medical Claims Data (PIMC) from the National Health Research Institute, Taiwan. The PIMC compiled all the health care utilization records during 1996-2006 for patients who had at least 1 psychiatric hospitalization during 1996-2001. The inclusion criteria required that patients (1) could be observed at least 6 months after the first dose of RLAI, (2) had a primary diagnosis of schizophrenia, and (3) were continuously treated with RLAI for at least 6 months. Patients who received at least 75 mg RLAI total for a 3-month time period were considered continuously treated. The differences in number of acute admissions, hospital days, and emergency room visits between the pre-and post-RLAI periods were compared.Results. A total of 253 from 91,104 patients met the inclusion criteria. As compared to the 6-month pre-RLAI period, the total number of acute admissions was reduced by 35% (136 vs. 88 times, p = .0007), and total hospital stays were reduced by 47% (5856 vs. 3080 days, p = .0002) in the 6-month post-RLAI period. A reduced number of emergency room visits was also observed (80 vs. 67 times) but was not significantly different (p = .24). Since the average hospital stay in acute psychiatric settings was 33 days in Taiwan (data on file; Department of Health, Executive of Yuan, Taiwan; 2007), a secondary analysis to eliminate prolonged hospitalization was conducted by excluding patients who stayed longer than 90 days per admission in the pre-RLAI period. The case number was therefore slightly decreased (N = 237), but the differences in acute admissions (115 vs. 80, p = .0010) and hospital days (3701 vs. 2160 days, p = .0026) remained significant.