COVID-19 vaccines before the 2026-27 season: the protection is against severity, and it holds

Design

systematic review of randomised trials, comparative observational studies and safety studies; four databases searched August 2025 to June 2026

Population

155 eligible publications on US-licensed COVID-19 vaccines and US SARS-CoV-2 epidemiology, from 11,829 references screened

Primary outcome

vaccine effectiveness against hospitalisation and other COVID-19 outcomes, and adverse events of special interest

Effect

hospitalisation VE 53.0% (95% CI 37.0-65.0) in adults 65+, 54.0% (1.0-78.0) in adults 18-64; paediatric ED visits 76.0% (58.0-87.0)

This annual synthesis screened 11,829 references and included 155 publications — 15 randomised trials, 84 comparative observational studies, 38 single-group safety studies and 18 burden descriptions — on US-licensed COVID-19 vaccines. Against hospitalisation, updated vaccines were associated with a 53.0% reduction in adults aged 65 and over (95% CI 37.0 to 65.0, 2025-26 season) and 54.0% in adults aged 18 to 64 (95% CI 1.0 to 78.0, 2024-25 season).

The more decision-relevant numbers concern people who cannot be relied on to mount their own response. Vaccination in pregnancy was associated with a 58.0% reduction in the mother's own emergency or urgent care encounters (95% CI 24.0 to 77.0) and 50 to 54% fewer COVID-19 hospital contacts in her infant's first two months, and — importantly for how the conversation is framed — this held regardless of the trimester in which she was vaccinated. In children aged 9 months to 4 years, effectiveness against emergency department visits was 76.0% (95% CI 58.0 to 87.0). In immunocompromised adults, protection against intensive care admission or death ran from 32.0% to 53.0% depending on the condition.

On safety, adverse events of special interest matched previous reviews, and no study conducted under the extended primary-series dosing interval reported raised myocarditis or pericarditis risk. Almost all the effectiveness evidence here is observational, so residual confounding by health-seeking behaviour cannot be excluded — but the consistency across populations, outcomes and seasons is what carries the conclusion, not any single estimate.

• Offer vaccination in pregnancy at whatever gestation the woman presents — infant protection did not depend on trimester
• Frame the benefit as prevention of severe disease and hospitalisation, not of infection
• For immunocompromised adults, give the condition-specific range rather than a single headline figure
• The 18-64 interval reaches 1.0% at its lower bound; treat that estimate as imprecise, not as equal to the older-adult one
• Availability, product and schedule differ in India from the US-licensed products reviewed here — check current CDSCO-approved options before quoting these figures locally

The statistics, in plain English

A vaccine effectiveness of 53% means hospitalisations were about half as common among vaccinated people as unvaccinated — it is a relative figure, so what it is worth to an individual depends on their baseline risk. The 18-64 estimate (54%, 95% CI 1.0 to 78.0) has a lower bound almost touching zero, meaning the data are compatible with almost no benefit in that group; the older-adult estimate is far more precise. Observational effectiveness studies compare people who chose vaccination with people who did not, and those groups differ in ways that adjustment cannot fully remove.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.