How useful are drug-drug interaction alerts? An interview study with Australian hospital prescribers
Background: Drug-drug interaction (DDI) alerts have been integrated into Australian hospital electronic medication management systems as a strategy to prevent medication errors and reduce patient harm. Although these alerts are believed to facilitate and improve prescriber behaviour, research suggests that up to 90% of DDI alerts are overridden by prescribers in practice
Aims: This study aims to identify the barriers and facilitators to the optimal use of DDI alerts and explore the experiences and perceptions of Australian hospital prescribers towards these alerts.
Methods: Semi-structured interviews were conducted at two hospital sites and participants were recruited using a snowball sampling approach. Prescribers were asked if DDI alerts were useful, how DDI alerts impacted prescriber workflow or decision-making and suggestions to improve DDI alerts. Interviews are audio-recorded and transcribed, and interview data analysed using qualitative content analysis and coded independently by three researchers to reach a consensus on the main themes identified.
Results: Although alerts were perceived to be useful in theory, prescribers identified several areas where DDI alerts could be improved, particularly in alert design. DDI alerts were reported to have a mixed impact on patient safety and prescriber workflow, with users identifying particular contexts where DDI alerts might be useful (e.g. complex polypharmacy patients and high risk interactions) and other contexts where alerts were not (e.g. prescribers receiving alerts for drug combinations commonly used in practice).
Conclusions: The implementation of DDI alerts in hospital electronic medication management systems is important to help prescribers prevent medication errors. However, these alerts are also associated with many risks, such as alert fatigue. Significant work is needed to tailor DDI alerts to the needs and preferences of prescribers. This way, we can optimise alert usability and effectiveness, as well as improve medication safety and patient outcomes overall.