Impact of Rapid Influenza Testing at Triage on Management of Febrile Infants and Young Children

Abstract

OBJECTIVE: To determine the impact of an emergency department (ED) triage protocol for rapid influenza testing of febrile infants and children on additional diagnostic testing, ED charges and patient time in the ED. METHODS: A trial of triage-based rapid influenza A and B testing of febrile infants and children 3 to 36 months of age presenting to an urban ED during December 2002 to March 2003 was performed. Children with a temperature of 39 degrees C or higher or history of fever 102 degrees F or higher at home were included. Those with obvious focal infection, potential immunodeficiency, and indwelling medical devices were excluded. The intervention group, tested for influenza at triage (TT) was compared with a nonintervention group consisting of those receiving usual care (SP). A subanalysis comparing influenza-positive children was performed. RESULTS: Of 1007 eligible subjects a total of 719 (71%) patients were in the SP group and 288 in the TT group. There were significant differences in respiratory syncytial virus rapid test (RSV; 18%-7%) and chest radiographs (CXRs; 26%-20%) tests in the TT group. In addition, significant increases in obtaining a complete blood count (relative risk [RR] 12.0; 95% confidence interval [CI] 2.9-49), blood culture (RR, 12.0; 95% CI, 3.0-51.0), RSV testing (RR, 0.9.2; 95% CI, 3.4-25.0), urinalysis (RR, 5.7; 95% CI, 2.0-16.0), CXR (RR, 2.2; 95% CI, 1.04-4.5), time in the ED (195 vs 156 minutes; 95% CI, of the difference 19-60), and medical charges ($666 vs $393; 95% CI, of the difference 153-392) were seen among those testing positive for influenza in the SP group. CONCLUSIONS: A triage protocol for rapid influenza testing for febrile infants and children appears to significantly decrease additional testing, time in the ED, and charges in children testing positive for influenza.