Background Medication discrepancies on admission is a common occurrence in hospitalised patients which can cause problems during hospitalisation.¹ Studies have shown that medication verification on admission results in fewer discrepancies. Most medication verification studies have been carried out with a general hospital population and do not include psychiatric patients. Medication reconciliation studies in psychiatric patients are scarce. One study showed that medication verification, using a structured medication history in psychiatric patients, resulted in a more accurate overview of medication on admission.²

Purpose To reduce clinically relevant medication discrepancies in psychiatric patients admitted to the medical psychiatric unit (MPU), using a standardised or customised medication verification tool by the hospital pharmacy.

Material and methods Patients admitted to the MPU were randomised in group A, B or C. In group A medication verification was done by the physician, in group B verification was done by a pharmacy technician using a standardised tool and in group C the pharmacy technician used a medication verification tool which was customised for the MPU. Medication discrepancies were assessed by comparing clinically prescribed medication to the outpatient medication records. All medication discrepancies were reviewed by a panel (two clinical pharmacists and a psychiatrist). This panel determined the clinical relevance of the medication discrepancies using the National Coordinating Council for Medication Error Reporting and Prevention index. Categories E to I were considered clinically relevant.

Results At the time of the interim analysis, 45 out of 124 patients were included (33% males; 67% females).

Thirty-five patients had at least one discrepancy, the mean age was 55 years and the mean number of medications was 7.1. In total, 98 discrepancies were found. Of these 27 (28%) were determined clinically relevant.

Conclusion The interim analysis shows relatively more clinically relevant discrepancies in group C. This may indicate that using a customised medication verification tool could lead to fewer clinically relevant medication discrepancies and, therefore, an improvement in clinical care. Statistical testing will be used after inclusion to determine whether these initial findings can be confirmed.

References and/or acknowledgements 1. https://www.ncbi.nlm.nih.gov/pubmed/15738372

2. https://www.ncbi.nlm.nih.gov/pubmed/23959814

No conflict of interest.