Effect of Point-of-care Influenza Testing on Management of Febrile Children

Abstract

Objectives: To determine the effect of point-of-care testing (POCT) for influenza on the physician management of febrile children who are at risk for serious bacterial illness (SBI) on the basis of age and temperature and who are presenting to a pediatric emergency department (ED) during an influenza outbreak. Methods: Patients 2-3 months of age with temperature of ≥38°C and patients 3-24 months of age with temperature of ≥39°C who were presenting to a pediatric ED during an influenza outbreak were enrolled into a prospective, quasi-randomized, controlled trial. Influenza testing was performed on enrolled patients by either the POCT or the standard-testing (ST) methods. The two groups were compared in terms of laboratory testing, chest radiography, antibiotic use, visit-associated costs, pediatric ED lengths of stay, inpatient admission, and return visits to the pediatric ED. Similar analyses also were performed on the resulting subgroups of patients on the basis of method of testing (POCT or ST) and test result (positive or negative). Results: Of 767 eligible patients, 700 (91%) completed the study. No significant differences were demonstrated between the POCT and ST groups with respect to laboratory tests ordered, chest radiographs obtained, antibiotic administration, inpatient admission, return visits to the pediatric ED, lengths of stay, or visit-associated costs. In the subgroup analysis, the adjusted odds ratios (ORs) for blood culture in influenza test-positive to -negative patients were 0.59 and 0.71 in the POCT and ST groups, respectively (p = 0.088). The adjusted ORs for urine culture in influenza test-positive to -negative patients were 0.46 and 0.67 in the POCT and ST groups, respectively (p = 0.005). Conclusions: When using a strategy of performing influenza testing on all patients at risk for SBI who presented to a pediatric ED during an influenza outbreak, the method of testing (POCT or ST) did not appear to significantly alter physician management, cost, or length of stay in the pediatric ED. However, if the interaction of the method of testing and the test result (positive or negative) were considered, a positive POCT for influenza was associated with a significant reduction in orders for urinalyses and urine cultures. © 2006 Society for Academic Emergency Medicine.