Influenza vaccine effectiveness in preventing hospitalization among Beijing residents in China, 2013-15.

BACKGROUND: Estimates of influenza vaccination effectiveness (VE) are valuable for populations where the vaccine has been promoted in order to support vaccination policy and to permit evaluation of vaccination strategies. Such studies would be important for China due to limited data available during seasons when the vaccine strains matched or mismatched the circulating viruses. METHODS: We conducted a test-negative study in hospitals in Beijing. Patients admitted to five hospitals in the city were enrolled during the winter influenza seasons of 2013-14 and 2014-15. Influenza virus infections were determined by PCR, and influenza vaccination records were extracted from a centralized electronic immunization registry. Influenza VE was estimated by logistic regression adjusting for age group, sex and chronic conditions, and matched by calendar week. RESULTS: A total of 2368 inpatients were recruited during the study period with a vaccination coverage in the control group of 12.8%. The overall estimate of influenza VE was 46.9% (95% CI: -20.4%, 76.6%) for the 2013-14 season and 5.0% (95% CI: -53.0%, 41.0%) for the 2014-15 season. Estimates of VE were relatively higher in children aged 6-17 years than older persons across two influenza seasons while estimates of VE for both adults and elderly were relatively low. CONCLUSIONS: Our findings were consistent with expected influenza vaccination effectiveness in seasons when the vaccine matched or mismatched circulating viruses. Strategies to increase influenza vaccine coverage could provide a public health benefit.