193 - Quality Improvement Initiative for Non-Purulent Cellulitis Management in Urgent Care Setting: Provider-level Performance Feedback

Background: Methicillin-resistant Staphylococcus aureus (MRSA) has emerged as a common cause of skin and soft-tissue infections (SSTIs). This has resulted in an 88% increase in MRSA-directed antibiotic use in emergency departments. However, the majority of cellulitis presents as non-purulent due to Group A streptococci. An unintended consequence is that many with non-purulent cellulitis receive sub-optimal antibiotics and unnecessary diagnostics. Clinical guidelines at our institution recommend beta-lactam antibiotics and discourage empiric MRSA coverage for non-purulent cellulitis. The aim of this study is to use an audit-feedback intervention to optimize urgent care providers management of mild/moderate non-purulent cellulitis.

Methods: We identified all consecutive patients presenting to our urgent care with a diagnosis of lower extremity non-purulent cellulitis using ICD coding. We conducted a prospective pre and post-intervention study from 10/2018-3/2019 and 11/2019-4/2020 respectively. Intervention included review of practice guidelines with providers and feedback from pre-intervention phase. To assess individual practitioners’ prescribing habits, a comprehensive scoring system focused on empiric antibiotic selection, antibiotic duration, and diagnostics was combined with patient demographics. Scores classified non-purulent SSTI treatment as follows: 0-2 = good, 2.5-5 = fair, and >5.5 = poor (Figure 1). Figure 1: Provider Cellulitis Management Scoring Sheet


Results: There was statistically significant provider score improvement in the post-intervention phase with greater percentage of good cases (40% to 69%) and no poor cases (Figure 2). For IV antibiotics for mild and moderate cellulitis, there was decreased use of overly broad antibiotics (Figure 3). Antibiotic duration of greater than 7 days decreased from 68% to 52%. Combination antibiotic therapy decreased from 12% to 4%. There was also a statistically significant decrease in use of wound cultures but no change in ultrasound use. Figure 2: Overall Provider Scores
Figure 3: IV Antibiotic Use


Conclusion: Audit-feedback intervention was effective for improving antibiotic usage and decreasing treatment duration. Providing physicians with data on their practice patterns relative to those of their peers and clinical practice guidelines can improve management of non-purulent cellulitis.