Admin_Adham
TOPLINE
In a large real-world analysis of patients hospitalised for COVID, about 75% received antibiotics, although clinically suspected coinfections were reported in fewer than 10% of cases. Antibiotic use during hospitalisation was linked to a higher risk for 28-day in-hospital mortality across severity groups.
METHODOLOGY
- Researchers conducted a multicountry cohort study from 2020 to 2023 using real-world data to assess whether empirical antibiotic use was associated with in-hospital mortality among patients hospitalised for COVID and to describe bacterial coinfections, antibiotic use, and resistance.
- The analysis used harmonised patient-level hospital data from 64 countries and included 592,898 hospitalised patients of all ages with suspected or confirmed SARS-CoV-2 infection; of these, 451,291 had records of antibiotic use (mean age, 61.9 years; 251,990 male individuals).
- Empirical antibiotic use was defined as treatment started within 24 hours of hospital admission or started later if the clinician reported the intent as empirical.
- The study outcome was 28-day in-hospital mortality, with follow-up from admission until death, discharge, transfer, or day 28, whichever occurred first.
TAKEAWAY
- Among the 451,291 patients with data on antibiotic use, 341,125 (75.6%) received antibiotics; however, a bacterial coinfection was suspected or confirmed in only 47,050 of the 592,898 patients (7.9%).
- Antibiotic use was most common in the World Health Organization (WHO) Eastern Mediterranean (83.0%) and African (82.7%) regions and least common in the Western Pacific region (32.8%). Treatment was mostly empirical, and most antibiotics belonged to the Watch group of the WHO Access, Watch, and Reserve classification.
- After adjustment, empirical antibiotic use was associated with a higher risk for 28-day in-hospital mortality than no empirical use in both mild-to-moderate disease (adjusted hazard ratio [aHR], 2.21; P < .001) and severe-to-critical disease (aHR, 1.08; P < .001).
- Antibiotic use was higher in severe-to-critical disease than in mild-to-moderate disease (P < .0001). Of 4237 bacteria with resistance data, 1815 (42.8%) were multidrug resistant, including 53.4% of Enterobacterales resistant to third- and fourth-generation cephalosporins.
IN PRACTICE
"[The study] findings highlight the widespread and unnecessary use of broad-spectrum antibiotics among patients hospitalised for COVID-19 worldwide, including those with non-severe illness and no evidence of co-infection. Low diagnostic capacity and high MDR [multidrug-resistant pathogen] prevalence underscore the urgent need for enhanced diagnostic infrastructure and stronger diagnostic and antimicrobial stewardship," the authors wrote.
SOURCE
The study was led by Daniele Dona, MD, PhD, University of Padua, Padua, Italy. It was published online on October 23, 2026, in The Lancet Microbe.
LIMITATIONS
The study was observational, so data entry errors and missing data could have affected the reported link between antibiotic use and outcomes. Illness severity was recorded only at admission, not at the time antibiotics were started. Moreover, infection onset was reported inconsistently, potentially biasing results and preventing a clear distinction between community- and hospital-acquired infections.
DISCLOSURES
The study did not receive any funding. The authors reported having no relevant conflicts of interest.
This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.
