Discussion

The findings of this study highlight significant discordance in antimicrobial prescribing patterns despite the availability of guidelines in the ED of a large tertiary care teaching hospital, with an overall discordance of 80%. This is consistent with previous studies indicating high rates of discordance in ED empiric antimicrobial prescribing. ^{Reference White, Clark, Sellick and Mergenhagen2,Reference Thaller, Dempsey, Levine and Shepard9–Reference Feist, Panecaldo, Kale-Pradhan, Delgado, Giuliano and Johnson12} While the predominant use of penicillins and cephalosporins aligns with common treatment guidelines, the high rate of discordance suggests suboptimal prescribing practices. The primary driver of discordance was prolonged therapy duration beyond recommended guidelines. For example, local guidelines recommend a 5-day course for cephalosporins for cellulitis and nitrofurantoin for UTI, yet a median duration of 7 days was prescribed. This may be attributed to Computerized Physician Order Entry (CPOE), which provides a default duration of 7 days for most antimicrobials. This can be optimized by systematizing antimicrobial durations by specific conditions or by shortening the default duration to 5 days and automating alerts for reassessment of therapy.

The study overlapping with the COVID-19 pandemic may have influenced case capture and outcomes due to major shifts in healthcare utilization. During this period, decreased ED visits may have resulted in a greater proportion of higher acuity patients. The CDC reports from 2019 to 2020, there was a 25% decrease in outpatient prescriptions due to fewer ED visits. ¹³ In addition, healthcare systems experienced changes in workflow and staffing pressures that may have influenced clinical approach and prescribing practices due to uncertainty around changing protocols.

This study identified COPD exacerbations and SSTIs as having the highest discordance rates. These findings mirror previous research showing significant discordance in the treatment of SSTIs in ED, where up to 71% of patients received antibiotics discordant with the Infectious Disease Society of America (IDSA) guidelines. ^{Reference Kamath, Sudhakar, Gardner, Hemmige, Safar and Musher11} Another retrospective study reported that only 39% of patients who were admitted to the ED with SSTI received guideline directed therapy. ^{Reference Feist, Panecaldo, Kale-Pradhan, Delgado, Giuliano and Johnson12} A possible reason for high discordance rates in SSTIs is its difficulty to diagnose, leading to variation in treatment approach. ^{Reference Linder, Nicolau and Nailor10}

The ED is a unique practice area which has high acuity cases with high patient turnover, further complicating the implementation of effective antimicrobial stewardship practices. A notable finding in this study was the lack of documented rationale for discordance in 87% of the cases. This may be attributed to the setting of the ED being fast paced with high patient volume. However, this gap in documentation can hinder efforts to address underlying reasons behind non-adherence to guidelines. Moreover, inappropriate antibiotic prescribing is one of the most modifiable risk factors that can contribute to the progression of resistance rates, ^{Reference Bao, Dubrovskaya and Jen5} which is why it is essential for appropriate documentation by prescribers. Previous studies have shown that the role of pharmacists in the ED have significantly improved antimicrobial stewardship practices. A retrospective cohort study which looked at patients who were admitted to the ED found that for patients diagnosed with intraabdominal infections, guideline-based empiric therapy was implemented 62% of the time in the presence of a pharmacist compared to 44% of the time in the absence of a pharmacist. ^{Reference Kulwicki, Brandt, Draper, Weise and Dumkow14} These findings suggest that similar interventions could be beneficial at our hospital to improve guideline concordance. In addition, one of the identified barriers to effective antimicrobial stewardship in the ED is the lack of diagnostic clarity. ^{Reference Paul, Pulia and Pulcini1} Approximately 41% of patients have pending laboratory and microbiological data at the time of discharge from ED, which may lead to diagnostic uncertainty. ^{Reference Bao, Dubrovskaya and Jen5} This uncertainty may lead clinicians to prescribe more broadly, which may lead to overutilization of antibiotics. ^{Reference Koro, Borgert, Abbott and Venugopalan15} This is also observed in this study where 96% of discordant agent cases involved unnecessarily broad coverage for Pseudomonas, MRSA, and atypical pathogens. This finding raises concerns about the potential for increasing antimicrobial resistance and questions whether overly broad empiric antimicrobials are appropriate for treating uncomplicated cases in the ED.

This study found no statistical significance between rates of ED revisit within 30 days of the same infection and antimicrobial concordance. Although revisitation rate was higher in the discordant group, 50% of revisits in both groups were due to worsening infection. These findings suggest that adherence to guideline therapy alone may not determine outcomes and factors such as patient presentation and clinical judgment will influence prescribing decisions.