Erros de medicação notificados em um hospital público de grande porte

Abstract

The general objective of this study was to analyze medication errors reported in a largepublic hospital, in the city of Salvador-Bahia. This is research with a cross-sectional andretrospective quantitative approach to notifications of medication errors, in the time window from 2019 to 2022. The study variables cover types of reported errors that occur in the process stage medicated; time of occurrence of reported errors; professional category that notifies; notified sector, where the medication error occurred; sector that reports errors arising from drug therapy and the degree of damage caused to the patient.Data analysis was carried out in STATA version 17, where all collected variables weresubjected to a descriptive analysis. For categorical variables, absolute (n) and relative frequencies (%) were calculated. To check the association between these variables, Pearson's Chi-square test or Fisher's exact test was applied. The level of statistical significance adopted was 5%. Our results demonstrated that there were quantitative reports of medication errors with differentiated values and, for these reasons, aggregatedbiennially (2019-2020 and 2021-2022). Regarding the period of occurrence of reportederrors, particularly in the 2021-2022 biennium, there is the daytime period, and the 63.7% reduction in error notifications during the night period of the second biennium trained is also highlighted. Regarding the type of error in the medication process stage,there were a greater number of notifications in the administration stage in all specific periods. In relation to the professional category that most reported medication errors, nurses and nursing technicians stood out in both biennia. Regarding the sectors of the hospital that received notifications of medication errors and the sectors that made notifications involving medication errors, similar results were observed in the two biennia investigated, with emphasis on the wards, intensive care units and the emergency, respectively in both notification situations. Regarding the degree of harm tothe patient related to erroneous drug therapy, the mild degree was more evident in bothbiennia investigated, with 97.1% and 93.42%, respectively. These results highlight the statistical significance in all variables stratified in relation to the variable of interest. This study contributed to characterizing medication errors reported in a large public hospital. The analysis of notifications related to medication errors reverberates in organizational learning, drives the creation of strategic actions, enabling causes to be identified and avoided, and suggests review and improvement of care processes