Which flu vaccine protects against contracting flu

The short version: no flu vaccine reliably prevents laboratory-confirmed infection, and the evidence shows live attenuated (nasal spray) vaccines outperform injected ones in head-to-head trials, but protection is modest and wanes quickly.

The best direct comparison comes from a 1984 randomized challenge study in young adults. The live attenuated cold-adapted nasal vaccine prevented infection in 13 of 16 people (81%), while the injected inactivated vaccine prevented infection in only 5 of 16 (31%) [1]. The nasal vaccine also nearly eliminated virus shedding, reducing it about 1000-fold compared to the shot [1]. A 2006 trial in children with recurrent respiratory infections found the live nasal vaccine was 53% more effective than the injected vaccine against matched strains, though it offered no advantage against H3N2 [2].

But these numbers come with serious caveats. The 1984 study was funded by the vaccine developers and used a controlled challenge, not real-world exposure [1]. The 2006 pediatric trial was funded by Wyeth and MedImmune, the manufacturers [2]. Real-world effectiveness is much lower. A 2011/12 UK study found the injected vaccine provided only 23% protection against H3N2, and that protection dropped from 53% to 12% within a single season [3]. In elderly Danes that same season, the shot showed negative effectiveness against influenza A (-11%), meaning vaccinated people were slightly more likely to get sick [5].

The mechanism matters. Injected vaccines produce mostly serum antibodies but weak mucosal immunity. The live nasal vaccine generates nasal IgA and cellular responses that block infection at the portal of entry [1]. But repeated annual vaccination blunts these responses: a 2020 pilot study found people who got the shot every year had significantly lower antibody-secreting cell responses than those who skipped years [4].

My call: the live attenuated nasal vaccine provides better protection against infection than the injected shot, but neither is reliable. Protection is modest, wanes within months, and the evidence base is thin, manufacturer-funded, and lacks long-term safety comparisons against truly unvaccinated people. Confidence: moderate for the head-to-head comparison, low for real-world effectiveness.