Avoiding alert fatigue in pulmonary embolism decision support: a new method to examine 'trigger rates'

Area of Science:

• Clinical Informatics
• Medical Decision Support
• Health Informatics

Background:

• Clinical decision support systems (CDSS) integrated into electronic health records (EHR) often suffer from poor physician adoption rates.
• A primary cause for low adoption is 'trigger fatigue,' where excessive or irrelevant alerts disrupt clinical workflow.

Purpose of the Study:

• To introduce and evaluate a novel usability process, Sensitivity and Specificity Trigger Analysis (SSTA).
• To optimize trigger rates for a pulmonary embolism (PE) decision support tool before its integration into the EHR, aiming to enhance adoption.

Main Methods:

• A retrospective chart review was conducted to assess physician ordering of CT angiography (CTA).
• Common pulmonary embolism symptoms and Wells criteria elements were analyzed as potential CDSS triggers.
• The predictive ability of each trigger for CTA ordering was evaluated using sensitivity and specificity metrics.

Main Results:

• The study identified that combining one or more pertinent symptoms with one or more Wells criteria elements yielded the highest sensitivity for triggering the PE CDSS.
• This combined approach also maintained a high specificity, minimizing false positive alerts.

Conclusions:

• Sensitivity and Specificity Trigger Analysis (SSTA) offers a formal method to optimize CDSS trigger rates prior to EHR integration.
• This methodology can significantly reduce inaccurate triggering, thereby increasing adoption rates for CDSS tools.