Mortality attributable to influenza is difficult to estimate because death certificates do not often include relevant data. Thompson et al have developed a statistical model using national mortality and viral surveillance data to estimate influenza deaths by type and subtype from 1976 to 1999 and, for the first time, deaths from respiratory syncytial virus (RSV) from 1990 to 1999. Using a new technique, they looked at data for underlying respiratory and circulatory deaths as these figures were more likely to include influenza related deaths than if only pneumonia and influenza deaths were considered. There was a significant increase in influenza associated deaths over this period (p<0.001). Influenza was estimated to be associated with a mean of 3.1% of underlying respiratory and circulatory deaths and RSV with 1%. Influenza A (H3N2) was associated with the most deaths, followed by RSV. The bulk of deaths from both viruses occurred among those aged over 65. Influenza caused more deaths in all age groups than RSV, except in children younger than 1 year.

This study underlines the importance of prevention measures against viral infections in our ageing population, and the need for development of an effective vaccine against RSV.