Antibiotic Prescribing Practices During the COVID-19 Pandemic: A Post-Hoc Analysis of the Adaptive COVID-19 Treatment Trial.

Background: The COVID-19 pandemic exacerbated antibiotic overuse; however, gaps remain regarding the timing of administration and whether prescribing is supported by evidence of bacterial infection. Methods: Antibiotic use was described within the Adaptive COVID-19 Treatment Trial (ACTT), a series of randomized controlled therapeutics trials in adults hospitalized with COVID-19 from Feb 2020 to May 2021. Proportions of participants receiving antibiotics by trial stage were described (overall, pre-randomization, post-randomization). Analyses were repeated for CDC's National Healthcare Safety Network (NHSN) categories and for antibiotics used for community-acquired pneumonia (CAP) according to IDSA guidelines. In ACTT-2 and ACTT-4, lab culture results were documented for Grade 3 or greater infection adverse events; proportions of antibiotic recipients with a positive culture were compared between trials. Results: Most (62%) of the 4074 participants received at least one antibiotic. Pre-randomization antibiotic use declined across ACTT stages, from 72% in ACTT-1% to 32% in ACTT-4. The largest decrease noted is within the NHSN category for community-acquired infection (decreasing significantly from 49% in ACTT-1% to 24% in ACTT-4). Post-randomization use was lower and declined modestly, from 28% to 18%. The percentage of post-randomization antibiotic recipients with a positive bacterial culture was 18% in ACTT-2% and 18% in ACTT-4. Conclusions: Antibiotic use decreased substantially over the first pandemic year; however, in-hospital prescribing remained common despite limited microbiologic confirmation of infection. Stewardship efforts during future respiratory viral pandemics should focus on reducing empiric antibiotic use for CAP-like illness at presentation and on strengthening reassessment, diagnostic stewardship, and de-escalation of later in-hospital antibiotic therapy.