The validation of an existing method of scoring the severity of medication administration errors for use in Germany

Abstract

Objective: The aim of this study was to assess the validity and reliability of an existing method of scoring the severity of medication administration errors, developed in the United Kingdom (UK), for use in Germany.

Method: 10 doctors, 10 nurses, and 10 pharmacists from German hospitals were asked to score the potential clinical significance of 49 cases of medication administration errors on a visual analogue scale.

Main outcome measure: Generalisability theory was used to determine the minimum number of judges required to obtain a reliable mean score, Validity was assessed by comparing the mean scores given by the judges to the known outcome of the errors for a subset of the cases. German results were compared to original UK data.

Results: The scores of 27 judges could be used (nine from each profession). At least three health professionals, one from each profession, were required to achieve a generalisability coefficient of 0.86, indicating acceptable reliability. The mean scores were found to be valid indicators of the potential severity of the errors German scores were significantly below UK scores for the same cases.

Conclusion: The mean score calculated from scores given by one doctor, one nurse and one pharmacist from the population of German health professionals was a valid and reliable measure of the potential clinical significance of medication administration errors. That German health professionals see cases as less dangerous than their UK counterparts is worthy of further investigation.