Assessing Medication CDS Usability: Pilot Results from 10 Outpatient Clinics.

Area of Science:

• Health Informatics
• Human Factors Engineering
• Patient Safety

Background:

• Medication-related clinical decision support (CDS) alerts are crucial for patient safety.
• Integrating human factors principles into CDS design is essential for effective alert management.
• Existing tools like the Instrument for Evaluating Human Factors Principles in Medication-Related Decision Support Alerts (I-MeDeSA) provide a foundation for assessment.

Purpose of the Study:

• To develop and pilot a human factors assessment for medication-related CDS alerts.
• To evaluate the integration of human factors principles in drug-drug interaction (DDI) alerts within outpatient electronic health record (EHR) systems.
• To identify areas for improvement in CDS alert design and implementation.

Main Methods:

• A human factors assessment was developed based on the validated I-MeDeSA tool, with modifications to questions and answer formats.
• The assessment was piloted in 10 outpatient clinics across the US, each using a different commercial EHR system.
• Clinics triggered high-severity DDI alerts and subsequently completed the human factors assessment.

Main Results:

• The final assessment comprised five sections and twelve questions covering alert timing, visual aspects, severity, content, and actions.
• The mean overall score across clinics was 62%, indicating substantial variability in human factors integration.
• While timing and visual aspects scored well, 40% of clinics lacked safeguards to prevent bypassing high-severity alerts in the 'actions' section.

Conclusions:

• There is significant variability in how outpatient clinics integrate human factors principles into DDI alerts.
• Some EHR systems lack essential medication safeguards, posing a risk to patient safety.
• The developed human factors assessment can be utilized by outpatient clinics for targeted safety improvement initiatives.