Medication administration error: magnitude and associated factors among nurses in Ethiopia

BACKGROUND The significant impact of medication administration errors affect patients in terms of morbidity, mortality, adverse drug events, and increased length of hospital stay. It also increases costs for clinicians and healthcare systems. Due to this, assessing the magnitude and associated factors of medication administration error has a significant contribution for improving the quality of patient care. The aim of this study was to assess the magnitude and associated factors of medication administration errors among nurses at the Felege Hiwot Referral Hospital inpatient department. METHODS A prospective, observation-based, cross-sectional study was conducted from March 24-April 7, 2014 at the Felege Hiwot Referral Hospital inpatient department. A total of 82 nurses were interviewed using a pre-tested structured questionnaire, and observed while administering 360 medications by using a checklist supplemented with a review of medication charts. Data were analyzed by using SPSS version 20 software package and logistic regression was done to identify possible factors associated with medication administration error. RESULT The incidence of medication administration error was 199 (56.4 %). The majority (87.5 %) of the medications have documentation error, followed by technique error 263 (73.1 %) and time error 193 (53.6 %). Variables which were significantly associated with medication administration error include nurses between the ages of 18-25 years [Adjusted Odds Ratio (AOR) = 2.9, 95 % CI (1.65,6.38)], 26-30 years [AOR = 2.3, 95 % CI (1.55, 7.26)] and 31-40 years [AOR = 2.1, 95 % CI (1.07, 4.12)], work experience of less than or equal to 10 years [AOR = 1.7, 95 % CI (1.33, 4.99)], nurse to patient ratio of 7-10 [AOR = 1.6, 95 % CI (1.44, 3.19)] and greater than 10 [AOR = 1.5, 95 % CI (1.38, 3.89)], interruption of the respondent at the time of medication administration [AOR = 1.5, 95 % CI (1.14, 3.21)], night shift of medication administration [AOR = 3.1, 95 % CI (1.38, 9.66)] and age of the patients with less than 18 years [AOR = 2.3, 95 % CI (1.17, 4.62)]. CONCLUSION In general, medication errors at the administration phase were highly prevalent in Felege Hiwot Referral Hospital. Documentation error is the most dominant type of error observed during the study. Increasing nurses' staffing levels, minimizing distraction and interruptions during medication administration by using no interruptions zones and "No-Talk" signage are recommended to overcome medication administration errors. Retaining experienced nurses from leaving to train and supervise inexperienced nurses with the focus on medication safety, in addition providing convenient sleep hours for nurses would be helpful in ensuring that medication errors don't occur as frequently as observed in this study.