A study of excess mortality during influenza epidemics in the Czech Republic, 1982-2000.

Influenza related mortality rates have been established in many countries; nevertheless, studies focusing on the Central European population have been rare to date. We assess mortality attributable to influenza by comparing all cause mortality and mortality due to diseases of the circulatory system during influenza epidemic and non-epidemic periods, as defined by acute respiratory infection surveillance data. Data on total mortality, mortality due to diseases of the circulatory system and surveillance data for influenza and other respiratory infections were used in a general linear model with a logarithmic link for dependence of left censored mortality data over time, and week as a categorical factor. Results of the analysis show statistically significant (p <0.001) differences in excess mortality rates between influenza epidemic and non-epidemic periods in the Czech Republic between 1982 and 2000. We estimate that 2.17% of all cause mortality, and 2.57% of mortality due to diseases of the circulatory system throughout the study period was attributable to influenza, with an estimated annual average of 2661 and 1752 deaths respectively. The highest numbers of deaths were reported during seasons when influenza A/H3N2 was the predominant circulating strain. Improving vaccination coverage against influenza is considered to be the primary strategy for prevention of influenza associated mortality.