Observational study on medication administration errors at a University Hospital in Brazil: incidence, nature and associated factors.

Background: Medication administration errors are frequent and cause significant harm globally. However, only a few data are available on their prevalence, nature, and severity in developing countries, particularly in Brazil. This study attempts to determine the incidence, nature, and factors associated with medication administration errors observed in a university hospital.

Methods: This was a prospective observational study, conducted in a clinical and surgical unit of a University Hospital in Brazil. Two previously trained professionals directly observed medication preparation and administration for 15 days, 24 h a day, in February 2020. The type of error, the category of the medication involved, according to the anatomical therapeutic chemical classification system, and associated risk factors were analyzed. Multivariate logistic regression was adopted to identify factors associated with errors.

Results: The administration of 561 drug doses was observed. The mean total medication administration error rate was 36.2% (95% confidence interval 32.3-40.2). The main factors associated with time errors were interruptions. Regarding technique errors, the primary factors observed were the route of administration, interruptions, and workload.

Conclusions: Here, we identified a high total medication administration error rate, the most frequent being technique, wrong time, dose, and omission errors. The factors associated with errors were interruptions, route of administration and workload, which agrees well with the results of other national and international studies.