Effectiveness of 2009 pandemic influenza A(H1N1) vaccines: a systematic review and meta-analysis

Background: The clinical effectiveness of monovalent influenza A(H1N1)pdm09 vaccines has not been comprehensively summarised. We undertook a systematic review and meta-analysis to assess vaccine effectiveness (VE) for adjuvanted and unadjuvanted vaccines. Methods: We searched healthcare databases and grey literature from 11 June 2009 to 12 November 2014. Two researchers independently assessed titles and abstracts to identify studies for full review. Random effects meta-analyses estimated the pooled effect size of vaccination compared to placebo or no vaccination for crude and adjusted odds ratios (OR) to prevent laboratory confirmed influenza illness (LCI) and related hospitalization. VE was calculated as (1-pooled OR) ∗ 100. Narrative synthesis was undertaken where meta-analysis was not possible. Results: We identified 9229 studies of which 38 at moderate risk of bias met protocol eligibility criteria; 23 were suitable for meta-analysis. Pooled adjusted VE against LCI with adjuvanted and unadjuvanted vaccines both reached statistical significance (adjuvanted: VE = 80%; 95% confidence interval [CI] 59–90%; unadjuvanted: VE = 66%; 95% CI 47–78%); in planned secondary analyses, VE in adults often failed to reach statistical significance and pooled point estimates were lower than observed in children. Overall pooled adjusted VE against hospitalization was 61% (95% CI 14–82%); in planned secondary analyses, adjusted VE attained statistical significance in adults aged 18–64 years and children for adjuvanted vaccines. Adjuvanted vaccines were significantly more effective in children compared to adults for both outcomes. Conclusions: Adjuvanted and unadjuvanted monovalent influenza A(H1N1)pdm09 vaccines were both effective in preventing LCI. Overall, the vaccines were also effective against influenza-related hospitalization. For both outcomes adjuvanted vaccines were more effective in children than in adults