expectancy at the ages when death occurs: in Sweden, a female death shortly after birth would contribute 83 years, whereas a death at age 83 would contribute 7.5 years. The average of such values, weighted by the number of deaths at each age, gives a life disparity of 9 (Zhang and Vaupel, 2009). Zhang and Vaupel (unpublished) performed analyses of the correlation between life disparity in a specific year and life expectancy in that year for men and women in 33 countries and regions. They found that during the 168 years from 1840 to 2007, 113 holders of record life expectancy also had the lowest life disparity. Countries with long life expectancy tend to have low life disparity because these countries have been successful in reducing premature deaths—doing so increases life expectancy and reduces life disparity. That is, efforts to avert deaths that occur at ages well below the life expectancy of a population appear to be especially effective in increasing life expectancy—and, simultaneously, reducing life disparity. Analyses of life disparity in Denmark show that a slowing of progress in reducing differentials in life spans occurred at about the same time as the slowing of progress in increasing life expectancy (see Figures 14-4a and 14-4b).
Analyses of cause-specific mortality for men and women in Denmark show that mortality rates from major causes of death, such as heart disease, have declined since the 1970s. However, lung cancer mortality increased for women throughout the second half of the 20th century. For men the increase was more pronounced until around 1980, when the rate stabilized. For alcohol-related mortality, an increase is seen from 1970 onward for both genders, again most pronounced for men. Denmark is now among the countries with the highest tobacco- and alcohol-related mortality rates in 20 OECD countries (see Figures 14-5a and 14-5b), when alcohol-related deaths are calculated from alcohol-related diagnoses from death certificates and tobacco-related deaths are calculated from the method of Peto et al. (1992).
These cause-specific mortality rates correspond to the trend in the incidence of major underlying diseases. Figure 14-6 shows the dramatic increase in lung cancer among women in Denmark compared with other countries in the same time period. Figure 14-7 shows the dramatic decline in heart disease mortality in all the study countries, with Denmark, however, still having the highest mortality among women at the end of the period.
Peto et al. (1992) developed a method that uses absolute age- and sex-specific lung cancer rates to indicate the approximate proportions of deaths due to tobacco not only from lung cancer itself but also, indirectly, from vascular disease and various other categories of disease. This method was applied by Brønnum-Hansen and Juel (2000) to Danish data from the early