Aims: To examine the prevalence of antibiotic prescription errors in three medical departments.
Background: Prescription errors are common and associated with significant adverse drug events (ADEs), morbidity and mortality, and health care expenditures.
Methods: A prospective observational cohort study was conducted in three medical departments, including consecutive patients with suspected or proven infections, and/or antibiotic prescriptions. The primary outcome was the proportion of prescription errors, defined as: contraindications, inadequate dose regimen, and unnecessary antibiotic treatment. Secondary outcomes included incidence of ADEs, proportion of potential drug-drug interactions (DDIs) with clinical relevance, and prevalence of inadequate monitoring for ADEs and therapeutic drug monitoring (TDM).
Results: We identified 327 patient-episodes in 295 patients. The most common infectious diagnoses were urinary tract infection and pneumonia. Among 633 prescriptions, 113 (18%) contained errors in 87 (27%) patient-episodes. The most common types of error were inappropriate dose adjustment for renal function and unnecessary treatment. There were 6 prescriptions with contraindications (0.9%). Laboratory monitoring was required in 259 patient-episodes but inadequate in 40 (15%). TDM was required in 40 patient-episodes, but was not performed in 25 (63%). There were 69 ADEs in 61 patient-episodes (19%). Compared to patients without ADEs, patients who developed ADEs had more prescription errors (p=0.055), more potential DDIs (p=0.012), and received more often antibiotics that needed monitoring and TDM.
Conclusions: Antibiotic prescription errors in medical departments are common and may be associated with significant ADEs. Our findings may help in prioritizing the customization of prescription computer decision support systems to improve antibiotic prescription.