A centralized automated-dispensing system in a French teaching hospital: return on investment and quality improvement

International Journal for Quality in Health Care, 2019, 31(3), 219–224 doi: 10.1093/intqhc/mzy152 Advance Access Publication Date: 11 July 2018 Article Article A centralized automated-dispensing system in a French teaching hospital: return on investment and quality improvement SARAH BERDOT1,2,3, VIRGINIE KORB-SAVOLDELLI1,3, EMMANUEL JACCOULET1, VINCENT ZAUGG1, PATRICE PROGNON1,4, LAETITIA MINH MAÏ LÊ1,4, and BRIGITTE SABATIER1,2 1 Pharmacy Department, Hôpital Européen Georges Pompidou, APHP, 20 Rue Leblanc, 75015 Paris, France, 2Equipe 22, Centre de Recherche des Cordeliers, UMR 1138 INSERM, 5 Rue de l’Ecole de Médecine, 75006 Paris, France, 3 Faculty of Pharmacy, Clinical Pharmacy Department, Université Paris Sud, EA4123, 5 Rue Jean-Baptiste Clément, 92290 Châtenay-Malabry, France, and 4Lip(Sys)2, EA7357, UFR Pharmacie, U-Psud, Univ. Paris-Saclay, 5 Rue JeanBaptiste Clément, 92290 Châtenay-Malabry, France Address reprint requests to: Sarah Berdot, Pharmacy Department, Hôpital Européen Georges Pompidou, APHP, 20 Rue Leblanc, 75015 Paris, France. Tel: +01-56-09-31-51; Fax: 01-56-09-25-59; E-mail: Editorial Decision 17 May 2018; Accepted 20 June 2018 Abstract Objectives: To evaluate the return on investment (ROI) and quality improvement after implementation of a centralized automated-dispensing system after 8 years of use. Design: Prospective evaluation of ROI; before and after study to evaluate dispensing errors; user satisfaction questionnaire after 8 years of use. Setting: The study was conducted at a French teaching hospital in the pharmacy department, which is equipped with decentralized automated medication cabinets in the wards. Participants: Pharmacy staff (technicians and residents). Intervention(s): Implementation of a centralized automated-dispensing robot. Main Outcome Measure(s): The true ROI was prospectively and annually compared to estimated returns calculated after implementation and upgrade of the robot; dispensing errors determined by observation of global deliveries and the satisfaction of users based on a validated questionnaire were evaluated. Results: Following the upgrade, we found little difference for the ROI (+1.86%). The payback period increased by almost 3 years. There was a significant reduction of dispensing errors, from 2.9% to 1.7% (P < 0.001). User satisfaction of the robot by the pharmacy staff was reported (score of 5.52 ± 1.20 out of 7). Conclusions: These systems are worthwhile investments and largely contribute to improving the quality and safety of the medication process. Key words: centralized automated-dispensing system, return on investment, medication process, satisfaction Introduction Centralized and decentralized automated-dispensing systems have been developed in hospital pharmacies to secure drug dispensing and administration process [1, 2]. In 2014, 37% of surveyed hospitals with more than 600 beds in the United States had a dispensing robot and all had automated-dispensing cabinets [2, 3]. These © The Author(s) 2018. Published by Oxford University Press in association with the International Society for Quality in Health Care. All rights reserved. For permissions, please e-mail: 219 Berdot et al. 220 Materials and Methods Setting The European Georges Pompidou Hospital (HEGP) is a teaching hospital in Paris, France with 714 beds and 106 day hospital places (43 clinical wards). The mean duration of hospitalization is 6.1 days. The hospital is equipped with a patient information system, integrating an electronic health record and a computerized physician order entry system (DxCare®, Medasys), and 50 automated medication cabinets (Omnicell® Inc.). The total number of referenced drugs managed in the pharmacy is 1700. Before implementation of the robot, the pharmacy team for the global medication dispensing process consisted of three pharmacy technicians, two pharmacy technician aides, three pharmacy residents and seven pharmacists. robot. Some drugs are not stored in the robot, such as plasmaderived medicinal products and narcotics drugs, due to weight, large dimensions, lack of a barcode on the drug box, or classification. Comparison of return on investment ROI is commonly used to gauge an investment’s profitability, because of its versatility and simplicity to calculate. An estimated ROI was initially calculated to justify the investment for the hospital financial department. We assessed the ROI of robot implementation by annually comparing the true ROI to the estimated ROI, calculated after implementation in 2008 (initial estimated ROI) and its upgrade in 2013 (estimated ROI with upgrade). Since 2013, the initial estimated ROI and estimated ROI with upgrade were combined. ROIs were calculated by deducting the costs of investment from the cost savings since implementation of the robot. The investment costs included buying and installation of the robot, annual maintenance and repairs, and upgrading of the robot. The cost savings were evaluated by the reduction of drug stocks and the decrease in pharmacy staff dedicated to global dispensing. Quality improvement Dispensing errors We conducted a before and after study to evaluate the benefit of the centralized automated-dispensing robot on dispensing errors. Observations were performed by a pharmacist before robot implementation and after 2 years of use. Twenty prepared global deliveries were examined on five consecutive days. Before implementation, deliveries were manually prepared by pharmacy technicians, whereas they have been ordered from the robot by pharmacy technicians after implementation and manually completed for drugs stored outside the system. We evaluated the mean number and standard deviation (SD) of drug units per global delivery. The error rate was calculated as the number of dispensing errors divided by the total unit doses ordered, multiplied by 100. Medication process Each electronic medication prescription by a physician is reviewed by a pharmacist. Drugs are dispensed by pharmacy technicians. Before implementation of the robot, the technicians manually dispensed drugs via decentralized automated-dispensing cabinets or urgent global deliveries. After implementation, the technicians dispensed drugs using the robot. Cabinets are refilled twice a week per unit. Since 2008, a centralized automated-dispensing robot (Rowa® system from ARX®) was implemented in the pharmacy to store and deliver drugs. Centralized automated-dispensing system The centralized automated-dispensing system consists of tandem robots with a duplicate stock for a total approximate capacity of 22 000 boxes. In 2008, each robot included a specific refrigerated unit, a loading unit (ProLog system) to input drug boxes using barcodes, and a dispensing system. In 2013, the system was upgraded to improve the quality process. In contrast to the first robot, the new system integrated the capacity to read datamatrix available on each drug box. It was equipped with three specific scanners (RO (...truncated)