Abstract
OBJECTIVES: To describe variation in empirical antibiotic selection in infants <60 days old who are hospitalized with skin and soft-tissue infections (SSTIs) and to determine associations with outcomes, including length of stay (LOS), 30-day returns (emergency department revisit or readmission), and standardized cost. METHODS: Using the Pediatric Health Information System, we conducted a retrospective study of infants hospitalized with SSTI from 2009 to 2014. We analyzed empirical antibiotic selection in the first 2 days of hospitalization and categorized antibiotics as those typically administered for (1) staphylococcal infection, (2) neonatal sepsis, or (3) combination therapy (staphylococcal infection and neonatal sepsis). We examined the association of antibiotic selection and outcomes using generalized linear mixed-effects models. RESULTS: A total of 1319 infants across 36 hospitals were included; the median age was 30 days (interquartile range [IQR]: 17-42 days). We observed substantial variation in empirical antibiotic choice, with 134 unique combinations observed before categorization. The most frequently used antibiotics included staphylococcal therapy (50.0% [IQR: 39.2-58.1]) and combination therapy (45.4% [IQR: 36.0-56.0]). Returns occurred in 9.2% of infants. Compared with administration of staphylococcal antibiotics, use of combination therapy was associated with increased LOS (adjusted rate ratio: 1.35; 95% confidence interval: 1.17-1.53) and cost (adjusted rate ratio: 1.39; 95% confidence interval: 1.21-1.58), but not with 30-day returns. CONCLUSIONS: Infants who are hospitalized with SSTI experience wide variation in empirical antibiotic selection. Combination therapy was associated with increased LOS and cost, with no difference in returns. Our findings reveal the need to identify treatment strategies that can be used to optimize resource use for infants with SSTI.