1Efforts to reduce mortality during the second half of the twentieth century in Europe have led to substantial gains in life expectancy. Yet the various countries of Europe have followed very different paths to achieve this result. Benoît Haudidieranalyses this diversity in relation to France and the former Federal Republic of Germany, two neighbouring countries with very close ties and which have recorded practically equivalent decreases in mortality.

2In fact, analysis of the main causes of death shows that the reasons for this decrease are not always the same. Cardiovascular disease, and cancer among women, two series of causes whose links with lifestyle and consumption are well known, have evolved more unfavourably in the former FRG, while France comes out worse for mortality from violent causes (road traffic accidents, suicides, etc.) and from infectious diseases. The cultural differences between the two countries, taken in the broadest sense, appear to play a role, despite growing similarities in lifestyle.

3By promoting a bilateral perspective, comparative studies such as the one presented here should contribute to progress in reducing mortality in either country.

4Life expectancy in France and the former Federal Republic of Germany (FRG) rose substantially in the second half of the twentieth century, as it did throughout the whole of western Europe. The evolution was very similar in both countries. Between 1958 and 1997, life expectancy at birth for men increased by 7.8 years in France and by 7.9 years in the former FRG, rising respectively from 66.8 to 74.6 years and from 66.5 to 74.4 years. For women, the increase was 9.2 and 8.9 years respectively, rising from 73.1 to 82.3 years in France and from 71.6 to 80.5 years in the former FRG. Over a period of 40 years, progress in the two countries was identical, with life expectancy remaining equal for males and becoming more favourable for French women. Though life expectancies in many western European countries converge for both males and females, only in France and the former FRG has life expectancy progressed in an identical or almost identical manner since the late 1950s. The fact that the two countries belong to two contrasting cultures — Latin and Germanic — makes this similarity all the more striking. The territory of the former FRG is used here both on practical grounds (it is difficult to reconstitute homogeneous statistics for the period prior to unification), and for a more fundamental reason: the two parts of Germany have followed two very different social models. The comparison therefore concerns two countries based on a western social model: France and the former FRG. The analysis of cause- and age-specific mortality was continued until 1995 because the conditions prevailing before unification continued to shape mortality trends for several years afterwards.

5Beyond the general indicator of life expectancy, which reveals identical overall trends in France and the former FRG, does the same rule apply to all aspects of mortality, notably with respect to age or cause of death? We will start by exploring the reasons for the interruption in the decline in infant mortality in the former FRG between 1968 and 1973. We will then focus on the similarities and differences in mortality from infectious diseases, violent causes and degenerative pathologies. Lastly, the differences between men and women that may appear when comparing mortality in the two countries will also be examined.

I – General observations and limitations of the study

6The similarity of overall mortality trends in the two countries conceals substantial age-specific variations (Figure 1). The differences are clearly visible. Before age 45 for males and before age 30 for females, mortality has fallen more in the former FRG than in France since 1975, whereas above these ages, progress has been greater in France throughout the period.

Figure 1

Standardized age- and sex-specific rates of mortality from all causes in France and the former FRG, 1950-95

Standardized age- and sex-specific rates of mortality from all causes in France and the former FRG, 1950-95

7This pattern is not surprising, since statistics for the whole of western Europe from 1950 to 1995 show that the growing convergence of life expectancies has in fact been accompanied (since 1965 at least) by a relative resistance to decline in the mortality of young people in Mediterranean Europe and of old people in northern Europe (Meslé and Vallin, 2002). But given the growing similarity between the living conditions of the population in the two countries and the increasing interpenetration of their political, economic and social organization, the true scale of the phenomenon deserves closer examination.

II – Infant mortality: a more irregular trend in the former FRG than in France

13Infant mortality — the mortality of children under the age of one year — decreased substantially in western Europe during the second half of the twentieth century. France and the former FRG are no exceptions (Figure 2), as the rate was divided by ten between 1950 and 1995, from more than 50‰ to 5‰ (respectively 5.2 and 4.9 in the former FRG and in France). Though the rates in the two countries were very similar at the beginning and end of the period, the pattern of change is nevertheless quite different. The decline merely slowed down in France during the 1980s, but rates levelled off durably between 1968 and 1973 in the former FRG. Although this halt was temporary, its effects were felt until the early 1980s. It is worth taking a closer look by examining changes in each component of infant mortality: early neonatal mortality (0-6 days), late neonatal mortality (7-27 days) and post-neonatal mortality (28-365 days).

Figure 2

Infant mortality rate and its main components in France and the former FRG, both sexes, 1950-95

Infant mortality rate and its main components in France and the former FRG, both sexes, 1950-95

14In both countries, as in all industrialized countries, the proportion of early neonatal deaths among all deaths under the age of one year (Figure 3) increased until around 1970 due to the faster decline of mortality from infectious disease, and of exogenous mortality in general (Bourgeois-Pichat, 1946). The medicalization of birth and progress in perinatal medicine halted this increase in the 1970s, and from then on, early neonatal mortality decreased faster than the other components of infant mortality. After a plateau of several years, the two curves in Figure 3 started moving in opposite directions.

Figure 3

Proportion of deaths occurring in the early neonatal period in France and the former FRG, 1950-95

Proportion of deaths occurring in the early neonatal period in France and the former FRG, 1950-95

15There are substantial differences between the two countries, however. In France, early neonatal mortality declined more rapidly as early as 1968, whereas late neonatal and post-neonatal mortality fell more slowly. In the former FRG, on the other hand, a slower decrease in early neonatal mortality was observed from 1965 on, while the decline in late neonatal and post-neonatal mortality came to an abrupt halt. In all three cases, the trend is even reversed, with a substantial rise for the first component between 1968 and 1970, and an even sharper increase for the two others (respectively from 1968 to 1977 and from 1969 to 1978).

16It looks very much as if the structural change brought about by new medical techniques — exactly as in France — was accompanied in the former FRG by a global aggravating factor powerful enough to cancel out the progress achieved in combating mortality during the first week of life and to reinforce the effect of postponement of deaths until a later age. According to certain German analysts (Rückert, 1972; Höhn, 1978), this halt in progress can be attributed to the new demographic characteristics of women of reproductive age: an increased mean age at childbearing, a growing proportion of single and foreign mothers. Though this hypothesis has some support, it goes against the fact that the relative share of births to unmarried and foreign women only started to become significant in 1970. Moreover, detailed simulations show that the excess mortality resulting from these factors amounts only to 0.3‰. Lastly, note that these new characteristics of births are not specific to the former FRG, though one of them — the relative proportion of babies born to foreign women — increased in the FRG more quickly than elsewhere due to the particularly rapid decline of overall fertility.

17It would appear that other factors contributed to the interruption in the decline of infant mortality from 1968 to 1973. The break of 1968 occurred at the very time when women born during or after the Second World War were reaching their most fertile ages. In a previous article, (Haudidier, 1995), a relative vulnerability (resistance to decline in mortality) was identified among these same birth cohorts for the period 1960-90. More generally, the break occurred at a time when dominant values were being challenged and when intergenerational conflict was increasing. It occurred in conjunction with the sudden drop in fertility and the spectacular increase, between 1965 and 1975, in alcohol consumption among young German women. Figure 4, which compares mortality from cirrhosis above age 40, partly illustrates this trend and shows the difference between birth cohorts in Germany and France.

Figure 4

Age-specific female rates of mortality from cirrhoses in France and the former FRG, 1950-95

Age-specific female rates of mortality from cirrhoses in France and the former FRG, 1950-95

18All in all, it would appear that the sudden halt in the decline in infant mortality in the former FRG can be linked to the state of health of some women of reproductive age and to growing alcohol consumption among the young. Figure 5 shows that the interruption of the decline in infant mortality is caused mainly by perinatal pathologies and congenital anomalies that are closely linked to the health of the mother. Among perinatal pathologies, those often associated with prematurity appear to have played a key role (Leutner, 1976). The link between prematurity and maternal alcohol consumption and smoking is well known.

Figure 5

Main causes of death of children under age one in France and the former FRG, both sexes together, 1950-95

Main causes of death of children under age one in France and the former FRG, both sexes together, 1950-95

19This halt in the decline of infant mortality is also observed in Austria, another Germanic country, but not in the former German Democratic Republic (GDR). Austrian women appear to have shared the fate of their West German counterparts, whereas the women of the GDR seem to have avoided a major conflict of generations.

III – Comparative evolution of the different pathologies

20The growing differentiation between the French and German age-specific mortality structures is accompanied by equally distinctive trends in cause-specific mortality. This is shown in Figure 6 which presents, for the different groups of causes of death and for each sex, the ratios of standardized mortality rates for all ages in the former FRG to those of France, indexed to the corresponding French rates (base 100).

Figure 6

Ratio of standardized sex- and cause-specific mortality rates in the former FRG to those in France (all ages) 1950-95

Ratio of standardized sex- and cause-specific mortality rates in the former FRG to those in France (all ages) 1950-95

21From an overall viewpoint, Figure 6 shows major differences in the patterns of evolution by cause for both sexes. The sharp increase in the relative excess mortality of the former FRG for cardiovascular diseases over the entire period contrasts, since the early 1970s, with a growing German advantage with regard to violent deaths and infectious diseases. Diseases of the digestive system and of the respiratory system (not shown here) are similar for both countries, however. With regard to cancers, despite the emergence of an advantage for German men and the persistent below-average mortality of French women, they are only a secondary differentiating factor of overall mortality between the two countries. Analysis of their evolution is important, however, given that — alongside cardiovascular diseases and cirrhoses — they account for a major share of the sex-specific mortality differences between the two countries. Infectious diseases, violent deaths, cardiovascular diseases, cancers and cirrhoses will be examined in succession.

1 – Infectious diseases: excess mortality in France

22Excess mortality due to infectious diseases in France is not a new phenomenon. It was observed from 1950 for all ages and both sexes (Figure 7). This excess mortality decreased and even disappeared in the 1960s, because of a slowing down in the German mortality decline. It reappeared in the early 1970s and has increased steadily since then (except over 60, where it peaked in the early 1980s).

Figure 7

Standardized age- and sex-specific rates of mortality from infectious diseases in France and the former FRG, 1950-95

Standardized age- and sex-specific rates of mortality from infectious diseases in France and the former FRG, 1950-95

23This resurgence of excess male mortality from infectious disease in France since the 1970s varies according to age. At 60 and over, it is due mainly to a slower decline in mortality from infectious disease in France than in the former FRG. At younger ages, excess mortality in France increases later, from 1983 on; the probability of dying increases in both countries at this time, though it is higher in France. The two patterns of evolution, above and below age 60, can be attributed to very different phenomena. The first, above age 60, is caused by a resistance to the decline in non-specified infectious diseases, but also by increased mortality due to certain pathologies such as septicaemia. The negative incidence of this set of pathologies is probably the result of new factors giving rise to higher lethality (notably, the appearance and spread of more resistant bacteria and nosocomial infections). The fact that resistance to decline emerges later in the former FRG may be linked to more scrupulous compliance with the rules of hygiene in German hospitals and a much less massive use of antibiotics.

24Below age 60, the emergence of the AIDS epidemic explains both the sudden renewed upward trend in the standardized male mortality rate in both countries and the substantial increase in French excess mortality. For this age interval, AIDS is the increasingly dominant infectious disease [2] and its incidence is much higher in France than in the former FRG [3]. Without questioning the impact of AIDS on excess mortality in France, the hypothesis of underestimation of German mortality for the 15-29 age group, itself due to the under-reporting of AIDS deaths, cannot be ruled out. There is a suspicious dip in the German curve associated with infectious diseases in 1988, and a sudden corresponding increase in deaths recorded under items 304 and 305 of the detailed ICD list (drug dependence and non-dependent abuse of drugs). French excess mortality from AIDS should be attributed to individual risk behaviour, more frequent than in the former FRG, rather than to other factors such as failings of the country’s healthcare system.

25The evolution of female mortality from infectious diseases in both countries runs clearly parallel to that of male mortality while being less pronounced. However, due to the early age of female AIDS mortality and the lower mortality in the former FRG, the risk of death from infectious disease at age 15-29 in France exceeds that of the FRG at ages 30-59 towards the end of the observation period.

2 – Violent deaths: a multi-faceted divergence between France and Germany

26Violence is the cause of death which has evolved most unfavourably for France over the entire period. The standardized mortality rates for the two sexes were similar in both countries in 1950. But after a period of almost fifty years, mortality from violent causes in France reached a level almost double that of the former FRG, for both males and females. As shown clearly in Figure 8, the difference between the two countries grows progressively wider, especially from the early 1970s, with a trend reversal for mortality from violent causes. The various components of the total — road traffic accidents, suicides and other violent deaths — have all contributed to this trend.

Figure 8

Standardized sex-specific rates of mortality from violent causes (all ages) by main type of cause in France and the former FRG, 1950-95

Standardized sex-specific rates of mortality from violent causes (all ages) by main type of cause in France and the former FRG, 1950-95

Road traffic accidents

27The evolution of road traffic accidents follows — with larger fluctuations — the overall pattern closely: rapid and almost identical increase in the probability of dying in the former FRG and in France during the first two decades (excluding the first three years of considerable increase in France), followed by a large drop due to the first oil shock and to new road safety measures, more pronounced in the former FRG than in France. Until 1970, mortality from road traffic accidents was the only cause of increased male mortality from violent causes in both countries and the main cause for females. It then became a differentiating factor between the two countries. By the end of the period, the difference was very large, with France totalling 8,000 annual road deaths (for a substantially smaller population), compared with 6,000 in the former FRG. What is the reason for this terrible death toll in France, despite the adoption of rules that were very similar to and, in some cases, even stricter than those of its German neighbour during the period [4]? Certain problems specific to France have been clearly identified, the main one being that of drunk driving, responsible for one-third of fatal road accidents [5]. After a marked decline, this proportion has not varied in France since 1980. In the former FRG on the other hand, it fell from 21% in 1975 to 14% in 1998. This situation is highly paradoxical, given that patterns of alcohol consumption in both countries have been converging since the 1960s (substantial decrease in France, increase in the former FRG).

28Other ills are less visible, but equally formidable over the long term: widespread flouting of laws and regulations, combined with laxity and inconsistency in road traffic policing [6]. No specific evidence is available to confirm that things are any different in the former FRG. Note however that the most important road safety laws enacted in the former FRG in 1973, 1974 and 1984 had a much more immediate and substantial impact on mortality than their French equivalents (Figure 8).

Suicides

29Generally speaking, comparative analysis of mortality from suicide is limited by the fact that data are not strictly comparable due to differences in the registration practices in each country and in the extent of dissimulation. In the case of France and the former FRG, this observation is especially true, since the dominant religious traditions have shaped very different attitudes and behaviour [7].

30It can nevertheless be said that in overall terms, the standardized rates are identical for men up to the end of the 1970s, but 50% higher in the former FRG for women (Figure 8). In the 1980s and 90s, standardized rates became higher in France, and the gap between the two countries increased for both men and women, due to the increase in French rates and the drop in German ones. By 1995, the standardized rate in France was around 40% higher than that of its German neighbour for both sexes.

31Analysis by age sheds light on the contrasting trends in the two countries (Figure 9, Table 1). In the 1950s, mortality from suicide is much higher in the former FRG before age 35 for men and up to age 55 or 60 for women. After a much larger relative increase in suicides among young people and adults in France up to 1975-79, the difference between the two countries decreases. For men in the 15-24 age group, the gap narrows down considerably for men at 15-24, and still appreciably at 25-34. For women, the gap decreases before age 45, but very little above that age.

Figure 9

Standardized age-and sex-specific rates of mortality from suicide in France and the former FRG, 1950-95

Standardized age-and sex-specific rates of mortality from suicide in France and the former FRG, 1950-95

Table 1

Suicides in the former FRG and in France in 1955-59, 1975-79 and 1990-94. Sex- and age-specific standardized mortality rate* (per 100,000) and ratio of standardized rates of the former FRG and France (%)

Age Sex- and age-specific standardized mortality rate* (per 100,000) 1955-59 1975-79 1990-94 Age Former FRG France Former FRG France Former FRG France Men 15-24 18.5 6.7 24.1 14.7 14.0 16.5 25-34 24.2 16.5 32.8 24.3 22.3 35.9 35-44 29.4 27.0 38.7 30.3 24.8 44.2 45-54 45.0 48.4 44.2 38.2 28.4 42.5 55-64 50.2 61.8 44.0 41.7 30.6 39.4 65-74 46.0 64.6 49.6 54.2 35.3 49.3 Women 15-24 7.8 4.0 8.8 5.4 4.1 4.8 25-34 9.9 5.3 11.8 8.9 7.2 9.9 35-44 14.1 7.6 15.8 11.0 8.7 13.8 45-54 20.9 13.4 22.4 13.4 12.1 17.1 55-64 22.9 18.6 26.3 17.9 12.8 18.0 65-74 22.8 22.1 27.4 21.4 16.0 18.2 Age Ratio of standardized rates of the former FRG and France (%) Men Women 1955-59 1975-79 1990-94 1955-59 1975-79 1990-94 15-24 275 164 85 195 163 85 25-34 147 135 62 187 133 73 35-44 109 128 56 186 144 63 45-54 93 116 67 156 167 71 55-64 81 106 78 123 147 71 65-74 71 92 72 103 128 88 *Based on the standard European population formerly used by the WHO. Sources : Database of causes of death in France on the INED server, based on cause-of-death statistics established by INSEE (1950-1967) and INSERM (since 1968). For the former FRG, data published by the German Federal Statistical Office.

Suicides in the former FRG and in France in 1955-59, 1975-79 and 1990-94. Sex- and age-specific standardized mortality rate* (per 100,000) and ratio of standardized rates of the former FRG and France (%)

32The substantial increase in the frequency of suicide among young people in France may be attributable to the breakdown in traditional values in the 1960s and 70s, especially visible among the baby-boom generations (Surault, 1995). This effect was apparently less pronounced in the former FRG, probably because suicide rates among young people and adults were already very high.

33In the second period, 1975-1995, the sharp drop in suicide rates among all age groups in the former FRG contrasts clearly with the rapid rise in France, notably between ages 25 and 44 for men and between ages 35 and 54 for women. For France, a close link has been established between the rapid rise in unemployment and the increase in suicidal behaviour (Nizard, 1998). Why is no such link observed in the former FRG, despite an equally depressed labour market since the early 1970s [8] (except among young people aged 15-24)? We must simply conclude that many other factors are involved in suicidal behaviour and that the correlation between unemployment and suicide implies a causal relationship only under certain clearly defined socio-cultural conditions.

Other violent deaths

34Though their relative share has decreased substantially over the whole period, violent deaths from causes other than those described above still accounted for around 50% of the total in both countries in 1995. They therefore play a preponderant role in the overall trends of mortality from trauma in each country and hence in the growing gap in mortality from violent causes observed between the two. The trend of other violent deaths, primarily non traffic-related accidental deaths, was a declining one in the former FRG and in France since the 1950s, but the slope is much steeper in the former FRG. This situation doubtless testifies to more effective accident prevention in the former FRG. However, one happy exception is worthy of note: for violent deaths below age 10 in the 1990s, the rates for France are not markedly higher.

3 – Cardiovascular disease: a growing advantage for France

35Contrary to mortality from infectious diseases and violence, mortality from cardiovascular disease is evolving much more favourably in France than in the former FRG. This situation is not unexpected, since France occupies a privileged position in Europe with respect to this pathology. The trends have been diverging since 1950, creating an ever wider gap between the two countries (Figure 10).

Figure 10

Standardized sex-specific rates of mortality from cardiovascular diseases at ages 30 and over by main components in France and the former FRG, 1950-95

Standardized sex-specific rates of mortality from cardiovascular diseases at ages 30 and over by main components in France and the former FRG, 1950-95

36From levels that were practically equal in 1950, excess male mortality in the former FRG had reached 70% by 1995. This difference is the result of contrasting trends between the two countries in mortality rates from “other cardiovascular diseases” (which include heart diseases). The decrease is rapid and continuous in France, while in the former FRG it is slower and, more importantly, preceded by an upward trend until the 1970s. Cerebrovascular mortality starts off at a much higher level in the former FRG [9], though the decline is similar in both countries. The acceleration of progress observed from the 1970s is more rapid in France.

37For women, the divergence for cardiovascular mortality in general is as large as for men, despite a generally much more favourable evolution for the female sex in both countries. As German mortality was higher than that of France from the start, the relative difference between the two countries at the end of the period is even larger than for men: by 1995, for all ages, excess mortality from cardiovascular diseases in Germany had reached 80% for women, compared with 60% for men.

38Since we are dealing with other cardiovascular diseases, the sudden jump in the German curve in 1968-70 is perhaps no more than a statistical artefact. It may be the result of the sharp reduction in the number of deaths attributed to senility thanks to more accurate reporting of causes of death following the 8th revision of the ICD.

39Given the scale of the divergence between France and Germany, one may well wonder whether different practices for the reporting and certification of causes of death may not be partly to blame. The comparison of cardiovascular mortality between different European countries (WHO, 1973) largely invalidates this hypothesis, however. It reveals strong similarities between the French and German coding practices for ischemic heart disease, the most common pathology in this category. The only noteworthy difference is a stronger propensity to report “other heart diseases” in France and hypertensive heart disease in the former FRG.

40Can the divergence between France and Germany be attributed solely to individual predisposition (obesity, hypercholesterolemia, type 2 diabetes, arterial hypertension) and risk factors (smoking, sedentary lifestyle, dietary imbalance) habitually associated with the onset of cardiovascular disease? The body mass indices recorded in Germany in 1995 [10] (Table 2) and the strong prevalence of hypercholesterolemia and hypertension (respectively 80% and 30%) suggest a predisposition of the Germanic population to cardiovascular pathologies which might itself result from a greater propensity to physical inactivity (Table 3), and perhaps also from heavy tobacco consumption. Strangely, on the other hand, the indicators of food consumption, are rather unfavourable to France, since both the mean number of calories consumed per person and per day and the proportion of saturated fats in overall fatty acids are higher than in Germany (Table 4). It is possible that these indicators are not entirely comparable, or that they are not adequate to represent the whole range of nutritional factors.

Table 2

Body mass indices (BMI) in the FRG and in France in 1995

Body mass index (BMI) (% of the population) Severely Underweight Underweight Normal BMI Obese Severely obese FRG France FRG France FRG France FRG France FRG France Men 0.4 0.6 1.8 5.4 69.1 72.9 23.2 16.5 5.4 4.6 Women 1.9 4.6 8.9 20.5 71.1 59.7 15.1 10.4 3.8 4.4 Source: EUROSTAT, Key figures on health, 2000.

Body mass indices (BMI) in the FRG and in France in 1995

Table 3

Indices of physical inactivity in 1997, in France and in the FRG, by level of education

Level of education FRG France Primary 14.0 2.2 Secondary 8.2 1.1 Higher 4.3 1.0 Note: The higher the index, the higher the propensity to physical inactivity. Source: EUROSTAT, Key figures on health, 2000.

Indices of physical inactivity in 1997, in France and in the FRG, by level of education

41Data relating to alcohol consumption is also difficult to interpret (Table 4). After movement in opposite directions — temporarily upward in the former FRG and downward in France — which certainly helps to explain the more favourable evolution of cardiovascular mortality in the latter country, average consumption in both countries reached similar levels in 1990, though the negative effects were apparently more harmful in the former FRG. The type and mode of consumption in each country must be taken into account: relatively moderate consumption of wine during meals in France; somewhat excessive consumption of beer and, to a lesser extent, wine outside mealtimes in the former FRG.

Table 4

Indicators of food consumption in the former FRG and in France

Indicators of food consumption Former FRG France Mean number of calories (per person and per day)* 1970 3166.1 3300.4 1995 3396.5 3535.7 Proportion of lipids in total adult energy intake, including alcohol (%) 38.8 37.0 Proportion of saturated fats in total fatty acids (%) 42.0 48.0 Alcohol consumption per person and per year (decalitres of pure alcohol) 1970 1980 1990 1970 11.7 16.7 1980 12.8 16.0 1990 11.8 13.2 * Recommended food intake is around 2,900 kilocalories for men (aged 25-50, around 176 centimetres in height and weighing 79 kilos) and 2,200 kilocalories for women (aged 25-50, around 163 centimetres and 63 kilos). Certain nutritional bodies recommend that saturated fats, damaging to health, should not represent more than 30 to 33% of the total energy from fatty acids in food. Sources: France: Agence française de sécurité sanitaire des aliments (AFSSA). Individual food consumption survey (ASPCC 1994) and individual and national food consumption survey (INCA survey 1999); Former FRG: Nutrition Survey and Risk Factor Analysis Study in West Germany 1987-1988 (VERA).

Indicators of food consumption in the former FRG and in France

To conclude, there is no simple causal relation between the available indicators of food consumption and the levels of cardiovascular mortality. It is nonetheless true that the difference in coronary heart disease mortality (Figure 11) between the Mediterranean countries (where it is low) and the northern countries (where it is much higher) can undoubtedly be attributed to diet, alongside other types of consumption or cultural factors (Keys, 1986; Kuschi et al., 1995), and that in the north/south gradient, the former FRG and France occupy positions that correspond to their geographical location.

Figure 11

Standardized sex-specific rates of mortality from heart disease (ages 40-84) in selected European countries, 1950-95

Standardized sex-specific rates of mortality from heart disease (ages 40-84) in selected European countries, 1950-95

4 – Cancers and cirrhoses: sex-specific differences in mortality between the former FRG and France

42Excepting cardiovascular diseases, two causes of death present very different patterns by sex between the two countries: cancers in terms of level, and cirrhoses in terms of trends.

Cancer mortality: generally similar levels for men and more unfavourable for German women

43On the cancer mortality scale in western Europe, the former FRG is among the high-mortality countries, for both men and women, whereas France is a high-mortality country for men and a low-mortality country for women. Figures 12a and 12b plot standardized mortality rates over time for all cancers and for the main sites.

Figure 12a

Standardized sex-specific rates of mortality from neoplasms at ages 30 and over in France and the former FRG, 1950-95

Standardized sex-specific rates of mortality from neoplasms at ages 30 and over in France and the former FRG, 1950-95

Figure 12b

Standardized sex- and site-specific rates of mortality from neoplasms at ages 30 and over in France and the former FRG, 1950-95

Standardized sex- and site-specific rates of mortality from neoplasms at ages 30 and over in France and the former FRG, 1950-95

44For males, the similar overall mortality in both countries during the 1950s (Figure 12a) is not found at the level of the sites (Figure 12b), with the exception of the “other sites”. Mortality from cancers of the aerodigestive tract is higher in France, in contrast with cancers of the lung, stomach and intestine which are markedly higher in the former FRG [11]. This balance gradually shifts in favour of the former FRG, due to the declining weight of cancer of the stomach and the much more marked increase in France associated with cancer of the lung — which catch up with the rates in the former FRG — and of “other sites”, for which mortality in France overtakes that of the former FRG from the 1960s on. These developments lead to an excess mortality from cancers in France of around 10% from the late 1970s, despite a rapid reduction in French excess mortality from cancers of the upper aerodigestive tract.

45In fact, globally speaking, apart from cancers of the lung, the upper aerodigestive tract and the oesophagus, male cancer mortality has increased only marginally in France and the former FRG, despite the effect of alcohol consumption and smoking on other sites (pancreas, kidney, bladder, etc.) (Figure 12a).

46Fom the 1950s, the pattern for women is different from that for men. It is characterized by an excess German mortality of over 20% for all sites except the upper aerodigestive tract. Despite a generally much more favourable overall trend — regular decrease due to the dominant weight of “other sites” — the gap between the two countries remained unchanged up to the end of the observation period.

47The relative advantage of French women with respect to cancers of the stomach and intestine, greater than that of French men (mortality respectively 35% and 25% lower than that of the former FRG) is not unexpected. It is explained by the importance of the nutritional factor [12]: as women are generally more attentive than men to the real needs of their body, it is probable that French women have reaped the benefits of a more balanced diet even more than their male compatriots.

48German women, on the other hand, are less well placed than German men (compared with France), for cancers associated with alcohol consumption and smoking: lung cancers from the early 1970s on, cancers of the upper aerodigestive tract and, in part, “other sites” over the entire period. For cancers of the upper aerodigestive tract, the relative disadvantage of German women is the highest. The contrast between the positions of women in relation to men in each country is striking (Figure 12b). This situation looks destined to persist, since a dual trend, divergent among women — to the detriment of German women — and convergent among men, has emerged in recent years.

Mortality from cirrhoses: reversal of the German and French positions, unfavourable evolution for German women

49Beyond the spectacular convergence of the two countries, comparison of standardized mortality rates from liver cirrhosis beyond age 30 shows that the German mortality rate exceeds that of France from around 1990 (Figure 13). Mortality from cirrhosis has been declining in France since 1970. A similar trend is observed, beginning several years later in the former FRG, but for men only; it is not perceptible for women. The differing trends for men and women in the former FRG appear to indicate that the growing similarity in male and female behaviour may counteract the effects of trends common to both sexes. Hence, the decline in alcohol consumption in German society since the mid-1970s contrasts with a trend towards more “masculine” behaviour among women and an associated increase in alcohol consumption for them. All in all, the two factors cancel each other out, as shown by the relative stability of the women since 1965. standardized mortality rate of German women since 1965.

Figure 13

Standardized sex-specific rates of mortality from cirrhoses at ages 30 and over in France and the former FRG, 1950-95

Standardized sex-specific rates of mortality from cirrhoses at ages 30 and over in France and the former FRG, 1950-95

Conclusion

50This study shows that despite the growing convergence of mortality patterns in western Europe, there are still a number of major differences in age-, sex- and cause-specific mortality trends between the former FRG and France, some of which have only emerged in recent years.

51For both sexes, mortality among the elderly and the very old is lower in France, and the difference between the two countries is increasing with time. On the other hand, among adolescents and young adults (up to age 30 for women and age 50 for men), mortality is lower in the former FRG, and the difference has increased substantially over the period.

52Cardiovascular disease is the only cause which has evolved much less favourably in the former FRG over the period as a whole. This divergence reflects a general divide between northern and Mediterranean countries rather than a difference specific to the two countries. Its scale is nonetheless surprising, given the growing convergence in behaviour and lifestyle observed between the two countries (notably with regard to alcohol consumption and smoking) over the last half century.

53With regard to violent deaths and infectious diseases, the former FRG has a clear advantage. Since 1970, there was a sharper drop in mortality from the former cause in that country, and a resistance to decline or a larger increase from the latter cause in France. Shortcomings in the culture of risk aversion and anticipation among the French is likely to have played a major role. For suicides and road traffic mortality, the divergence between France and Germany has probably been increased by weaker social cohesion in France. The evolution of mortality from suicide — more closely correlated with that of unemployment in France than in the FRG — appears to testify to this difference.

54Lastly, the excess mortality from degenerative pathologies (cardiovascular diseases and cancers) of German women contrasts with a more similar situation for men in both countries. This can be attributed to the very low level of female mortality in France, but also reflects a more general pattern, i.e., a smaller sex differential in mortality in northern Europe than in southern Europe, itself due to more similar lifestyles. In particular, the difference between the sexes is smaller in the former FRG than in France for mortality associated with alcohol consumption and smoking, notably for cirrhosis and cancers of the upper aerodigestive tract, the oesophagus and the lung. Moreover, the rapid increase in alcohol consumption in the former FRG up to 1975 may well account for the interruption in the decline in infant mortality from 1968 to 1973.

55All in all, the differences in trends for both age- and cause-specific mortality balance each other out. The two countries have maintained similar levels for men, and France holds on to its advantage for women. More generally, the fact that sex-specific differences between the two countries have remained stable over time reflects the strong resistance of both cultures to the growing uniformity of living conditions in western Europe.