Disclaimer: This is a user generated content submitted by a member of the WriteUpCafe Community. The views and writings here reflect that of the author and not of WriteUpCafe. If you have any complaints regarding this post kindly report it to us.
Invistics, the leading provider of cloud-based software solutions that improve inventory visibility and analytics across complex healthcare systems and global supply chains, announced the publication of its groundbreaking scientific study, “Detecting Drug Diversion in Health System Data using Machine Learning and Advanced Analytics,” in the latest edition of the peer-reviewed American Journal of Health-System Pharmacy.
The retrospective study, published Feb. 9, compared the use of machine learning and advanced analytics software to uncover drug diversion with traditional means of detection — including the use of monthly anomalous usage report — within 10 acute-care inpatient hospitals across four independent health systems.
Study co-authors included: Tom Knight (CEO, Invistics Corporation), Bernie May (SVP Healthcare Systems, Invistics Corporation), Don Tyson (Director of Pharmacy, Piedmont Athens Regional Medical Center; Scott McAuley (Executive Director of Pharmacy, Piedmont Healthcare) Pam Letzkus (Senior Director of Pharmacy, Scripps Health); and Sharon Murphy Enright (Principal, EnvisionChange, LLC).
Researchers extracted two datasets from each participating health organization’s health technology systems, and then applied supervised machine learning models to analyze 24 months of historical data. This included 27.9 million medication movement transactions by 19,037 nursing, 1,047 pharmacy, and 712 anesthesia clinicians — and 22 known, blinded diversion cases. Researchers then gauged when the machine learning model would have found, or detected, evidence of those known diversion cases.
The advanced analytics and machine learning technologies detected known diversion cases (n=22) in blinded data an average of 160 days faster than existing, non-machine learning detection methods had. Additionally, the machine learning model demonstrated 96.3% accuracy, 95.9% specificity, and 96.6% sensitivity detecting transactions at high-risk of diversion in the dataset.
“For healthcare systems that don’t yet utilize a drug prevention and detection program leveraging machine learning and advanced analytics tools, the research speaks for itself,” said Don Tyson, Director of Pharmacy at Piedmont Athens Regional Medical Center. “Advanced analytics and machine learning technology can improve the accuracy, efficiency, and effectiveness of any drug diversion prevention program and goes far beyond what can be addressed manually, especially when dealing with large amounts of data.”
“Identifying drug diversion quickly is critical to patient safety. Advances in technology have made it possible to detect and investigate potential diversion months earlier,” said Pam Letzkus, Senior Director of Pharmacy at Scripps Health. “As such, the research has big implications for patients and healthcare providers.”
Drug diversion in healthcare settings is a silent, yet pervasive epidemic with financial, clinical, and legal consequences. A May 2021 Porter Research survey indicates 96% of healthcare workers agree that drug diversion is occurring in U.S. hospitals, and nearly three-quarters (73%) of survey participants rated machine learning as an effective tool in identifying or preventing drug diversion — up from 65% in 2019.
“The findings prove that advances in machine learning and analytics are a real gamechanger – and can improve the detection of drug diversion in hospitals and other healthcare settings,” says Tom Knight, CEO, Invistics. “This is really important, considering the huge financial, clinical, and emotional burden that medication theft imposes on healthcare systems, patients, and families.
