Signs of Drug Diversion in Hospitals

Ask a pharmacy director how diversion gets caught and the honest answer is usually "by chance." A colleague notices something. A patient complains their pain never improved. A count comes up short one too many times. The signs were in the data all along; nobody was in a position to see them.

The most dependable signs of diversion are transaction patterns, and they fall into a handful of categories that a monitoring program can watch for systematically.

Dispensing irregularities

• Dispense without administration. A controlled substance leaves the automated dispensing cabinet, but no administration is ever documented. The single most direct red flag.
• Multiple dispenses in a short window. One staff member dispensing controlled substances for multiple patients within minutes deserves a look.
• Repeated max-dose dispensing. Always pulling the maximum allowed dose, especially when the pattern repeats shift after shift.

Administration irregularities

• Administration without a documented dispense. The medication appeared from somewhere.
• Different dispense and administration users. The person who pulled the drug is not the person who gave it, without a handoff that explains why.
• Missing pain score documentation. Administration of pain medication with no corresponding pain assessment, or pain scores that follow an abnormal documentation pattern.

Reconciliation failures

• Balance sheet discrepancies. Doses administered plus wasted plus returned should equal the amount dispensed. When it does not, something happened in between.
• Whole dose waste or return. Entire doses regularly "wasted" is one of the classic patterns, especially when paired with the next category.

Witness irregularities

• Witness off shift at the time of the waste. A waste witnessed by someone who was not working is not a witness.
• Witness in the wrong location. Same problem, geographic version.

Shift and location anomalies

• Off-shift activity. Medication activity by a user who is not currently on shift. This is only detectable when timekeeping data is correlated with dispensing data, which is exactly why many programs miss it.
• Wrong-location activity. A user working in a different location than the patient they dispensed for.

Patient status anomalies

• Activity on discharged or expired patients. Dispensing against a patient who left hours earlier is a strong indicator that the medication had another destination.
• Insulin without a glucose result. Insulin administered or dispensed with no corresponding glucose test is its own specific risk pattern.

Why individual signs are not enough

Any one of these events has innocent explanations: charting lag, a hectic code, a training gap. That is what makes manual review so draining. Peterson Health found that spot checks left significant gaps and that standard EHR and ADC reports could not capture the full picture.

The signal is in combination and repetition: the same user, multiple risk types, the same gaps recurring over time. That is the job of drug diversion monitoring software: DetectRx monitors 28+ risk types continuously, scores every flagged event by risk type, drug, and user history, and classifies results as High (80+), Medium (60-79), or Low (below 60) so investigators start where it matters. Renown Health went from monitoring 5 risk factors manually to 24 with continuous automated review.

There is a second-order benefit: reviewing every outlier surfaces workflow problems that are not diversion at all. In Peterson Health's words, DetectRx "found practices that may have needed to be improved or practices that we didn't know existed."

If you are evaluating your own program, score it against this list. Every category your current process cannot see is a category a diverter can use.

Risk types and customer results in this article come from the DetectRx whitepaper.