STUDY OBJECTIVE: To measure the rate of dispensing errors and to identify the types and sources of dispensing errors in a highly automated mail-service pharmacy practice. DESIGN: Descriptive analysis of a random sample of completed prescriptions. SETTING: A high-volume mail-service pharmacy practice comprising a network of prescription processing and dispensing pharmacies in the United States. MEASUREMENTS AND MAIN RESULTS: During September and October 2003, new and refill prescriptions were retrieved before shipping and evaluated for dispensing accuracy. Container contents were compared against the container label, and the label record was compared against the original prescription order. The overall dispensing error rate was 0.075% (16 dispensing errors among 21,252 prescriptions, 95% confidence interval 0.043-0.122). Fourteen errors involved incomplete or incorrect directions on the final label. All dispensing errors were associated with the initial stages of prescription processing (including order entry); no errors were associated with the mechanical stages of product dispensing. CONCLUSION: A highly automated mail-service pharmacy can achieve a dispensing error rate of less than 1 error/1000 prescriptions, which is substantially lower than the rates reported for retail pharmacies. A high degree of automation in the mechanical aspects of dispensing appears to be a key factor in achieving this high dispensing accuracy.
STUDY OBJECTIVE: To measure the rate of dispensing errors and to identify the types and sources of dispensing errors in a highly automated mail-service pharmacy practice. DESIGN: Descriptive analysis of a random sample of completed prescriptions. SETTING: A high-volume mail-service pharmacy practice comprising a network of prescription processing and dispensing pharmacies in the United States. MEASUREMENTS AND MAIN RESULTS: During September and October 2003, new and refill prescriptions were retrieved before shipping and evaluated for dispensing accuracy. Container contents were compared against the container label, and the label record was compared against the original prescription order. The overall dispensing error rate was 0.075% (16 dispensing errors among 21,252 prescriptions, 95% confidence interval 0.043-0.122). Fourteen errors involved incomplete or incorrect directions on the final label. All dispensing errors were associated with the initial stages of prescription processing (including order entry); no errors were associated with the mechanical stages of product dispensing. CONCLUSION: A highly automated mail-service pharmacy can achieve a dispensing error rate of less than 1 error/1000 prescriptions, which is substantially lower than the rates reported for retail pharmacies. A high degree of automation in the mechanical aspects of dispensing appears to be a key factor in achieving this high dispensing accuracy.