An invented account, built from the way record-based diversion in a hospital pharmacy vault is typically uncovered during a routine inventory. No part of it describes a real facility or person.
The setting
The central pharmacy of a regional hospital. Controlled substances are held in a vault and issued to cabinet refills, compounding and the operating rooms. One pharmacy technician, long-tenured and well liked, is the designated controlled substance technician: he receives deliveries, maintains the perpetual inventory, stages expired and returned product for the reverse distributor, and prepares the paperwork for anything destroyed on site. A pharmacist signs off on each of those steps. The vault is counted at every shift change, and it always balances.
The first signal
The biennial inventory. The pharmacy director assigned it, as policy required, to a pharmacist and a technician who did not normally work the vault. They counted everything on hand at close of business, exact counts for Schedule II, and compared the result with the perpetual. Almost everything matched. One product did not: a concentrated hydromorphone injection used mostly for patient-controlled analgesia pumps. The shelf count was lower than the perpetual by a quantity too large to be a miscount. A recount by the same pair confirmed it.
The look
Because the shift counts had balanced for two years, the director's first question was how the perpetual could be right every day and wrong at the end. The answer had to be that the perpetual and the shelf had moved apart recently, or that the shift counts had not been counting what they claimed to. She did what a DEA investigator would do: an accountability audit on that one product from the previous biennial inventory forward. Opening balance from the last inventory, plus every receipt from the order records and invoices, minus every disposition from the issue logs, cabinet refills, returns and destruction records, compared with the current count.
The reconstruction showed the gap had accumulated gradually, not in one event. And the dispositions that absorbed it were concentrated in one category: destruction. Over the period, the product had appeared on the on-site destruction records more often than any comparable drug, usually in small quantities described as expired or damaged, each entry signed by the controlled substance technician and witnessed by a pharmacist. On paper, everything was in order.
The confirmation
The director compared the internal destruction records with the records of the reverse distributor that collected the hospital's waste for the portion not destroyed on site, and with the hospital's own on-site destruction log. The reverse distributor's receiving records listed quantities for the product smaller than the quantities the internal paperwork said had been staged for pickup. For the on-site destructions, the witnessing pharmacists, interviewed one at a time, described the same routine: the technician would bring the sealed destruction container and the completed form to the pharmacist's desk for signature. The pharmacist signed as a witness to a form, not to a destruction. None had watched the product rendered unusable, and none had counted what went into the container.
The director then looked at the perpetual's edit history. The product's balance had been reduced on each destruction date by the quantity on the form. The shift counts had balanced because the perpetual had been reduced before each count, and the vials had left the shelf at the same time, into the destruction container and then, the director now believed, out of the building. The shift count was comparing the shelf with a record the same person controlled.
What the hospital did
The director treated the day the accountability audit identified an unexplained loss as the date of discovery. The written notice went to the local DEA office within one business day, and the Form 106 was filed within the 45-day window with the quantity from the reconstruction. The loss was reported to the chief executive under the hospital's policy and to the state board as its rules required. The technician was removed from controlled substance duties that day; his vault access and his role as the destruction preparer were ended, and the perpetual edit rights were restricted to two pharmacists. The interview, with human resources and counsel present, came after the reverse distributor records had been obtained in writing.
What would have surfaced it sooner
- Witnesses who witness. The destruction rule requires two employees to witness the destruction until the drug is non-retrievable, and to sign the record. Signing a form at a desk is not that. Every pharmacist involved had signed in good faith and none had understood the signature as an attestation of having watched.
- Destruction as a trended disposition. One product appearing on destruction records far more often than its peers is visible in a quarterly report of dispositions by type and product. Nobody ran one.
- Matching internal destruction records to the reverse distributor's receipts. The two records should agree, line by line. The first time anyone compared them was two years in.
- Segregation of duties in the vault. One person received, maintained the perpetual, staged returns and prepared destruction. Each of those is a point where a quantity is decided, and the same person decided all of them. The perpetual inventory pitfalls guide lists the failure modes this created.
- Independent biennial counters. The policy that put outsiders on the biennial inventory is the reason the loss was found at all. The lesson the hospital drew was to apply the same independence to a quarterly reconstruction of a few high-risk products, rather than waiting two years.
The record was too neat. A real destruction history has gaps, odd quantities and the occasional missing signature. This one had none, because one person wrote all of it.
This narrative is an illustrative composite. No real person, hospital, reverse distributor or event is described.