Flaw in AI medication monitoring software comes to light after drug diversion in Tenn. hospital

 Hundreds of hospitals use AI tools to identify drug diversion in their respective facilities, but a recent case at a Tennessee hospital highlights some of the shortcomings of this system, CBS News reported June 1.

Here are five things to know.

1. In 2025, a nurse at Chattanooga, Tenn.-based Erlanger Baroness Hospital was fired after administrators discovered he was using fentanyl left over after surgeries, according to a Tennessee Board of Nursing consent order. During a review, leaders found the nurse had been diverting fentanyl for four months. An audit of his dispensing records found five instances when the AI should have flagged missing drugs but did not. The review also found “additional inconsistencies between drug dispensing and waste documentation that should have been flagged by the automated monitoring system,” CBS News reported.

2. Drug diversion, defined as unlawfully taking a controlled substance from healthcare facilities, is a widespread challenge in medical facilities. It is estimated that up to 15% of healthcare workers divert drugs at least once, according to the Healthcare Diversion Network. Diversions have been linked to 13 disease outbreaks since 1985, the CDC found. Hospitals must confidentially report lost or stolen drugs to the Drug Enforcement Administration; they also can also report stolen drugs to state health agencies.

3. In recent years, hospitals have increasingly turned to AI-powered medication-monitoring software to identify missing medications. A 2022 study found that Sentri7, a commonly used software, could identify drug diversion faster than existing methods, but said the AI was not able to find all cases of diversion.

4. Two programs dominate the medication-monitoring field: More than 1,500 hospitals use Bluesight’s ControlCheck, and an additional 700 use Wolters Kluwer’s Sentri7 Clinical Surveillance programs.

5. Because AI technology is heavily proprietary, hospital officials often do not understand how it works. This lack of transparency can allow errors to be buried, leading to repeated instances at other facilities, David Rastall, DO, PhD, a neurologist and AI researcher at Baltimore-based Johns Hopkins Medicine, told CBS News. In this instance, it is unclear why Sentri7 did not catch the diversion. The hospital, software company and other investigating organizations involved declined CBS News’ request for comment.

https://www.beckershospitalreview.com/legal-regulatory-issues/a-flaw-in-ai-medication-monitoring-software-comes-to-light-after-drug-diversion-in-tennessee-hospital/