- Design
- Systematic review and random-effects meta-analysis of observational studies with GRADE assessment, PROSPERO registered, commissioned by WHO
- Population
- 2,150,518 patients with laboratory-confirmed seasonal influenza across 63 studies, mean ages 3.6 to 87.9 years
- Primary outcome
- Hospital admission and all-cause mortality
- Effect
- Admission: cardiovascular disease OR 2.72 (1.24-5.94), immunosuppression 2.70 (1.55-4.70), chronic respiratory disease 2.24 (1.90-2.64). Mortality: secondary bacterial infection 4.13 (1.53-11.14), malnutrition 3.29 (1.57-6.89), sepsis 3.19 (2.08-4.89)
Sixty-three observational studies enrolling 2,150,518 patients with laboratory-confirmed seasonal influenza, published between 2000 and October 2025, were pooled to support an update of the WHO influenza clinical guidelines. Only adjusted effect estimates were included, and certainty was graded.
For patients with non-severe influenza, the risks for hospital admission at moderate or high certainty were cardiovascular disease (OR 2.72, 95 per cent CI 1.24 to 5.94), HIV or other immunosuppression (2.70, 1.55 to 4.70), any neurological disease (2.31, 1.65 to 3.24), chronic respiratory disease (2.24, 1.90 to 2.64) and older age (1.72 per 10-year increase in adults, 1.02 to 2.89), with pregnancy (1.88), diabetes (1.84) and malignancy (1.75) adding further risk.
For patients already severe, the drivers of death are different and largely acquired rather than pre-existing: secondary bacterial infection (4.13, 1.53 to 11.14), malnutrition (3.29, 1.57 to 6.89), sepsis (3.19, 2.08 to 4.89) and acute kidney injury (2.92, 1.11 to 7.73) led, ahead of age 65 or over (2.47), cardiovascular disease (2.43), malignancy (2.21), neurological disease (2.08), liver disease (1.86), immunosuppression (1.79) and COPD (1.74).
That split is the useful part. The admission list tells you whom to treat early and vaccinate; the mortality list is dominated by complications you can look for and treat. Malnutrition sitting second is worth noting in Indian practice, where it is both common and rarely recorded, and secondary bacterial infection topping the list is an argument for looking rather than for empirically covering everyone.
- Use the admission risk list to decide who gets antiviral treatment early rather than symptomatic advice.
- In a severe case, actively look for secondary bacterial infection - it carried the highest odds of death.
- Record nutritional status; malnutrition ranked second among mortality risks and is usually not documented.
- Watch renal function serially in severe influenza; acute kidney injury nearly tripled the odds of death.
- Treat pregnancy, diabetes and malignancy as admission risks even when the illness looks mild.
The statistics, in plain English
These are pooled adjusted odds ratios from observational studies, so they describe association in populations where confounders were handled differently in each study - the GRADE ratings are the guide to how much weight each deserves. Note the width: cardiovascular disease at 1.24 to 5.94 and secondary bacterial infection at 1.53 to 11.14 are certain in direction and very loose in magnitude. The age estimate for admission, 1.02 to 2.89, only just excludes no effect despite the enormous sample, which reflects how differently age was modelled across studies.
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.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for pulmonology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free