The Acceptance of Interruptive Medication Alerts in an Electronic Decision Support System Differs between Different...

Area of Science:

• Clinical Informatics
• Pharmacovigilance
• Health Informatics

Background:

• Clinical decision support systems (CDSS) aim to reduce medication errors, such as drug-drug interactions (DDIs).
• High alert override rates in CDSS can be mitigated by alert tailoring and workflow-interruptive displays.
• The impact of alert intrusiveness on prescriber acceptance remains inconclusive.

Purpose of the Study:

• To assess the frequency of prescription changes resulting from interruptive alerts.
• To evaluate alert acceptance rates for various contraindicated prescription constellations with high risks for adverse drug events (ADEs).

Main Methods:

• Retrospective analysis of interruptive alerts from the AiDKlinik CDSS at Heidelberg University Hospital (March 2016 - August 2020).
• Evaluated alerts included contraindicated simvastatin DDIs, potentially inappropriate medications (PIM) for patients >65 years, and contraindicated drugs in hyperkalemia.
• Prescription changes were assessed based on alert acceptance or override with free-text reasons.

Main Results:

• Interruptive alerts occurred in 935 of 500,274 prescribing sessions.
• Overall, 57.5% of evaluable interruptive alerts were accepted, while 42.5% were overridden.
• Acceptance rates varied significantly by alert type: 85.7% for DDI alerts, 65.3% for hyperkalemia alerts, and 25.1% for PIM alerts.

Conclusions:

• 57.5% of high-risk interruptive medication alerts were accepted in this setting.
• Alert acceptance rates significantly differed based on alert type, with contraindicated simvastatin DDI alerts showing the highest acceptance.
• Findings underscore the need for tailored alert strategies to enhance medication safety.