Influenza A viruses of the H3N8 subtype are a major cause of respiratory disease in horses. Subclinical infection with virus shedding can occur in vaccinated horses, particularly where there is a mismatch between the vaccine strains and the virus strains circulating in the field. Such infections contribute to the spread of the disease. Rapid diagnostic techniques are available for detection of virus antigen and can be used as an aid in control programmes. Improvements have been made to methods of standardising inactivated virus vaccines, and a direct relationship between vaccine potency measured by single radial diffusion and vaccine-induced antibody measured by single radial haemolysis has been demonstrated. Improved adjuvants and antigenic presentation systems extend the duration of immunity induced by inactivated virus vaccines, but high levels of antibody are required for protection against field infection. In addition to circulating antibody, infection with influenza virus stimulates mucosal and cellular immunity; unlike immunity to inactivated virus vaccines, infection-induced immunity is not dependent on the presence of circulating antibody to HA. Live attenuated or vectored equine influenza vaccines, which may better mimic the immunity generated by influenza infection than inactivated virus vaccines, are now available. Mathematical modelling based upon experimental and field data has been applied to examine issues relating to vaccine efficacy at the population level. A vaccine strain selection system has been implemented and a more global approach to the surveillance of equine influenza is being developed.