Nurses' perceptions of causes of medication errors experienced

This research was planned as a descriptive study with the purpose of determining nurses' perceptions of causes of medication errors experienced by nurses. The research was conducted with a total of 590 nurses working in drug administration units of A, B, C and D hospitals in Antalya between the dates of April 23 and August 1, 2013. The survey, developed by Wakefield et al. to measure nurses' perceptions of medication errors, was used as the data collection tool after the researchers performed the language validity of, and added seven questions, to the survey based on the literature knowledge. The Statistical Package for Social Sciences (SPSS 16) was used for analysing data obtained. Factor analysis for structural validity and Cronbach's alpha coefficient for reliability were calculated. The Kendall Coefficient of Concordance to assess the feasibility and intelligibility of the survey items, and the skewness value to assess the survey's conformity to the normal distribution were measured. The survey sub-scale score-total score correlation was determined. Frequency, percentage (%) distribution, mean and standard deviation values of data were measured in the descriptive analyses. The correlations between data were analysed using 'Pearson Chi-Square test (X²)' in the hypothesis analyses. Normality test, Anova and T test was performed for data of continuous type. Inadequate number of nurses (82%), excessive workloads of nurses (79.8%), illegibility of physicians' drug orders (71%), were determined as causes of medication errors indicated at the highest rate by nurses. It was found out that 26.1% of the nurses made medication errors, 55.8% of them witnessed to medication errors. It was discovered that 68.8% of the nurses were not aware of medication errors they made. A statistically significant relationship was found between departments of the nurses and making medication errors and witnessing to medication errors. It was found that the rates of reporting medication errors by nurses with an undergraduate and above education were high. As a result, it has been concluded that determination of causes of medication errors is important for preventing recurrence of medication errors, establishing standards for drug safety. The survey, used as the data collection tool in the research, was found valid and reliable together with items added in Turkish. Cronbach alpha value of the research was 0.92 and the factor analysis results were found consistent with the original survey. Keywords: Medication errors, Perception, Nurse, validity and reliability