Enhanced Free-Text Search for Aggregated Medication Error Report Analysis and Risk Detection

Abstract

Objectives Detecting medication errors (MEs) and learning from them are the key elements of medication safety management in health care. While the aggregation of the data and learning across the ME reports could help detect and manage organizational risks, the inconsistent and partly missing structural data complicate the analysis. The objective of this study was to examine whether an analysis of free-text data of aggregated ME reports could contribute to the detection of organizational risks. Methods A retrospective, cross-sectional analysis of ME reports from a patient safety incident reporting system in a tertiary hospital 2017–2021. Clustering of characteristics and variables of ME reports with an enhanced free-text search of the 10 most frequent active substances (TOP10) related to ME reports using Microsoft Excel. Validity analysis of the four most frequent active substances of the search results (TOP4). Evaluation of the possible impact of the enhanced free-text search method on ME report analysis and risk detection. Results The enhanced free-text search increased significantly the number of relevant ME reports of TOP10 active substances from 698 reports to 1578 reports. The validity of the enhanced free-text search results in TOP4 active substances was more than 74%. The enhanced free-text search revealed also new ME findings. Conclusions Enhanced free-text search can contribute to the aggregate analysis of clustered ME reports and to the improvement of ME risk detection. The enhanced free-text search method enables more comprehensive analysis of the free-text data with commonly available software and provides new insights into medication safety improvement.