Multidisciplinary system for detecting medication errors in antineoplastic chemotherapy

Abstract

Objective: To analyze medication errors (MEs) in a multidisciplinary system with a Computerized Pharmacotherapy Process (CPP) in cancer patients.

Design: A longitudinal, prospective 2-year (January 2003 -to December 2004) cohort study was made in adult patients administered antineoplastic treatment in Services of Oncology and Haematology. MEs were identified by double cross-validation of each stage of the pharmacotherapeutic process (prescription, preparation, dispensing, administration, and follow-up) carried out by the multidisciplinary team (physician, pharmacist, nurse) with CPP assistance.

Variables: Number of MEs per 1000 patient-days, percentage according to the stage of the pharmacotherapeutic process and the severity of intercepted ME (scored from 1 = no damage to the patient, to 5 = patient death).

Results: A total of 1311 patients were receiving treatment, and MEs were identified in 225. Out of a total of 13,158 patient-days, 276 MEs were detected, equivalent to 20.9 MEs per 1000 patient-days; of these, 16.8 MEs per 1000 patient-days (80%) were intercepted and did not affect any patient. The detected ME distribution according to pharmacotherapeutic stage was: prescription 75.7%, preparation 21.0%, dispensing 1.8%, administration 1.1%, and follow-up 0.4%. ME distribution according to severity was: grade 1 : 15.9%, grade 2 : 49.6%, grade 3 : 33.7%, grade 4 : 0.7%, and grade 5 : 0%. The system intercepted 98.9% of all MEs with severity >or=3 (MEs with a potential for causing patient damage).

Conclusions: The multidisciplinary system with a well-established CPP detects 20.9 MEs per 1000 patient-days and intercepts 98.8% of all MEs with a potential for causing patient damage.