In 13.8 million older French adults, a high-dose flu shot was linked to fewer flu hospitalizations

The immune system gets slower and less sharp with age. One consequence is that vaccines work less well in older people: the body responds less strongly to a shot. That matters for the flu, which sends many older adults to hospital every winter. To make up for it, manufacturers developed enhanced flu vaccines for older people, among them a high-dose vaccine that contains more of the viral protein the immune system learns to recognise.

Randomized trials have suggested that such vaccines protect better than standard-dose ones. But most countries have few real-world numbers that compare the products directly in the same population. France offered an unusual opportunity: in the 2023–2024 season, one high-dose and three standard-dose vaccines were all in use, free of charge for this age group, and the choice was left to the doctor or pharmacist. So each product was given to people of all ages. A team at EPI-PHARE, a research group of the French medicines agency and the national health insurance, used this to compare them. The study is a preprint, which means it has not yet been peer-reviewed.

What they did

The team used the national health insurance records, which cover almost everyone living in France. They included all 13,788,297 adults born in 1958 or earlier in mainland France who had used any health service in the previous year. They left out about 97,000 people in nursing homes whose vaccinations could not be tracked individually. They then looked for two things between December 2023 and April 2024: a hospital stay with a flu diagnosis, and death during such a stay.

Because nobody was randomly assigned a vaccine, the team had to make the groups comparable by statistics. They used two designs:

• Vaccine against vaccine. People who had already received a high-dose or a standard-dose shot before the epidemic began were matched on sex, age group, region and nursing-home residence, and balanced for illnesses such as diabetes, heart disease and cancer, and for how often they saw a doctor.
• Vaccine against no vaccine. Every week of the season, people newly protected by a shot (counted from 14 days after it) were paired with similar people who had not been vaccinated yet. This is a way of imitating a series of trials with observational data.

What they found

• At the start, 5.7 million people (41%) had been vaccinated; about a third of them had the high-dose vaccine. Over the season there were 25,214 flu hospitalizations and 2,199 deaths during them.
• Compared with no vaccine, vaccination was linked to a 39% lower risk of flu hospitalization and a 45% lower risk of dying in hospital with flu.
• Against hospitalization, the high-dose vaccine was linked to 50% protection, the standard-dose vaccines to 36%. Put differently: about 610 people needed a high-dose shot to prevent one flu hospitalization, against about 1,100 for a standard shot. For deaths, this analysis showed no clear difference between vaccine types.
• In the direct comparison, high-dose recipients had a 27% lower risk of flu hospitalization and a 27% lower risk of dying with flu in hospital than standard-dose recipients. About 1,600 people would have to receive the high-dose instead of a standard shot to prevent one hospitalization.
• The high-dose vaccine did better than two of the three standard vaccines; against the third, the team found no clear difference.
• Protection from any flu vaccine became smaller with age. Between 64 and 69, high-dose and standard shots could not be told apart; from 70 on, the high-dose vaccine was linked to more protection.

What to keep in mind

• Observational, not randomized. Vaccinated and unvaccinated people differ in ways records cannot capture. The team checked this with “negative control” outcomes that a flu shot should not affect, hospital stays for injuries and for cataracts, and found that vaccination status was still linked to them. The authors take this as a sign of leftover confounding and think it more likely makes the vaccines look more effective than they are. The comparison between vaccine types was less affected by this.
• One season. The 2023–2024 epidemic was almost entirely influenza A. The authors note that the results may not hold when other virus strains circulate.
• The records did not say which flu strain each patient had. Other outcomes the team looked at, such as pneumonia and heart problems, gave results similar in size to the injury control, so they are hard to interpret.
• The 27% advantage of the high-dose vaccine is in the lower range of what a pooled analysis of randomized trials found (about 27% to 49%).
• The study did not look at side effects. Flu vaccines are given by doctors and pharmacists, and which one a person receives depends on national rules; this study is not advice on which to choose.

Why it matters

Most flu vaccine studies in older people either compare products or measure protection against no vaccine. This one did both, for almost an entire national population, with very narrow margins of error. Its message is in line with earlier trials: flu vaccines give older adults moderate protection against the worst outcomes, and a vaccine designed for an aging immune system was linked to extra protection, especially from 70 on. The authors say their results support choosing vaccine products by age group. As a preprint, the study still has to pass peer review.