Misordered dosing is a serious and preventable public health problem. Extensive deployment of electronic health records and computerized order entry systems has significantly reduced medication order errors and inefficiencies in inpatient settings. However, new research also introduces new sources of error related to the interaction between the provider and the platform.
Meanwhile, for dosage Order errors, manual reviews of incoming pharmacy orders are the “gold standard” for improving drug use and minimizing prescription errors, and are hospital-based manual reviews of drug orders. Clinical pharmacist In addition, computerized dosing ordering by doctors can be affected by factors such as warning fatigue and can lead to malpractice.
To address these errors and inefficiencies, a team led by Martina Balestra, a former postdoc and adjunct professor at the Center for Urban Sciences and Progress (CUSP) at New York University’s Tandon Institute of Technology, includes Professor Oded Nov. In the field of technology management and innovation of NYU Tandon, and Ji Chen, Eduardo Iturrate, Yindalon Aphinyanaphongs of NYU Grossman and NYU Langone, Machine learning model Identify drug orders that require pharmacy intervention, using only provider behavior and other contextual features that may reflect these new causes of inefficiency, rather than the patient’s medical records. ..
Their study, “Prediction of Inpatient Pharmacy Order Interventions Using Provider Behavior Data,” recently published at JAMIA Open, used a major metropolitan hospital system as a case study. The team collected data on the provider’s behavior in EHR systems and pharmacy orders.Using this dataset, researchers then built machine learning-based classifications. model To identify orders that are likely to require pharmacist intervention.
The classification model developed by the team focuses on clinician data, whereas previous models that predict dosing sequence errors capture data from patient medical records. This reduces the risk to patient data privacy and security. With proper adjustment, this model and similar models can significantly reduce the workload of pharmacists and increase patient safety.
Martina Balestra et al, Prediction of Inpatient Pharmacy Order Intervention Using Provider Action Data, Jamia Open (2021). DOI: 10.1093 / jamiaopen / ooab083
NYU Tandon Institute of Technology
Quote: Researchers ordered an inpatient pharmacy for intervention obtained on October 19, 2021 from https: //medicalxpress.com/news/2021-10-flagging-inpatient-pharmacy-intervention.html (2021) October 19th) I found a new way to flag
This document is subject to copyright. No part may be reproduced without written permission, except for fair transactions for personal investigation or research purposes. The content is provided for informational purposes only.
Researchers find novel means of flagging inpatient pharmacy orders for intervention Source link Researchers find novel means of flagging inpatient pharmacy orders for intervention