Influenza vaccination and mortality: prospective cohort study of the elderly in a large geographical area.

The 50% reduced overall mortality previously associated with influenza vaccination among the elderly was based on studies neither fully taking into account systematic differences between individuals who accept or decline vaccination nor encompassing the entire general population. A population-based prospective cohort study was performed in Stockholm County (Sweden), including all persons aged > or =65 yrs (n = approximately 260,000), over three influenza seasons: 1998/1999, 1999/2000 and 2000/2001. The relative risks of mortality among vaccinated versus unvaccinated individuals were estimated using Cox's proportional hazards regression adjusted for, and stratified by, demographic factors and comorbid conditions during the three seasons and the respective following off-seasons. Influenza vaccination was associated with an unadjusted reduction in all-cause mortality during the three seasons of 50, 46 and 42%, respectively, which decreased to 14, 19 and 1%, respectively, following adjustment for confounders and differences in mortality between vaccinated and unvaccinated individuals following the influenza season. The numbers needed to treat to prevent one death, during the three seasons, were 297, 158 and 743, respectively. Vaccination remains the most important measure for prevention of influenza complications in elderly people, although the effectiveness in reducing all-cause mortality in elderly persons is lower than previously thought.