Medications that are unaccounted for in hospitals: discrepancies in electronic records

Reasons for inconsistencies need to be investigated to reduce the levels of medications that go missing

Undocumented medication use in hospitals potentially poses risks for patients, staff, and the hospital organisation.1,2 A variety of reasons may explain why medications are unaccounted for, including diversion3 and workarounds.4,5

Medications that are unaccounted for are invisible to the hospital unless their supply and use are closely monitored. Supply and use have typically been monitored using paper‐based documentation systems that afford only limited control of medication use, and in regular, resource‐intensive audits. Compared with paper‐based systems, electronic systems provide more stringent control of medication prescribing, dispensing, and administration, decision support functionalities guiding behaviour, and automated documentation. Further, electronic systems afford easier access to data that can be audited to identify patterns of use and discrepancies between records, which can then be investigated. Electronic systems are common in North America and are increasingly used elsewhere, and data extracted from these systems have been used to identify medication supply discrepancies, particularly for monitoring controlled medications.6

The study by Walker and colleagues reported in this issue of the MJA7 assessed the extent to which medications are unaccounted for in hospitals in Victoria. The authors extracted data from electronic dispensing (Merlin, iPharmacy) and prescribing systems (Cerner Millennium) for a sample of medications supplied and administered to patients in two general medical wards, two surgical wards (one general, one orthopaedic), and the emergency departments of each of four Victorian hospitals during 2019. Almost one‐fifth of the medications supplied were not accounted for.7

The medications examined by Walker and her colleagues were among those more frequently prescribed in hospitals. Unaccounted use was particularly high for oral antibiotics, which has important implications for antibiotic stewardship.8 Substantial discrepancies with regard to controlled drugs were identified in individual hospitals, despite governance systems for monitoring their use.9 Discrepancy patterns were otherwise similar in the four hospitals, with levels of missing medications higher in emergency departments than in medical and surgical wards, and for oral medications than for equivalent parenteral formulations.

Walker and colleagues had anticipated that missing medications would impose a financial burden on hospitals, but found that the total cost across the four hospitals was $27 800, or only 0.02% of their 2019 medication budgets. The medications examined in the study were inexpensive, but, as the authors point out, the estimated financial damage did not take into account possible indirect costs (eg, safety‐related incidents), externalities (such as their environmental impact), or societal effects (eg, on antimicrobial resistance).

Why medications were missing was not explored by Walker and her colleagues. The authors suspect that self‐treatment by staff members (or of their friends and families) may explain the undocumented use of oral antibiotics and analgesics. Knowing that these medications are inexpensive may also contribute to clinicians taking them for self‐care.10 The authors also propose that some discrepancies in the supply and use of parenteral medications may be explained by waste, including, I imagine, the dead spaces in syringes.11 They noted that the four hospitals had no “data on doses that were lost, wasted, returned to the pharmacy, or destroyed”, and that the higher levels of missing medications in the four emergency departments might be explained by a variety of factors, including their “hectic environment” and patients being transferred elsewhere with medications. Electronic systems being temporarily unavailable (“downtime”) can also be a reason for incomplete documentation of medication use.

There may indeed be good reasons for unaccounted hospital medication use. There is always a degree of misfit between the workflows recommended, understood, and modelled in medication systems and workflows in real practice.4,12

Electronic systems may give the illusion of control, but some medication use is invisible to these systems. The illusion of control may result in fewer hospital audits being conducted, and thus even less control and more invisible practices not captured by data systems, which may also have repercussions for algorithms built upon their data. Although extracting data is easier with electronic than paper‐based systems, their analysis requires substantial “data work”, informed by knowledge of the medications examined and of local contexts of supply and use, as evidenced by the study of Walker and her colleagues.7

Identifying discrepancies in medication use records is the first step to improving practices and system design, and to reducing risks for patients and hospitals. The reasons for discrepancies need to be investigated if we are to recommend actions that reduce the levels of medications unaccounted for, improve data quality, and reduce the need for data work.