What Happened

A nurse anesthetist at Erlanger Baroness, the largest hospital in Chattanooga, Tennessee, admitted diverting unused fentanyl left over after surgical procedures for months — while the hospital's AI-powered medication monitoring system failed to flag the missing drugs.

According to a Tennessee Board of Nursing consent order, co-workers reported in June 2025 that the nurse appeared impaired while on duty in the surgery center — slurred speech, extreme tiredness, swaying while standing, and difficulty keeping his eyes open. In the days that followed he failed a drug test and was fired.

When questioned by state investigators, he admitted he began diverting "unused fentanyl that would otherwise have been wasted after surgical procedures" in March 2025. He used the fentanyl waste once or twice a week at first, then daily by June of that year. Erlanger audited his dispensing record over those four months and found approximately five instances where the automated monitoring system did not flag missing drugs, plus "additional inconsistencies between drug dispensing and waste documentation that should have been flagged."

He signed the consent order settling the state investigation in November 2025, and it was released publicly in December 2025. The nursing board placed his license on probation while he completed drug counseling. He has not been charged with a crime.

What Went Wrong

The case is notable because the hospital had invested in an AI-powered medication monitoring system — the newest line of defense against drug diversion — and it failed to raise alarms for months. Experts interviewed about the case said the theft of leftover drug waste is one of the most well-known diversion methods, making the missed detections hard to explain. Key failure points:

  • Over-reliance on automated surveillance. The monitoring system was treated as the safety net, but it missed missing drugs and dispensing-versus-waste inconsistencies for roughly four months. Automated tools surface signals; they do not replace human investigation.
  • Operating room blind spot. Drug diversion experts note that automated monitoring is generally more effective in emergency departments and ICUs than in operating rooms, where drugs are dispensed and charted differently. Waste in the OR is a known, high-risk diversion vector that requires focused controls.
  • Weak waste documentation. Leftover fentanyl that should have been wasted with a witness was instead available to be taken. The audit found "inconsistencies between drug dispensing and waste documentation" — the exact pattern waste-witnessing is designed to prevent.
  • Humans caught what the software missed. The diversion was detected by coworkers observing impairment, not by any surveillance signal. Behavioral red flags remain a first-line detection mechanism.

How It Could Have Been Prevented

  • Do not rely on automated monitoring alone. Pair software alerts with a named human analyst who reviews, triages, and investigates every discrepancy — especially in the OR and procedural areas.
  • Enforce witnessed waste for all leftover controlled substances after surgical procedures, and reconcile waste documentation against dispensing records on a daily basis.
  • Build unit-specific surveillance: monitor waste rates, documented versus administered doses, and after-hours or unassigned activity in anesthesia workflows, not just general dispensing patterns.
  • Require an independent review of the monitoring system's output, including testing it against known diversion scenarios, and do not assume a "learning phase" excuses missed detections.
  • Train anesthesia and perioperative staff on behavioral signs of impairment — slurred speech, excessive fatigue, head nodding — and create a safe channel to report concerns.
  • Investigate every discrepancy promptly. A single unexplained variance, left unaddressed, can become a months-long diversion pattern.

Related Guidance

Sources