The Burden of a Highly Targeted Alert

Abstract

Background

Interruptive alerts in clinical decision support (CDS) systems are intended to guide clinicians in making informed decisions and adhering to best practices. However, these alerts can often become a source of frustration, contributing to alert fatigue and clinician burnout. Traditionally, an alert's burden is often assessed by evaluating the total number of times it is seen by end-users, which can overlook the true impact of highly interruptive workflows. This study demonstrates how an alert burden metric was employed to pinpoint an ineffective and burdensome alert, ultimately leading to its deactivation.

Objectives

This study aimed to evaluate the effectiveness of a burden metric in identifying high-impact, low-value alerts and prioritizing improvement efforts for a CDS governance team.

Methods

A clinical informatics team employed Phrase Health's “Phrase Burden Index” (PBI) to assess alert burden and identify areas requiring intervention within the alert library.

Results

The team used the PBI to identify a breast cancer survivorship alert that fired 3,550 times in 2023, with the desired alert action chosen in only 0.00056% of alert firings. An investigation identified that this alert targeted a single clinician over the span of several years, and the CDS governance team promptly decommissioned the alert.

Conclusion

This case highlights the value of continuous CDS monitoring, effective governance, and advanced analytics to identify and mitigate alert fatigue. Insights from this failure provide guidance for enhancing future CDS design, evaluation, and clinician engagement.

Keywords

clinical decision support - burden - alert fatigue - electronic health records - governance

Protection of Human and Animal Subjects

This case study did not involve human subject participants.

Publikationsverlauf

Eingereicht: 27. Dezember 2024

Angenommen: 02. April 2025

Accepted Manuscript online:
03. April 2025

Artikel online veröffentlicht:
30. Juli 2025

• References

• 1 Aaron S, McEvoy DS, Ray S, Hickman TTT, Wright A. Cranky comments: detecting clinical decision support malfunctions through free-text override reasons. J Am Med Inform Assoc 2019; 26 (01) 37-43
• 2 AHRQ | PSNet (Patient Safety Network). Alert Fatigue. September 7, 2019. Accessed December 9, 2024 at: https://psnet.ahrq.gov/primer/alert-fatigue
• 3 McGreevey III JD, Mallozzi CP, Perkins RM, Shelov E, Schreiber R. Reducing alert burden in electronic health records: State of the art recommendations from four health systems. Appl Clin Inform 2020; 11 (01) 1-12
• 4 Orenstein EW, Muthu N, Weitkamp AO. et al. Towards a maturity model for clinical decision support operations. Appl Clin Inform 2019; 10 (05) 810-819
• 5 Sittig DF, Sengstack P, Singh H. Guidelines for us hospitals and clinicians on assessment of electronic health record safety using SAFER Guides. JAMA 2022; 327 (08) 719-720
• 6 HealthIT.gov. SAFER Guides. Accessed December 9, 2024 at: https://www.healthit.gov/topic/safety/safer-guides
• 7 Bates DW, Kuperman GJ, Wang S. et al. Ten commandments for effective clinical decision support: making the practice of evidence-based medicine a reality. J Am Med Inform Assoc 2003; 10 (06) 523-530