Interim 2023/24 influenza A vaccine effectiveness: VEBIS European primary care and hospital multicentre studies, September 2023 to January 2024

Influenza A viruses circulated in Europe from September 2023 to January 2024, with influenza A(H1N1)pdm09 predominance. We provide interim 2023/24 influenza vaccine effectiveness (IVE) estimates from two European studies, covering 10 countries across primary care (EU-PC) and hospital (EU-H) settings. Interim IVE was higher against A(H1N1)pdm09 than A(H3N2): EU-PC influenza A(H1N1)pdm09 IVE was 53% (95% CI: 41 to 63) and 30% (95% CI: −3 to 54) against influenza A(H3N2). For EU-H, these were 44% (95% CI: 30 to 55) and 14% (95% CI: −32 to 43), respectively.

As at week 5 2024, influenza virus is circulating in Europe, with median influenza test positivity among sentinel primary care networks at 34%, and influenza A as the main virus type [1].Most countries are reporting dominance of influenza A(H1N1)pdm09 virus [1].Here, we present interim influenza vaccine effectiveness (IVE) from European primary care and hospital multicountry studies between September 2023 and January 2024.

Influenza vaccination in Europe
In the northern hemisphere for the 2023/24 season, the A/Victoria/4897/2022 (H1N1)pdm09-like clade 5a.2a.1 virus was recommended by the World Health Organization (WHO) as the influenza A(H1N1)pdm09 egg-based vaccine strain virus.For influenza A(H3N2) virus egg-based vaccines, the recommendation was an A/Darwin/9/2021 (H3N2)-like clade 2a.2 virus [2], the same as for 2022/23.Influenza vaccination target groups in the European Union (EU) include older adults, certain occupational groups, individuals with chronic conditions, and those at increased risk of influenza complications and severe disease [3].In around one third of European Union/European Economic Area (EU/EEA) countries, vaccination of healthy children is also recommended, e.g. in Ireland among those 2-17 years, Spain among those ≥ 6 months-4 years, and Romania among those ≥ 6 months-17 years.
The I-MOVE (Influenza -Monitoring Vaccine Effectiveness in Europe) primary care (EU-PC) and hospital (EU-H) multi-country studies have estimated IVE since 2008/09 (EU-PC) and 2012/13 (EU-H).These studies are now coordinated through the ECDC Vaccine Effectiveness, Burden and Impact Studies (VEBIS) project.Within these two multi-country studies, we assessed interim IVE for the 2023/24 season.Our results aim to inform the end-February 2024 WHO Vaccine Strain Selection Committee.

Study design and setting
Nine of 11 EU-PC and seven of 12 EU-H participating study sites in 10 countries reported sufficient influenza cases (≥ 10 eligible influenza cases) to be included in interim IVE analyses (Figure 1).
Using the test-negative design, patients meeting a common case definition were systematically selected for recruitment and swabbing [4,5].For EU-PC, we included patients meeting the EU acute respiratory infection (ARI) case definition: patients with sudden onset of symptoms and at least one of four respiratory symptoms (cough, sore throat, shortness of breath, coryza).
For EU-H, we included patients meeting a severe ARI (SARI) case definition: hospitalised for at least 24 h with at least one of three symptoms (fever, cough, shortness of breath) and with symptom onset within 10 days of hospital admission.Reverse transcription (RT)-PCR testing was used for influenza virus detection and type A subtyping.Patients testing negative were designated as controls, and those testing positive as cases.Controls presenting before the week of onset of the first influenza (sub)type/clade-specific case were excluded from the analysis of that (sub)type.Eight of nine sites in EU-PC and three of seven sites in EU-H selected all (or a random sample of) influenza viruspositive specimens above specified quantification cycle (Cq) values for haemagglutinin genome segment or whole genome sequencing, followed by phylogenetic analysis to determine clade distribution.

Vaccination definitions and patient characteristics
Vaccination information was obtained through GP or hospital records, by data linkage with national influenza vaccination databases, or patient interview, depending on the study site.We defined vaccinated patients as those having received 2023/24 influenza vaccine at least 14 days before onset of symptoms, and excluded those vaccinated < 14 days before symptom onset. In

Influenza vaccine effectiveness
Using logistic regression, we estimated an odds ratio (OR) for vaccination adjusted for a priori potential confounding variables of study site, age, sex, presence of chronic conditions and onset date.We calculated VE as 1-OR x 100.For models with small sample sizes (< 10 cases or controls per parameter), we performed a sensitivity analysis using Firth's penalised logistic regression [6].Where differences between this and the original estimate were ≥ 10%, estimates were not presented.
Limitations include small sample size for IVE against A(H3N2), as influenza A(H1N1)pdm09 circulation has dominated the season thus far, resulting in low precision and some stratified estimates unable to be calculated.The low proportion of available sequenced viruses reduced precision for clade-specific IVE estimates for A(H1N1)pdm09.As with all observational studies, unmeasured confounding and selection bias cannot be ruled out.The influenza season continues in Europe, and end-of-season estimates will provide more precise IVE estimates against circulating influenza strains and clades.

Conclusion
Overall, up to 53% and 44% of vaccinated individuals in primary care or hospital settings, respectively, were protected against mild and severe influenza during the 2023/24 season.Influenza vaccination should be promoted in line with national guidelines and recommendations in all European countries with ongoing influenza virus circulation.Sites included in EU-PC analysis against influenza A and A(H1N1)pdm09: Croatia, France, Germany, Ireland, the Netherlands, Portugal, Navarre, Spain and Sweden.Against influenza A(H3N2): France, Germany, Ireland, the Netherlands and Spain.Sites included in EU-H analysis of IVE against influenza A and influenza A(H1N1)pdm09: Belgium, Croatia, Germany, Malta, Navarre, Portugal and Spain.Against influenza A(H3N2): Navarre and Spain.a Age-specific or target group-specific IVE was not included for overall or (sub)type-specific IVE in some study sites, where sample size did not allow estimation of IVE.b Vaccine brand information was available for 77% (1,005/1,307) of vaccinated controls in EU-PC: all vaccines were quadrivalent, 9% (n = 88) were cell-based, 1% (n = 14) were live attenuated influenza vaccine (LAIV), 5% (n = 53) were adjuvanted vaccine, 12% (n = 118) were highdose vaccine, 73% (n = 732) were unadjuvanted, standard dose, egg-propagated, inactivated.In EU-H 77% (1,230/1,595) of vaccinated controls had available vaccine brand information: all vaccines were quadrivalent, 61% (n = 749) egg-propagated, inactivated, unadjuvanted, standard dose vaccines, 18% (n = 221) were high dose, 12% (n = 145) adjuvanted, 8% (n = 101) cell-based and 1% (n = 14) LAIV.c The IVE models were adjusted by study site, age, sex, presence of chronic conditions and onset date.The best functional forms of the continuous variables age and onset date (categories, splines, linear terms) were selected using the Akaike information criterion.d Groups targeted by seasonal influenza vaccination include older adults (aged 55, 60 or 65 years or older), those in medical risk groups and healthy children (in Ireland among those aged 2-17 years and in Spain among those aged ≥ 6 months to 4 years), as defined locally in the studies and study sites.

Figure 1
Figure 1Sites a participating in VEBIS European primary care and hospital influenza vaccine effectiveness studies, September 2023− January 2024 (n = 15) Number of influenza cases and test-negative controls by week of onset, VEBIS European (A) primary care and (B) hospital studies, September 2023-January 2024 (n = 17,016 a ) DiscussionOur results from two well-established European multicountry studies indicate 2023/24 interim VE estimates against influenza A in primary care and hospital settings were 51% and 38% among all ages, and 53% and Figure2a Numbers for subtypes A(H1N1)pdm09 and A(H3N2) are subsets of the total number of influenza A cases.

Table 1
Influenza viruses characterised by genetic clade and amino acid substitutions, VEBIS European primary care and hospital studies, September 2023-January 2024 (n = 222) European primary care multicentre VEBIS study; EU-H: European hospital multicentre VEBIS study; NC: not calculated (percentages not shown where denominators < 60); VEBIS: Vaccine Effectiveness, Burden and Impact Studies.a At time of publishing (Feb 2024), not all specimens from the study period were processed.

Table 2
Interim vaccine effectiveness against influenza A, A(H1N1)pdm09 and A(H3N2), by age group and target group for vaccination, VEBIS European primary care and hospital studies, September 2023-January 2024 CI: confidence interval; EU-H: European hospital multi-country VEBIS study; EU-PC: European primary care multi-country VEBIS study; IVE: influenza vaccine effectiveness; Vacc: vaccinated; VEBIS: Vaccine Effectiveness, Burden and Impact Studies.