The influenza vaccine forms the basis of efforts to prevent the occurrence of influenza virus infection. However, vaccine effectiveness (VE) differs every season, which complicates efforts to combat the spread of infection. To develop a robust method to analyse variations in VE, we assessed VE among adult patients with influenza using a test-negative, case-control study design that evaluated vaccination records and the corresponding results of rapid influenza diagnostic tests during the 2013/14 and 2014/15 influenza seasons. During the 2013/14season, the adjusted VEs against influenza A and B viruses were 54.9% (95% confidence interval [CI] = 24.2% - 73.2%) and 56.6% (95% CI = 19.1% - 76.7%), respectively. In contrast, during the 2014/15season, the adjusted VE against the influenza A (H3N2), virus was -2% (95% CI = -66% - 37.5%). Moreover, only a few patients were infected with the influenza B virus, thus, the VE against influenza B could not be assessed. The low VE during the 2014/15 season could be attributed to antigenic drift in the circulating influenza A (H3N2) viruses and mutations in the egg-adapted vaccine strains. Estimation of the VE against the influenza virus using this test-negative, case-control study design was simple and easy, and this study design had a precision similar to that of a randomized control trial. Therefore, this study design could be employed to predict VE through out the influenza season and may be used as the basis of influenza prophylaxis.