The European I-MOVE Multicentre 2013-2014 Case-Control Study. Homogeneous moderate influenza vaccine effectiveness against A(H1N1)pdm09 and heterogenous results by country against A(H3N2).

Abstract

Background: In the first five I-MOVE (Influenza Monitoring Vaccine Effectiveness in Europe) influenza seasons vaccine effectiveness (VE) results were relatively homogenous among participating study sites. In 2013-2014, we undertook a multicentre case-control study based on sentinel practitioner surveillance networks in six European Union (EU) countries to measure 2013-2014 influenza VE against medically-attended influenza-like illness (ILI) laboratory-confirmed as influenza. Influenza A(H3N2) and A(H1N1)pdm09 viruses co-circulated during the season. Methods: Practitioners systematically selected ILI patients to swab within eight days of symptom onset. We compared cases (ILI positive to influenza A(H3N2) or A(H1N1)pdm09) to influenza negative patients. We calculated VE for the two influenza A subtypes and adjusted for potential confounders. We calculated heterogeneity between sites using the I(2) index and Cochrane's Q test. If the I(2) was <50%, we estimated pooled VE as (1 minus the OR)×100 using a one-stage model with study site as a fixed effect. If the I(2) was >49% we used a two-stage random effects model. Results: We included in the A(H1N1)pdm09 analysis 531 cases and 1712 controls and in the A(H3N2) analysis 623 cases and 1920 controls. For A(H1N1)pdm09, the Q test (p=0.695) and the I(2) index (0%) suggested no heterogeneity of adjusted VE between study sites. Using a one-stage model, the overall pooled adjusted VE against influenza A(H1N1)pdm2009 was 47.5% (95% CI: 16.4-67.0). For A(H3N2), the I(2) was 51.5% (p=0.067). Using a two-stage model for the pooled analysis, the adjusted VE against A(H3N2) was 29.7 (95% CI: -34.4-63.2). Conclusions: The results suggest a moderate 2013-2014 influenza VE against A(H1N1)pdm09 and a low VE against A(H3N2). The A(H3N2) estimates were heterogeneous among study sites. Larger sample sizes by study site are needed to prevent statistical heterogeneity, decrease variability and allow for two-stage pooled VE for all subgroup analyses.