5PSQ-118 Intelligent infusion pumps for cytostatic drug delivery: development of a drug library and first steps of implementation

<h3>Background and Importance</h3> The incorporation of smart infusion pump technology allows the incorporation of medication error reduction software that includes a drug library. <h3>Aim and Objectives</h3> The aim of this study is to design and implement a drug library for the administration of intravenous oncohaematological treatments in an oncohaematology day hospital for its incorporation into the software of the intelligent pumps and to carry out initial monitoring. <h3>Material and Methods</h3> Prospective, single-centre, multidisciplinary intervention study comprising pharmacists, IT staff and nurses. Pre-intervention phase: installation, drug library design and staff training from November 2022 to February 2023. Intervention phase: implementation of the software, monitoring of alerts and modification of alerts until March 2023. Software was initially installed on two pumps and upgraded weekly to full 27 pumps. Plum 360 15.1x pumps with Hospira MedNet software were used. <h3>Results</h3> In the pre-intervention phase, relative and absolute limits per drug were defined. The maximum possible dose was calculated for a body surface area of 2m2 or 100kg. For drugs with different doses per protocol, the maximum dose was chosen. With this calculation, the rates (mL/h) for the relative upper limits were defined. The absolute upper limit was calculated by adding 1% to the relative upper limit. Lower limits were only defined for the following drugs: vinblastine, vincristine, epirubicin, bendamustine, etoposide, methotrexate. In the intervention phase, a total of 85 alerts were detected, 47 from absolute upper limits, 38 from relative upper limits and 0 from lower limits. Of the total number of alerts, 60 (70.5%) were caused by inadequate selection of the entry, the remaining 25 (29.5%) by too strict limits. The limits were adjusted and all alerts were reviewed with the nursing staff. The final pharmacotheca consisted of 87 entries made up of 58 drugs. 2701 infusions were performed, of which 263(9.7%) were performed without drug definition. <h3>Conclusion and Relevance</h3> The design of the drug library requires a thorough review of existing protocols and drugs. Lower limits were defined for drugs administered as bolus, vesicants or when their usual rate of administration was 999mL/h. The implementation was considered successful, its monitoring has allowed corrective measures to be established and has contributed to increasing patient safety. <h3>References and/or Acknowledgements</h3> <h3>Conflict of Interest</h3> No conflict of interest