Seasonal influenza vaccine efficacy and its determinants in children and non-elderly adults: A systematic review with meta-analyses of controlled trials

CONTEXT: The true level of influenza vaccine efficacy is controversial and many factors may influence its estimation. OBJECTIVES: To estimate the efficacy of vaccination of children and non-elderly adults for the prevention of influenza and to explore the impact of type of vaccine, age, degree of strain matching, influenza type and case ascertainment methods on vaccine efficacy estimates. DATA SOURCES: Medline and EmBase databases until October 2011. References of relevant articles were also reviewed. STUDY SELECTION: Controlled trials evaluating seasonal influenza vaccines and presenting incidence of laboratory-confirmed influenza illness were eligible. Studies exploring efficacy after experimental challenge, presenting duplicate data, employing group randomization, or focusing on special populations were excluded. DATA EXTRACTION: The vaccine effect on influenza prevention was evaluated by calculating Mantel–Haenszel risk ratios (RR) and using random-effects models. Vaccine efficacies were calculated for each comparison as (1−RR)×100. RESULTS: Thirty studies were included in one or more of a total of 101 analyses, comprising 88.468 study participants. There was evidence of heterogeneity in 49% of the analyses. Summary vaccine efficacy was 65% against any strain, 78% against matched strains and 55% against not-matched strains. Both live-attenuated and inactivated vaccines showed similar levels of protection against not-matched strains (60% and 55%, respectively). Live-attenuated vaccines performed better than inactivated vaccines in children (80% versus 48%), whereas inactivated vaccines performed better than live-attenuated vaccines in adults (59% versus 39%). There was a large difference (20%) in efficacy against influenza A (69%) and influenza B (49%) types for not-matched strains. Summary estimates of vaccine efficacy were highest when ascertainment was based on culture confirmation. CONCLUSION: Influenza vaccines are efficacious, but efficacy estimates depend on many variables including type of vaccine and age of vaccinees, degree of matching of the circulating strains to the vaccine, influenza type, and methods of case ascertainment.