Clinical utility of a host-protein test for suspected infection in the pediatric emergency department: a pragmatic pre-/post-implementation study

Abstract

Background – Identifying infectious etiology in pediatric emergency medicine can be challenging, often leading to antibiotic misuse. MeMed BV (MMBV) is a host-protein test that accurately differentiates bacterial from viral infections, but real-world utility data in children are limited. Methods – We performed a single-center retrospective study of preschool-aged children enrolled in two pragmatic cohort studies, where MMBV was ordered at physician discretion. In the “pre” study, MMBV results were not available (standard of care, SC arm), whereas in the “post” study, MMBV results were available in a timely manner (MMBV arm). The primary endpoint was change in antibiotic prescribing rate among outpatients with viral MMBV results. Secondary endpoints included prescribing rate changes in a lower respiratory tract infection (LRTI) subcohort and changes in hospital length of stay (LOS). Results – The study cohort included 1, 022 children in the SC arm and 474 in the MMBV arm. Antibiotic prescribing in outpatients with viral MMBV results decreased from 20.3% in the SC arm to 7.5% in the MMBV arm (p = 0.007). Prescribing increased in outpatients with bacterial MMBV results (p = 0.111), resulting in a net decline across outpatients in the SC versus MMBV arms (25.0% vs. 16.8%; p = 0.071). In the LRTI subcohort, there was a prescribing decreased (84.0% to 67.9%; p = 0.002) irrespective of patient disposition. Despite more severe clinical presentation, patients in the MMBV arm had a shorter length of stay (LOS) compared with patients in the SC arm (3.1 ± 1.9 vs. 3.6 ± 1.9 days; p < 0.001). Conclusions – Implementation of MMBV in routine pediatric emergency care was associated with optimized antibiotic use and shorter hospital LOS.