Browse the corpus

Vignettes Identify Variation in Antibiotic Use for Suspected Early Onset Sepsis. BACKGROUND AND OBJECTIVES: There is widespread unwarranted antibiotic use and large individual provider variation in antibiotic use in NICUs. Vignette-based research methodology offers a unique method of studying variation in individual provider decisions. The objective with this study was to use a vignette-based survey to identify specific areas of provider antibiotic use variation in newborns being evaluated for early onset sepsis. METHODS: This study was undertaken as part of a statewide multicenter neonatal antibiotic stewardship quality improvement project led by a perinatal quality improvement collaborative. A web-based vignette survey was administered to identify variation in decisions to start and discontinue antibiotics in cases of early onset sepsis. RESULTS: The largest variation was noted in 3 of the 6 vignette cases. These cases highlighted variation in (1) decisions to start antibiotics in a case describing a well-appearing newborn with risk factors and an elevated C-reactive protein, (2) decisions to start antibiotics in the case of a newborn with risk factors plus mild respiratory signs at birth, and (3) decisions to stop antibiotics in the case of the newborn with a history of sepsis risk factors and mild clinical respiratory signs that resolved after 72 hours. CONCLUSIONS: Clinical vignette assessment identified specific areas of variation in individual provider antibiotic use decisions in cases of suspected early onset sepsis. Vignettes are a valuable method of describing individual provider variation and highlighting antibiotic stewardship improvement opportunities in NICUs.