Skip to content
Buy This GLP
← Research
Evidence

The benefit did not scale with the dose

Semaglutide users had lower COVID and flu severity than matched metformin users. The advantage was the same at the lowest dose as at the highest.

Neil Sanders7 min read
Composite severity, by weekly dose — COVID P = .50, flu P = .320.25–0.5 mg1.0 mg1.7–2.4 mgidenticalMore drug bought no more protection. Read that twice.

Two findings sit awkwardly together in this study, and the awkwardness is the interesting part: a large relative benefit, and no sign of it growing with more drug — the opposite of what a dose-dependent effect looks like.

What was compared

Adults with a documented COVID-19 or influenza infection, classified by whether they had been taking semaglutide beforehand or metformin without prior GLP-1 exposure. [1] Matching covered demographics, body mass index, glycated hemoglobin, vaccination status, prior infection, antiviral use, healthcare utilization and baseline comorbidities, leaving 7,092 matched pairs for COVID and 1,920 for influenza.

That is a more thorough matching list than most studies of this kind manage, and it is why the results are worth taking seriously rather than dismissing — it is the opposite end of the spectrum from a matched comparison that let follow-up differ by two hundred days.

The numbers, with their bases

COVID 30-day mortality ran 0.3% against 0.8%, relative risk 0.39. Hospitalization 5.3% against 7.4%. The composite severity outcome 8.2% against 10.6%. In influenza, mortality 0.2% against 0.7%, hospitalization 6.5% against 9.6%.

Five fewer deaths per thousand infections is what a relative risk of 0.39 means on a base of 0.8%. Intensive care admission and mechanical ventilation were not significantly different in either cohort, which are the outcomes least sensitive to who ends up coded as hospitalized.

Where it held up best

Acute kidney injury was significantly lower in both cohorts after adjustment for multiple comparisons, at relative risks of 0.73 and 0.58. Acute coronary syndrome and acute respiratory failure were not.

The benefit also held among unvaccinated patients and among adults over 65, and remained among people who received standard COVID therapeutics — which is the kind of consistency across subgroups that a confounding explanation has to work harder against, though it does not eliminate it.

What a buyer takes from it

Nothing to act on. Nobody should buy a GLP-1 to reduce their risk from influenza, and the authors ask for prospective evaluation rather than claiming the case is made.

What is worth carrying is the check. When a benefit does not increase with more of the drug, that is either a clue about mechanism or a clue about confounding, and the study design decides which — the same question behind any comparison run at doses somebody chose. The absolute figures here are also a reminder that a large relative risk on a small base is a small number of people.

Frequently asked

Does semaglutide protect against severe COVID or flu?
This matched observational study found lower mortality and hospitalization against metformin users. It is not randomized, and the authors call for prospective evaluation.
How big was the mortality difference?
0.3% against 0.8% for COVID and 0.2% against 0.7% for influenza — roughly five fewer deaths per thousand infections.
Why does a flat dose-response matter?
It supports the authors' reading that the effect is not driven by weight loss. It is equally consistent with the benefit belonging to the kind of patient prescribed the drug rather than to the drug.
Did the most serious outcomes improve?
Intensive care admission and mechanical ventilation were not significantly different in either cohort.

Sources

  1. [1] Matson R, et al. (2026). Semaglutide use linked to lower COVID-19 and influenza severity with better renal outcomes after pandemic or seasonal infection Biology Methods and Protocols. PMID 42540132

Where to get it

Price the injectable sellers

The desk lists every seller that publishes an injectable figure, with the advertised price struck against the one a buyer is billed.

Open the price desk

More in Evidence