Does Use of Electronic Alerts for Systemic Inflammatory Response Syndrome (SIRS) to Identify Patients With Sepsis Improve Mortality

Abstract Purpose The objective of this study was to assess whether earlier antibiotic administration in patients with systemic inflammatory response syndrome (SIRS) and evidence of organ dysfunction identified through electronic alerts improves patient mortality. Methods This is a retrospective observational cohort study of adult patients admitted across 5 acute-care hospitals. Mortality, Premier CareScience TM Analytics Expected Mortality Score, and clinical and demographic variables were obtained through the electronic medical record and Premier (Premier Healthcare Solutions, Inc, Charlotte NC) reports. Patients with 2 SIRS criteria and organ dysfunction were identified through an automated alert. Univariate and multivariate logistic regression was performed. Results Of those with SIRS and organ dysfunction, 8146 patients were identified through the electronic Best Practice Alert (BPA). Overall 30-day mortality rate was 8.7%. There was no significant association between time to antibiotic administration from BPA alert and mortality ( P = 0.21) after adjusting for factors that could influence mortality, including age, heart rate, blood pressure, plasma lactate levels, creatinine, bilirubin levels, and the CareScience TM Predicted Mortality Risk Score. Female gender (odds ratio [OR] 1.31, 95% confidence interval [CI] 1.06-1.63) and facility were also independently associated with mortality. Conclusion The use of alerts in the electronic medical record may misclassify patients with SIRS as having sepsis. Time to antibiotic administration in patients meeting SIRS criteria and evidence of end-organ dysfunction through BPA alerts did not affect 30-day mortality rates across a health system. Patient severity of illness, gender, and facility also independently predicted mortality. There were higher rates of antibiotic use and Clostridioides difficile infection in patients with BPA alerts.