WHO updates influenza guidelines with targeted antivirals and clear limits on add-on drugs

Influenza remains a major cause of illness and death, especially in older adults and people with chronic conditions. New randomised trial data, ongoing concern about zoonotic (novel influenza A) infections, and wider access to rapid diagnostic tests prompted WHO to update its guidance. This BMJ Rapid Recommendation summarizes the WHO guideline first published in September 2024.

The panel asked which medications work for non-severe and severe influenza, including zoonotic disease, which drugs prevent infection in exposed contacts, and which testing strategies best support fast, accurate treatment decisions. The evidence came from four systematic reviews of randomised trials, a systematic review of prognostic factors for progression to severe disease, and a decision-analysis model comparing diagnostic pathways. A panel judged free of important conflicts of interest developed the recommendations.

The guideline contains 29 recommendations: 21 on antivirals and six on adjunctive therapies. For seasonal influenza, WHO conditionally recommends oseltamivir for severe illness and baloxavir for non-severe patients at high risk of progression, with treatment within 48 hours of symptom onset. Prophylaxis with baloxavir, laninamivir, oseltamivir or zanamivir is conditionally recommended for anyone exposed to zoonotic influenza and for extremely high-risk people exposed to seasonal influenza. Antibiotics are strongly recommended against in non-severe influenza. In severe influenza, the panel conditionally recommends against corticosteroids, macrolides, mTOR inhibitors, NSAIDs and passive immune therapy. A recommendation on diagnostic testing is also made for both non-severe and severe disease.

The guidance supports a risk-stratified approach: treat those who are severely ill or likely to deteriorate, and avoid routine add-on drugs. The main caveat is that the trial evidence was low to very low certainty, which explains why most recommendations are conditional. This summary is based on the abstract only, so the details of the risk thresholds and testing pathways are not covered here.