Racial Differences in Sepsis Recognition in the Emergency Department

Abstract

OBJECTIVES: We assessed racial differences in sepsis recognition in a pediatric emergency department (ED) with an established electronic sepsis alert system. METHODS: Quality-improvement data from June 1, 2016 to May 31, 2017 was used in this retrospective cohort study. All ED visits were included for non-Hispanic black (NHB) and non-Hispanic white (NHW) patients. The sepsis pathway was activated through the alert, 2 stages and a huddle, or outside of the alert using clinician judgment alone. We evaluated racial differences in the frequency of alerts and sepsis pathway activation within and outside of the alert. Multivariable regression adjusted for high-risk condition, sex, age, and insurance. RESULTS: There were 97 338 ED visits: 56 863 (58.4%) and 23 008 (23.6%) from NHBs and NHWs, respectively. NHWs were more likely than NHBs to have a positive second alert (adjusted odds ratio [aOR] 2.4; 95% confidence interval [CI] 2.1–2.8). NHWs were more likely than NHBs to have the sepsis pathway activated (aOR 1.4; 95% CI 1.02–2.1). Of those treated within the alert, there was no difference in pathway activation (aOR 0.93; 95% CI 0.62–1.4). Of those recognized by clinicians when the alert did not fire, NHWs were more likely than NHBs to be treated (aOR 3.4; 95% CI 1.8–6.4). CONCLUSIONS: NHWs were more likely than NHBs to be treated for sepsis, although this difference was specifically identified in the subset of patients treated for sepsis outside of the alert. This suggests that an electronic alert reduces racial differences compared with clinician judgment alone.