Comparison of manual and automated sepsis screening tools in a pediatric emergency department

Abstract

OBJECTIVES: To compare the performance and test characteristics of an automated sepsis screening tool with that of a manual sepsis screen in patients presenting to a pediatric emergency department (ED). METHODS: We conducted a retrospective cohort study of encounters in a pediatric ED over a 2-year period. The automated sepsis screening algorithm replaced the manual sepsis screen 1 year into the study. A positive case was defined as development of severe sepsis or septic shock within 24 hours of disposition from the ED. We calculated the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and positive and negative likelihood ratios with 95% confidence intervals (CIs) for each. RESULTS: There were 122 221 ED encounters during the study period and 273 cases of severe sepsis. During year 1 of the study, the manual screen was performed in 8910 of 61 026 (14.6%) encounters, resulting in the following test characteristics: sensitivity of 64.6% (95% CI 54.2%-74.1%), specificity of 91.1% (95% CI 90.5%-91.7%), PPV of 7.3% (95% CI 6.3%-8.5%), and NPV of 99.6% (95% CI 99.5%-99.7%). During year 2 of the study, the automated screen was performed in 100% of 61 195 encounters, resulting in the following test characteristics: sensitivity of 84.6% (95% CI 77.4%-90.2%), specificity of 95.1% (95% CI 94.9%-95.2%), PPV of 3.7% (95% CI 3.4%-4%), and NPV of 99.9% (95% CI 99.9%-100%). CONCLUSIONS: An automated sepsis screening algorithm had higher sensitivity and specificity than a widely used manual sepsis screen and was performed on 100% of patients in the ED, ensuring continuous sepsis surveillance throughout the ED stay.