Study objectives: This study aimed to assess total influenza mortality among the elderly (> or = 75 years old) in France, and to evaluate how many deaths may have been avoided through vaccination during the past 10 years.
Design: The monthly mortality rates related to different causes among the elderly were obtained from the national mortality statistics for the period 1978-90. For each cause, the proportion of the registered death rate attributable to influenza was estimated using time series models. Each model analysed the registered death rate for the considered cause as a linear function of the registered influenza death rate for that month, the secular trend, and the seasonal variations. This yielded yearly regression coefficients for influenza. Formulas were subsequently developed to estimate the death rates avoided as a result of influenza vaccination according to the level of vaccine coverage and the hypothetical effectiveness of the vaccine.
Main results: Between 1980 and 1990 registered influenza death rates ranged from 11-81 per 100,000. The number of deaths attributable to influenza but registered as resulting from another cause was up to eight times the number of deaths registered as influenza. Total influenza death rates were estimated as ranging from 28 per 100,000 (1988-89) to 482 per 100,000 (1985-86). At the same time it was estimated that the use of influenza vaccine avoided from 7 per 100,000 deaths in 1981-82 to 697 per 100,000 deaths in 1989-90, depending on the intensity of the epidemic, the vaccine coverage, and the vaccine effectiveness.
Conclusions: These results support the policy of promoting influenza vaccination among the elderly.