Clinical Impact of a Sepsis Alert System Plus Electronic Sepsis Navigator Using the Epic Sepsis Prediction Model in the Emergency Department

Adam R Schertz; Sydney A Smith; Kristin M Lenoir; Karl W Thomas

doi:10.1016/j.jemermed.2023.02.025

Clinical Impact of a Sepsis Alert System Plus Electronic Sepsis Navigator Using the Epic Sepsis Prediction Model in the Emergency Department

J Emerg Med. 2023 May;64(5):584-595. doi: 10.1016/j.jemermed.2023.02.025. Epub 2023 Mar 8.

Authors

Adam R Schertz¹, Sydney A Smith², Kristin M Lenoir², Karl W Thomas¹

Affiliations

¹ Department of Internal Medicine, Section on Pulmonology, Critical Care, Allergy, and Immunologic Diseases, Wake Forest University School of Medicine, Winston-Salem, North Carolina.
² Department of Biostatistics and Data Science, Division of Public Health Science, Wake Forest University School of Medicine, Winston-Salem, North Carolina.

PMID: 37045722
DOI: 10.1016/j.jemermed.2023.02.025

Abstract

Background: The Epic Sepsis Prediction Model (SPM) is a proprietary sepsis prediction algorithm that calculates a score correlating with the likelihood of an International Classification of Diseases, Ninth Revision code for sepsis.

Objective: This study aimed to assess the clinical impact of an electronic sepsis alert and navigator using the Epic SPM on time to initial antimicrobial delivery.

Methods: We performed a retrospective review of a nonrandomized intervention of an electronic sepsis alert system and navigator using the Epic SPM. Data from the SPM site (site A) was compared with contemporaneous data from hospitals within the same health care system (sites B-D) and historical data from site A. Nonintervention sites used a systemic inflammatory response syndrome (SIRS)-based alert without a sepsis navigator.

Results: A total of 5368 admissions met inclusion criteria. Time to initial antimicrobial delivery from emergency department arrival was 3.33 h (interquartile range [IQR] 2.10-5.37 h) at site A, 3.22 h (IQR 1.97-5.60; p = 0.437, reference site A) at sites B-D, and 6.20 h (IQR 3.49-11.61 h; p < 0.001, reference site A) at site A historical. After adjustment using matching weights, there was no difference in time from threshold SPM score to initial antimicrobial between contemporaneous sites. Adjusted time to initial antimicrobial improved by 2.87 h (p < 0.001) at site A compared with site A historical.

Conclusions: Implementation of an electronic sepsis alert system plus navigator using the Epic SPM showed no difference in time to initial antimicrobial delivery between the contemporaneous SPM alert plus sepsis navigator site and the SIRS-based electronic alert sites within the same health care system.

Keywords: Epic Sepsis Prediction Model; Predicting Sepsis Score; sepsis alert; sepsis bundle; sepsis navigator.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Emergency Service, Hospital
Humans
Retrospective Studies
Sepsis* / diagnosis
Sepsis* / drug therapy
Software
Systemic Inflammatory Response Syndrome / diagnosis
Systemic Inflammatory Response Syndrome / drug therapy