# Physicians’ Perspectives on Prescription Alerts: A Journey Towards Reducing Fatigue

## TL;DR

This study explores how too many medical alerts in electronic records can overwhelm doctors, leading to alert fatigue, and suggests reducing non-critical alerts to improve safety and efficiency.

## Contribution

The study introduces a mixed-methods approach combining surveys and data analysis to reduce non-critical alerts and assess their impact on physician behavior.

## Key findings

- 76% of physicians found alerts helpful, but 81% felt overwhelmed by their volume.
- Reducing 13,000 non-critical alerts per month did not significantly change prescription behavior.
- Alerts related to administrative or cost issues were most problematic for clinicians.

## Abstract

Background

Alerts in electronic medical records (EMR) and computerized decision support systems play crucial roles in enhancing patient safety by notifying healthcare providers of potential risks. However, an excessive number of alerts can lead to alert fatigue, in which clinicians become desensitized to notifications, potentially compromising patient safety. This study examined the experiences of physicians with alerts, focusing on their perceptions of and challenges posed by alert fatigue.

Methods

A mixed-methods approach was employed, combining an anonymous questionnaire survey with an interrupted time series (ITS) analysis. This study targeted physicians to gather insights into their experiences of using alerts. Data were collected through a Google Forms survey, and prescription data from the EMR were analyzed to assess the impact of the strategic reduction in non-critical alerts. Based on the survey feedback and research team discussions, the intervention targeted specific alerts contributing to alert fatigue, removing two types: “Powerful Drugs” and “Multiple Prescriptions.”

Results

While 76% of physicians found alerts helpful, 81% reported being overwhelmed by their volume, leading to alert fatigue. Notably, 55% of physicians admitted to dismissing alerts without reading them. The ITS analysis showed that a reduction of approximately 13,000 non-critical alerts per month did not significantly alter prescription behavior. The most problematic alerts were those related to administrative or cost issues, which contributed to the overall fatigue experienced by clinicians.

Conclusion

This study highlights the significant impact of alert fatigue on healthcare providers and underscores the need for a more streamlined alert system. Healthcare institutions can better manage alert fatigue and improve clinician efficiency and patient safety by focusing on reducing non-critical alerts and incorporating feedback from frontline clinicians.