Prospective hospital-based case-control study to assess the effectiveness of pandemic influenza A (H1N1)pdm09 vaccination and risk factors for hospitalization in 2009-2010 using matched hospital and test-negative controls

Background: We performed a case-control study to estimate vaccine effectiveness (VE) for prevention of hospitalization due to pandemic influenza A(H1N1)pdm09 (pH1N1) and to identify risk factors for pH1N1 and acute respiratory infection (ARI) in 10 hospitals in Berlin from December 2009 to April 2010. Methods: Cases were patients aged 18-65 years with onset of ARI <= 10 days before admission testing positive for pH1N1 by PCR performed on nasal and throat swabs or by serological testing. Cases were compared to (1) matched hospital controls with acute surgical, traumatological or other diagnoses matched on age, sex and vaccination probability, and (2) ARI patients testing negative for pH1N1. Additionally, ARI cases were compared to matched hospital controls. A standardized interview and chart review elicited demographic and clinical data as well as potential risk factors for pH1N1/ARI. VE was estimated by 1-(Odds ratio) for pH1N1-vaccination >= 10 days before symptom onset using exact logistic regression analysis. Results: Of 177 ARI cases recruited, 27 tested pH1N1 positive. A monovalent AS03-adjuvanted pH1N1 vaccine was the only pandemic vaccine type identified among cases and controls (vaccination coverage in control group 1 and 2: 15% and 5.9%). The only breakthrough infections were observed in 2 of 3 vaccinated HIV positive pH1N1 patients. After exclusion of HIV positive participants, VE was 96% (95%CI: 26-100%) in the matched multivariate analysis and 46% (95%CI: -376-100%) in the test-negative analysis. Exposure to children in the household was independently associated with hospitalization for pH1N1 and ARI. Conclusions: Though limited by low vaccination coverage and number of pH1N1 cases, our results suggest a protective effect of the AS03-adjuvanted pH1N1 vaccine for the prevention of pH1N1 hospitalization. The use of hospital but not test-negative controls showed a statistically protective effect of pH1N1-vaccination and permitted the integrated assessment of risk factors for pH1N1-infection. To increase statistical power and to permit stratified analyses (e.g. VE for specific risk groups), the authors suggest pooling of future studies assessing effectiveness of influenza vaccines for prevention of severe disease from different centres.