327. Assessment of Bacterial Co-infection Rates and Antibiotic Exposure in COVID-19 Patients

BACKGROUND: COVID-19 pandemic data suggest risk for bacterial co-infection upon hospital presentation remain extremely low. Despite low co-infection rates, antibiotics are prescribed for most patients. Current data are limited regarding institutional-specific change in antibiotic use over the course of the pandemic. Given the low rates of co-infections, Saint Luke’s Health System’s COVID-19 Treatment Taskforce developed a COVID-19 evaluation and treatment order set which included procalcitonin (PCT) . As co-infection literature emerged, active education was provided, and order sets were modified to provide passive education regarding co-infection rates. We aimed to assess antibiotic practice changes as data and strategies to influence use evolved during the pandemic. METHODS: This was a multi-center, single health-system retrospective cohort study. Ten community hospitals and 1 academic medical center were included in analysis. Inclusion criteria were age ≥18 years, admitted during April or September 2020 and had a positive COVID-19 result on admission. Patients were excluded if they were readmitted for COVID-19 related issues. Both primary and secondary outcomes were analyzed from the first 7 days after admission. The primary outcome was rate of respiratory bacterial co-infections. This was determined through sputum and blood cultures, urinary antigens including Streptococcus pneumoniae and Legionella, and PCT. Secondary outcomes included rate of antibiotic use, antibiotic days of therapy (DOT), length of therapy, and antibiotic use trends. Baseline Characteristics [Image: see text] RESULTS: A total of 294 patients were included with 69 patients in April 2020 and 225 in September 2020. Primary and secondary results are shown in Table 2. Rate of culture-confirmed bacterial co-infection when examining April 2020 was 4.38% and 4.44 % in September 2020. Antibiotic uses, antibiotic DOT, and length of therapy were all significantly lower in September 2020 compared to April 2020. [Image: see text] CONCLUSION: Our results show bacterial co-infections were extremely low in our health system. Despite positive trends in antibiotic use, prescribing remained high. More targeted interventions to decrease antibiotic exposure in COVID-19 patients are needed. DISCLOSURES: All Authors: No reported disclosures