AI flagged nurse stealing drugs; hospital ignored alerts
Get the Health newsletter
Daily health & science — research, biotech, public health, the studies worth knowing. Free.
- Adventist Health in Bakersfield, CA had a travel nurse in late September 2024 who was observed walking barefoot through the ICU, talking to herself, and acting abrasively toward patients and families
- The nurse stole medications from a secured cabinet for personal use, including fentanyl and morphine that patients believed they were receiving through IV drips that provided no relief
- Hospital managers at Adventist Health ignored alerts from machine learning software designed to flag potential drug theft by staff, according to auditors
- CMS investigators investigated the incident in November 2024 after a complaint, finding the nurse had documented the drugs as administered to patients while actually self-administering them
- AI drug-tracking technology can identify employee theft and connect individuals with substance abuse resources, but the Adventist case shows the systems require humans to act on the warnings
- Hospitals serve as major repositories of addictive substances that also function as crucial medical treatments, creating ongoing risk of diversion by staff
Why it matters: A machine learning system flagged a nurse stealing drugs at Adventist Health, but managers ignored the warnings — letting a compromised clinician treat ICU patients for weeks. For hospitals investing in AI surveillance to curb drug diversion, the lesson is blunt: the technology is only as effective as the staff willing to review its alerts.
Ask SkimNews
