Error Detection and Cost Savings With an Image-Based Workflow Management System Connected to a Computerized Prescription Order Entry Program for Antineoplastic Compounding

OBJECTIVE: The aim of the study was to analyze both the prevalence of errors with the implementation of an image-based workflow management system during the antineoplastic compounding process, and the estimated costs associated with the negative clinical outcome if the errors had not been intercepted. METHODS: Three months after the implementation of Phocus Rx system at a hospital pharmacy department, the identification, classification (type, preparation stage, and cause), and potential severity degree (from negligible to catastrophic) of the errors intercepted were determined. The probability of an error causing an adverse event if it had reached the patient (from nil [0] to high [0.6]) and its consequences was estimated by a team of clinical pharmacists and physicians. Cost-effectiveness analysis from the hospital's perspective was performed. RESULTS: Overall, 9872 antineoplastic medications were prepared using Phocus Rx. The total compounding error rate was 0.8% (n = 78, 56 [69.2%] were related to incorrect dose, 20 [28.2%] to incorrect drug preparation or conditioning technique, and 2 [2.6%] were wrong drugs). Approximately 70% of the detected errors were classified as undetectable via the previous verification practice, with 11.55% judged to be potentially severe (n = 9) and 51.3% moderate (n = 29). Likelihood of occurrence of an adverse event was medium (0.4) to high (0.6) for 37.2% of the errors. Estimated cost ratio and return on investment were O4.21 and 321%, respectively. CONCLUSIONS: The implementation of Phocus Rx prevented antineoplastic preparation errors that would have reached the patient otherwise. In addition, acquisition of this technology was estimated to be cost-effective.