Event analysis for automated estimation of absent and persistent medication alerts

• Verfasst von: Bittmann, Janina [VerfasserIn]
• Verfasst von: Scherkl, Camilo [VerfasserIn]
• Verfasst von: Meid, Andreas [VerfasserIn]
• Verfasst von: Haefeli, Walter E. [VerfasserIn]
• Verfasst von: Seidling, Hanna [VerfasserIn]
• Titel: Event analysis for automated estimation of absent and persistent medication alerts
• Titelzusatz: novel methodology
• Verf.angabe: Janina A Bittmann, Dr sc hum, Camilo Scherkl, Andreas D Meid, PD Dr sc hum, Walter E Haefeli, Prof Dr med, Hanna M Seidling, Prof Dr sc hum
• E-Jahr: 2024
• Jahr: 04.06.2024
• Umfang: 6 S.
• Illustrationen: Illustrationen
• Fussnoten: Gesehen am 15.11.2024
• Titel Quelle: Enthalten in: JMIR medical informatics
• Ort Quelle: Toronto : [Verlag nicht ermittelbar], 2013
• Jahr Quelle: 2024
• Band/Heft Quelle: 12(2024), Artikel-ID e54428, Seite 1-6
• ISSN Quelle: 2291-9694
• DOI: doi:10.2196/54428
• Datenträger: Online-Ressource
• Sprache: eng
• K10plus-PPN: 1908768231

Abstract

Background: Event analysis is a promising option to estimate the acceptance of medication alerts issued by computerized physician order entry systems with integrated clinical decision support systems (CPOE-CDSS), particularly when alerts cannot be interactively confirmed in the CPOE-CDSS due to its system architecture. Medication documentation is then reviewed for documented evidence of alert acceptance, a time-consuming process, especially when performed manually. Objective: We present a new approach of an automated event analysis and apply it to a large dataset generated in a CPOE-CDSS with passive, non-interruptive alerts. Methods: Medication and alert data generated over 3.5 months within the CPOE-CDSS at Heidelberg University Hospital were divided into 24-hour time intervals in which alert display was correlated with associated prescription changes. Alerts were considered as “persistent” if they were displayed in every consecutive 24-hour time interval due to a respective active prescription until patient discharge and as “absent” if they were no longer displayed during continuous prescriptions in the subsequent interval. Results: Overall, 1,670 patient cases with 11,428 alerts were analyzed. Alerts were displayed for a median of three consecutive 24-hour time intervals with alerts for drug-allergy interactions displayed the shortest, and the longest for potentially inappropriate medication for the elderly (PIM). A total of 56.1 % of all alerts (n = 6,413) became absent, and among them, alerts for drug-drug interactions were the most common (80.9 %, n = 1,915) and PIM alerts the least common (39.9 %, n = 199). Conclusions: This new approach to estimate alert acceptance based on event analysis can be flexibly adapted to the automated evaluation of passive, non-interruptive alerts. This enables large datasets of longitudinal patient cases to be processed, and to derive the ratios of persistent and absent alerts, compare and prospectively monitor them.