5. Changing mortality
During the post-war period, the mortality of the oldest-old has declined substantially. It may therefore be helpful to examine at this point the general nature of this decline in order to better appreciate the changing parameters in the later chapters.
A general view of the change emerges in Figure 5 which shows the progress made from one decade to the next, by detailed age in the core group of thirteen countries. The decline has accelerated notably over time for both sexes and at all ages below 100 years. Throughout the period, the decline has been faster for women than men.The age pattern of decline is very clear and consistent for women, the relative benefits for the 80-year-old being always more than twice that for 95-year-old. Around age 100 a definite decline is still observed but it is more modest and an acceleration is not obvious. The pattern for men is more ambiguous. Up to 1970s, the decline was hardly at all age-selective but between the last decades it took on the form observed for women.
The decline by country and aggregate is given in Table 5 by age groups and illustrated in Figures 6a and b.The curves in Figure 6a, showing a 20-year change in individual countries, produce for males a rather ambivalent picture which includes some contrasting patters. In the first frame, England & Wales, West Germany and Italy display an almost level decline of 10 - 15 percent while the reductions in France and Japan are strongly concentrated among the youngest. In the second frame, Sweden shows an even 10 percent decline while in Norway and Denmark the reduction increases, in Finland and Iceland decreases with age. In the third group, the East European countries show actual mortality increase at younger ages and slight decline after that but Scotland and New Zealand a fairly constant 10 percent decrease.
In the last frame, a contrast is evident between gains that increase in the Netherlands, stay level in Belgium and Austria and decrease in Portugal and particularly in Switzerland. In a general way, where the decline is concentrated on oldest ages, it is altogether slight, vz. Eastern Europe, the Netherlands and Norway. In countries with largest overall gains, these are most impressive among the 80-year-old as witnessed by Japan, Switzerland, France and Finland.
These diverging age patterns are connected with what happens to the younger elderly men. In Monograph I in this series, a connection was made with death rates at ages 60 - 79 published by the United Nations Demographic Yearbook, ( various editions), and it showed that in countries with large declines for octogenarian men (Japan, France, Finland), as large or still larger declines were recorded for the younger elderly. On the other hand, in the Netherlands, Norway, Denmark and Eastern Europe where the decline was smaller for octogenarians, there was some actual increase in the mortality of the younger elderly men (Kannisto, 1994, p. 46).
The changes regarding females are confirmed in Figure 6b as adhering closely to one single pattern of large declines at age 80 which then become gradually smaller. This common pattern operates on widely different levels. Relatively small gains are recorded for Eastern and Southern Europe and Austria. The countries with the largest gains are the same four as with males though in a different order: Switzerland, Finland, France and Japan. In Norway and the Netherlands, the slight gains for men, particularly the younger, contrast with large gains for women, and again particularly the younger ones. The mean values of qx by age group and country are given in Annex Table 5.
If the present stage of epidemiological transition is a manifestation of delayed senescence or at least delayed terminal frailty, then it can be expected to have more effect on the relatively younger elderly and less at older ages where vitality ebbs. The characteristic age pattern of the new stage of mortality transition is therefore one of larger gains among the younger, observed among women in all countries of the database and among men where overall progress has been good. In some countries, a contrasting tendency of obviously different etiology has caused the death rates of middle-aged and elderly men to stagnate or rise, sending ripple effects to older ages and so interfering with the general decline.
Updated by V. Castanova, 1 November 1999