Influenza vaccine effectiveness in Wisconsin during the 2007-08 season: Comparison of interim and final results

Background: During the 2007-08 influenza season, we reported an interim vaccine effectiveness (VE) estimate of 44% for preventing medically attended influenza. In this analysis we report results for the entire season and compare them with the interim estimate. Methods: Patients with feverishness, chills, or cough <8 days duration were prospectively recruited over 10 weeks and tested for influenza by real-time reverse transcriptase PCR (rRT-PCR). Case-control analyses were performed using data from patients with rRT-PCR confirmed influenza (cases) and ill patients without influenza (test-negative controls). VE was estimated as 100 X (1 - adjusted odds ratio) in a logistic regression model adjusting for age, week, and high risk medical condition. A sample of influenza isolates was antigenically characterized. Results: Influenza was detected by rRT-PCR in 865 (44%) of 1972 patients; 73% were type A and 27% were type B. VE was 37% (95% Cl, 22-49%) overall and 44% (95% Cl, 27-58%) among participants tested 0-3 days after illness onset. VE was 39% (95% Cl, 2-62%) in children 6-59 months old and 37% (95% CI, -2% to 61%) in adults >= 50 years old. VE was 41% (95% Cl, 24-53%) for influenza A and 31% (95% Cl, 3-51%) for influenza B. All 24 characterized influenza A viruses were antigenically matched to the H3N2 vaccine strain, although 14 viruses exhibited mild antigenic drift. There was a lineage mismatch with the vaccine strain for all 39 characterized influenza B viruses. Conclusions: The 2007-08 influenza vaccine provided modest protection against medically attended influenza in this population. The interim estimate of VE after 17 days closely approximated the final season VE, supporting the potential use of interim VE estimates while influenza seasons are still in progress. (C) 2011 Elsevier Ltd. All rights reserved.