Provider risk factors for medication administration error alerts: analyses of a large-scale closed-loop medication administration system using RFID and barcode

• Published in: Pharmacoepidemiology and drug safety Vol. 25; no. 12; pp. 1387 - 1396
• Main Authors: Hwang, Yeonsoo, Yoon, Dukyong, Ahn, Eun Kyoung, Hwang, Hee, Park, Rae Woong
• Format: Journal Article
• Language: English
• Published: England Blackwell Publishing Ltd 01.12.2016; Wiley Subscription Services, Inc
• Subjects: Automatic Data Processing; closed-loop medication administration; Humans; Logistic Models; Medical errors; Medical Order Entry Systems; medication error; Medication Errors - prevention & control; Medication Errors - statistics & numerical data; Medication Systems, Hospital; Nurses - organization & administration; patient safety; Pharmaceutical Preparations - administration & dosage; pharmacoepidemiology; point-of-care system; Point-of-Care Systems; Prescription drugs; Radio frequency identification; Radio Frequency Identification Device; radio-frequency identification (RFID); Risk Factors; Time Factors; Work Schedule Tolerance; point-of-care system; patient safety; closed-loop medication administration; radio-frequency identification (RFID); medication error; pharmacoepidemiology
• ISSN: 1053-8569; 1099-1557; 1099-1557
• DOI: 10.1002/pds.4068

Purpose To determine the risk factors and rate of medication administration error (MAE) alerts by analyzing large‐scale medication administration data and related error logs automatically recorded in a closed‐loop medication administration system using radio‐frequency identification and barcodes. Methods The subject hospital adopted a closed‐loop medication administration system. All medication administrations in the general wards were automatically recorded in real‐time using radio‐frequency identification, barcodes, and hand‐held point‐of‐care devices. MAE alert logs recorded during a full 1 year of 2012. We evaluated risk factors for MAE alerts including administration time, order type, medication route, the number of medication doses administered, and factors associated with nurse practices by logistic regression analysis. Results A total of 2 874 539 medication dose records from 30 232 patients (882.6 patient‐years) were included in 2012. We identified 35 082 MAE alerts (1.22% of total medication doses). The MAE alerts were significantly related to administration at non‐standard time [odds ratio (OR) 1.559, 95% confidence interval (CI) 1.515–1.604], emergency order (OR 1.527, 95%CI 1.464–1.594), and the number of medication doses administered (OR 0.993, 95%CI 0.992–0.993). Medication route, nurse's employment duration, and working schedule were also significantly related. Conclusion The MAE alert rate was 1.22% over the 1‐year observation period in the hospital examined in this study. The MAE alerts were significantly related to administration time, order type, medication route, the number of medication doses administered, nurse's employment duration, and working schedule. The real‐time closed‐loop medication administration system contributed to improving patient safety by preventing potential MAEs. Copyright © 2016 John Wiley & Sons, Ltd.