COVID-19 Pneumonia: When Should Antibiotics Be Started?

When does WHO recommend avoiding antibiotics for COVID-19?

The June 2025 WHO guideline distinguishes between illness severity and the likelihood of bacterial infection. For non-severe COVID-19 with low suspicion of bacterial coinfection, its recommendation against empirical antibiotics is strong. For severe disease under the same circumstances, the recommendation is conditional, reflecting greater uncertainty. Empirical treatment means prescribing before the suspected bacterial cause has been confirmed. [WHO clinical management guideline](https://www.who.int/publications/i/item/B09467).

This guidance does not apply when bacterial infection is meaningfully suspected. Clinicians assess the patient's history, examination and laboratory findings, and reassess as the illness evolves. Severe COVID-19 alone does not establish that antibiotics will help; a bacterial complication can nevertheless develop later. The guideline therefore supports an individualized prescribing decision rather than a permanent decision made at admission. [WHO guideline, antibiotic recommendations](https://iris.who.int/bitstream/handle/10665/381920/B09467-eng.pdf?sequence=1).

What evidence shows that antibiotics were overused during the pandemic?

The figures come from a WHO analysis announced in April 2024, using approximately 450,000 hospital records from 65 countries collected between January 2020 and March 2023. They describe historical hospital care, not prescribing rates in October 2026. WHO also reported frequent use of antibiotics in its Watch category, which includes medicines with greater potential to drive resistance. [WHO antibiotic-use analysis](https://www.who.int/news/item/26-04-2024-who-reports-widespread-overuse-of-antibiotics-in-patients--hospitalized-with-covid-19).

The large difference between antibiotic exposure and documented bacterial coinfection supports concern about unnecessary prescribing. However, these aggregate percentages cannot determine whether an individual prescription was justified: clinicians sometimes must act before diagnostic uncertainty is resolved. The practical implication is to make the suspected bacterial diagnosis explicit when starting treatment and reconsider that assessment as additional information becomes available.

How does more selective prescribing protect future treatment?

CDC's February 2025 account of COVID-19 and antimicrobial resistance describes a broader hospital problem: resistant infections increased alongside disrupted infection prevention, staffing pressures and greater use of invasive devices. Antibiotic prescribing was part of this picture, but the surveillance findings do not establish it as the sole cause. Preventing infections and improving prescribing must work together. These are historical findings rather than a newly announced October 2026 outbreak. [CDC pandemic resistance report](https://www.cdc.gov/antimicrobial-resistance/data-research/threats/covid-19.html).

For patients, the distinction is between treating the coronavirus and treating a bacterial complication. Antibiotics do not act against SARS-CoV-2. COVID-19 treatment depends on severity and individual risk, and may include antiviral medicines or hospital respiratory support. Antibiotics remain valuable when a bacterial infection complicates the illness; withholding an unnecessary antibiotic should still be accompanied by an appropriate treatment and monitoring plan. [WHO COVID-19 treatment information](https://www.who.int/news-room/questions-and-answers/item/coronavirus-disease-covid-19).