Objective: To assess the type, frequency and potential clinical significance of medication-administration errors.
Design: Prospective study using the observation technique as described by the American Society of HealthSystem Pharmacists but eliminating the disguised aspect.
Setting: Medical intensive care unit (ICU) in a university hospital.
Patients and participants: 2009 medication administration interventions by nurses.
Interventions: Pharmacist-performed observation of preparation and administration of medication by nurses, comparison with the original medical order and comparison with the data available in the literature.
Measurements and results: 132 (6.6% of 2009 observed events) errors were detected. Their distribution is as follows: 41 dose errors, 29 wrong rate, 24 wrong preparation technique, 19 physicochemical incompatibility, 10 wrong administration technique and 9 wrong time errors. No fatal errors were observed, but 26 of 132 errors were potentially life-threatening and 55 potentially significant.
Conclusion: According to this first observation-based study of medication administration errors in a European ICU, these errors were due to deficiencies in the overall organisation of the hospital medication track, in patient follow-up and in staff training.