Overall and stratified estimates of influenza vaccine effectiveness in Europe for the season 2010-2011Archived

ECDC Scientific Advance 2011-12-12

Overall and stratified estimates of influenza vaccine effectiveness in Europe for the season 2010-2011 I-MOVE Multi-Centre Case Control Study 2010-11: Overall and Stratified Estimates of Influenza Vaccine Effectiveness in EuropeKissling E, Valenciano M, Cohen JM, et al. PLoS ONE, 2011; 6(11): e27622

This paper gives the results from the third season of the ‘Influenza Monitoring Vaccine Effectiveness in Europe’ (I-MOVE) Project. This is a multi-centre case-control study based on sentinel practitioner surveillance networks in eight European Union (EU) member states. The paper estimates 2010/11 influenza vaccine effectiveness (VE) against laboratory-confirmed as influenza. The methodology used by the authors included a systematic sampling whereby practitioners swabbed ILI/Acute Respiratory Infection (ARI) - affected patients within seven days of symptom onset. They compared influenza-positive patients to influenza laboratory-negative patients among those meeting the EU ILI case definition. A valid vaccination was counted when the time lapse between receiving a dose of vaccine and symptom onset was longer than 14 days. For the statistical analysis multiple imputation with chained equations was used to estimate missing values. The authors then employed logistic regression with study as fixed effect and calculated the resulting influenza VE adjusting for potential confounders. Apart from the overall influenza VE, the authors also estimated stratified VE values ranked by influenza type/subtype, age group and target group for vaccination. In total, 2019 cases and 2391 controls were included in the analysis.

The overall adjusted VE was 52% (N = 4410; 95% CI: 30-67).  The VE against the predominant influenza A(H1N1) was 55% (95% CI: 29-72) and against the influenza B VE was 50% (95% CI: 14-71). When considering the age group strata, the adjusted VE against all influenza subtypes was 66% (95% CI: 15-86) among those aged 0-14 years old, 41% (95% CI: -3-66) among those aged 15-59 years old and 60% (95% CI: 17-81) among those aged ≥60 years old. Finally, the stratification by target group for vaccination showed that the overall VE was 56% (N=1004; 95% CI: 34-71), the VE against influenza A(H1N1) was 59% (95% CI: 32-75) and the VE against influenza B was 63% (95% CI: 31-81).

The authors concluded that moderate protection was given by the trivalent influenza vaccines used for the winter season 2010-11 against laboratory-confirmed influenza across Europe. In addition, the authors note that it was possible to calculate the adjusted and stratified influenza VE estimates from this multi-centre case-control study thanks to the large sample size they had available from using the same methodology in eight countries.

Moreover, they state that I-MOVE is a clear example of how a network can provide precise summary VE measures across Europe.

ECDC Comment (12 December 2011):

Some initial findings of this study were published in March in Eurosurveillance, and these are now the definitive results for this season. This is a remarkable achievement in that by successfully combining data from several countries the I-MOVE collaboration, Epiconcept and ECDC has been able to stratify the results in a way that normally cannot be done with results from a single country. Previous studies have been summarized in  Scientific Advance documents published in two parts, Part 1 and Part 2. Estimating, understanding and communicating about influenza vaccine effectiveness are difficult tasks (1-3). Effectiveness is best defined as ‘the likelihood of an intervention preventing an outcome when it is applied in the field’. This is often determined by well designed observational studies allowing for adjustment for possible confounding factors. High serological titres  after immunisation can be confused with high effectiveness (70-90%) when the actual effectiveness estimates for seasonal influenza vaccination are like those documented by Osterholm et al. (1-6). There is a need to be realistic and not overly optimistic and include independent estimates of effectiveness in post-marketing surveillance of vaccines (7). There are also difficulties in the end points of studies of influenza vaccines, whether they are laboratory confirmed infections, hospitalisations or deaths. Non-specific end-points like ILI are confusing and should be avoided (2). It is sometimes stated that large randomised trials should be conducted in order to resolve these uncertainties (8), but this is both unfeasible and unethical as these trials would entail offering placebos to people who would benefit from an intervention of proven effect. In addition, and because of the variability observed in VE between influenza seasons since the 2009 pandemic (9), any single trial could not be generalized to current and future experience. Well-conducted trials contribute to this field but they have to be complemented by observational studies (1).