Prioritizing emergency department antibiotic stewardship interventions for skin and soft tissue infections using judgment analysis

Infect Control Hosp Epidemiol. 2025 Jan 20:1-9. doi: 10.1017/ice.2024.211. Online ahead of print.

Abstract

Objective: Skin and soft tissue infections (SSTIs) account for over 2.8 million annual emergency department (ED) visits and often result in suboptimal antibiotic therapy. The objective of this study was to evaluate a set of interventions in minimizing inappropriate prescription of antibiotics for presumed SSTIs in the ED.

Design: Case vignette survey.

Participants: A national sample of emergency medicine (EM) physicians.

Methods: Each vignette described a clinical scenario of a presumed SSTI (cellulitis or abscess) and included a unique combination of zero to five interventions (outpatient follow-up, inappropriate antibiotic request flag, thermal imaging for cellulitis or rapid wound MRSA PCR for abscess, patient education/shared decision-making, and clinical decision support). Out of 64 possible vignettes, we asked participants to respond to eight vignettes. Following each vignette, we asked participants if they would prescribe an antibiotic in their everyday practice (yes/no). We built adjusted hierarchical logistic regression models to estimate the probability of prescribing an antibiotic for each intervention and vignette.

Results: Surveys were completed by 113 EM physicians. The thermal imaging, rapid wound MRSA PCR, and patient education/shared decision-making interventions showed the largest decrease (15-20%) in antibiotic prescribing probability. Vignettes with a combination of both a diagnostic intervention (thermal imaging or rapid wound MRSA PCR) and a patient education/shared decision-making intervention had the lowest prescribing probabilities.

Conclusion: We recommend future research focuses on the development and integration of novel diagnostic tools to identify true infection and incorporate shared decision-making to improve diagnosis and management of SSTIs.