Antibiotic prescribing trends in the US during the first 11 months of the COVID-19 pandemic.

Background: The study aims to compare antibiotic prescribing trends for U.S. COVID-19 patients, categorized by disease severity, and non-COVID-19 population with similar symptoms during 2019-2020 pandemic.

Methods: A retrospective observational cohort design using Symphony Health (January-November 2020). Sample population included about 13.3 million patients with at least one prescription claim ±6 months from date of diagnosis of COVID-19 or COVID-19 like symptom. Cohorts were categorized based on diagnosis codes; COVID-19 positive cohorts 1 to 3 with severe, mild, and no symptoms, respectively and non-COVID-19 cohorts 4 and 5 with severe and mild symptoms, respectively. Descriptive statistics were calculated for demographic characteristics and acute antibiotic utilization (≤7 days) including total number of antibiotics, weekly rate of prescribing, and proportion of fills in three "appropriateness" categories (always appropriate, potentially appropriate, never appropriate).

Results: Three cohorts with a positive COVID-19 diagnosis code constituted a total of about 1.8 million patients (13.53%). About 22.79% of COVID-19 positive groups had severe symptoms, 24.43% had moderate symptoms and the majority, 52.78%, had no symptoms. In the analytical sample of 13 million, about 4.2 million antibiotic prescriptions were prescribed to 2.5 million patients (19%) within 7 days of the first diagnosis of either COVID-19 or COVID-19-like symptoms. Within the COVID-19 positive cohorts, about 11% received an antibiotic prescription, while the non-COVID-19 cohorts, about 19.70% received an antibiotic. Among patients with antibiotic prescriptions, about 37.01% were prescribed an antibiotic "appropriately", 39.46% were prescribed a "potentially appropriate" antibiotic and about 22.64% received an "inappropriate" antibiotic. Among patients prescribed antibiotics, azithromycin was the most common, ranging from 21.80 to 44.80% for each cohort.

Conclusions: Although the overall proportion of COVID-19 patients receiving antibiotics was much lower than non-COVID-19 patients, the findings suggest use of antibiotics persisted despite guidelines against widespread use, particularly for patients with moderate and mild COVID-19 symptoms.