Dispensing errors in community pharmacy: frequency, clinical significance and potential impact of authentication at the point of dispensing

Abstract

Abstract Objective Our aims were to explore the nature of dispensing errors in community pharmacy, and investigate the potential impact of authentication at the point of dispensing using barcodes or radio-frequency identification tags. Objectives were to develop a dispensing error definition for use in the community setting, to describe the incidence, types and clinical significance of dispensing errors in UK community pharmacies and to assess the likely impact of authentication at the point of dispensing. Setting Eleven UK community pharmacies. Method A definition of a dispensing error was developed using the Delphi technique. A research pharmacist checked dispensed items awaiting collection and compared them against the original prescription to identify dispensing errors. An expert panel assessed the potential clinical significance of the errors identified. We then predicted the likely impact of three different systems of authentication at the point of dispensing: (a) stand-alone; (b) linked to patient medication records (PMR); and (c) linked to electronic transfer of prescriptions (ETP). Key findings We found a content error in 49 (1.7%) of 2859 dispensed items, and a labelling error in 46 (1.6%). The majority (67%) were of minor clinical significance. We estimated that a stand-alone system would prevent about one in five content errors and very few labelling errors, a PMR-linked system would prevent one-quarter of content errors and one-third of labelling errors, and an ETP-linked system would prevent nearly half of content and labelling errors. While none of the three systems would have prevented the one serious error identified, 22–60% of moderate errors could have been prevented, depending on the system used. Conclusion Content errors were identified in 1.7% of dispensed items, and labelling errors in 1.6%. Authentication at the point of dispensing could prevent a significant proportion of these, particularly if the system were linked to PMR or ETP. An experimental study is now needed.

Ähnliche Arbeiten