Influenza vaccines are currently the best method available to prevent seasonal influenza infection. In most European countries one dose (or two doses for children) of seasonal vaccine is given from September to December to the elderly and other target groups for vaccination. In Europe, influenza seasons can last until mid-May (1), and it is expected that vaccination conveys protection on the individual for the duration of the season. In 13/15 reviewed studies on the length of vaccine-induced protection among the elderly, using anti-haemagglutination antibody titres as a proxy for seroprotection levels, seroprotection rates lasted at least >4 months after vaccination (2).
However in the 2011-12 influenza season various studies in Europe reported a decrease in influenza vaccine effectiveness (VE) against A(H3N2) over time within the season (3–5). In the United States, a decrease in VE against A(H3N2) with time since vaccination was suggested in the 2007-8 influenza season (6).
The observed decrease of VE over time can be explained by viral change (notably antigenic drift) occurring in the season. Drift in B viruses may be slower than in A viruses (7), and A(H3N2) viruses undergo antigenic drift more frequently than A(H1N1)pdm09 viruses (8).
The decrease of VE over time can also be explained by a waning of the immunity conferred by the vaccine independently from viral changes. If vaccine-induced protection wanes more rapidly during the season, then depending on the start and duration of the influenza season, the decline of VE may cause increases in overall incidence, hospitalisations and deaths. Changes to vaccination strategies (timing and boosters) may be needed.
As anti-haemagglutination antibody titres are not well defined as a correlate of protection (9,10), vaccine efficacy (as measured in trials) or vaccine effectiveness observational studies may be one way to measure vaccine-induced protection. These studies require a large sample size to model VE by time since vaccination and currently, most of the seasonal observational studies lack the precision required to provide evidence for waning immunity.
In this study we pooled data across five post-pandemic seasons (2010/11-2014/15) from the I-MOVE (Influenza - Monitoring Vaccine Effectiveness) multicentre case control studies (1,3,11,12), to obtain a greater sample size to study the effects of time since vaccination on influenza type/subtype-specific VE. We measure influenza type/subtype-specific VE by time since vaccination for the overall season, but also in the early influenza phase; under the hypothesis that virological changes are fewer in the early season, but waning of the vaccine effect should be present regardless of time within the influenza phase.
- influenza season