Memorandum by Emily Grundy, Professor
of Demographic Gerontology, Centre for Population Studies, London
School of Hygiene & Tropical Medicine
This Directive has implications for the insurance
and pensions industry who currently take into account the generally
longer life expectancy of women than men in pricing their products.
The evidence supplied here relates to possible changes in the
extent of gender differentials in mortality in the future.
1. GENDER DIFFERENCES
IN MORTALITY;
CURRENT PATTERNS
IN EUROPE
AND THE
USA
Currently women live longer than men in virtually
all populations, the exceptions being a few in which the status
of women is very low. However the extent of this differential
varies considerably. In Russia, Latvia and a number of other Eastern
European countries female life expectancy at birth exceeds that
of men by a decade, partly reflecting the heavy toll of excess
alcohol consumption among men (Velkoff and Kinsella 1993). The
gap between men and women in Southern Europe is less marked, possibly
because nutritional and lifestyle factors may protect men from
risks which some of their peers elsewhere are exposed to. Further
life expectancy at age 65 is also consistently higher for women
than for men, with women generally enjoying at least three or
four years of additional life. (see Figures 1 and 2).
2. COHORT INDICATORS
OF GENDER
DIFFERENCES IN
SURVIVAL TO
LATER LIFE
The period indicators shown in these Figures
are based on current rates of mortality. Examining the fate of
particular cohorts highlights the divergent destinies of boys
and girls born in the late 19th and early 20th centuries even
more strikingly. This is illustrated in Table 1, which shows survivorship
to age 80 for men and women born between 1861 and 1921 in England
and Wales. Of those born in 1861, a mere 10 per cent of boys and
16 per cent of girls lived to age 80. Among those born 50 years
later in 1911, 23 per cent of males and 41 per cent of females
reached this age while nearly half the women (but only 29 per
cent of the men) born in 1921 became octogenarians. The trends
underlying these variations by gender are complex and still not
fully understood. They include not only substantial improvements
in survival through infancy and adulthood, but also large changes
in the relative survival chances of adult men and women.
3. CHANGES IN
GENDER DIFFERENTIALS
IN LIFE
EXPECTANCY OVER
TIME
The tendency for mortality decline to be associated
with a greater gender divergence in life expectancy is illustrated
in Table 2. Between 1900-01 and the early 1970s the gap between
male and female life expectancy at birth increased from less than
one year to more than five years in Japan and from less than four
to over seven years in France. In both these countries this difference
increased further in the last part of the 20th century, although
in France (and Sweden) this peaked in the mid 1990s. In the UK
and the USA, by contrast, the sex differential in life expectancy
at birth has narrowed slightly since 1970-71. In both these countries
the gender difference in further life expectancy at age 65 was
also greatest in the early 1970s, but in the other countries shown
in Table 2 (and elsewhere) this gender difference has continued
to increase.
4. EXPLANATIONS
FOR GENDER
DIFFERENCES IN
MORTALITY
Reasons for the widening gap between male and
female mortality during most, or all, of the twentieth century
have been reviewed by Waldron (1986; 1993) and Verbrugge (1989)
and include, in the early phases of the mortality transition,
declines in causes of death specifically or predominantly affecting
women (such as maternal mortality, cancer of the uterus and respiratory
tuberculosis) and possibly changes in the intra household allocation
of resources by gender. Gender differences in health related behaviour,
particularly smoking, in exposure to occupational hazards and
perhaps the greater susceptibility of men to stresses associated
with socio-economic change have been suggested as important reasons
for the widening gap between male and female mortality for much
of the 20th century. In particular the 20th century epidemic of
coronary heart disease affected only men in most industrialised
countries. In England and Wales, for example, the sex ratio of
age standardised mortality from coronary heart disease remained
close to 1.5 during the 1920s, 30s and 40s but then rose rapidly
peaking at 3.5 in 1971 (Lawlor et al 2001). One hypothesised explanation,
apart from differences in stress and in smoking already referred
to, is possible gender variation in both consumption of red meat
("preferentially" allocated to the male breadwinner)
and biological response to dietary fat (Lawlor et al 2001).
Changes in the relative propensity of men and
women to smoke are undoubtedly an important factor in the recent
slight narrowing of the sex differential in mortality in England
and Wales and the USA (and some other western populations). Figure
3 shows the ratio of male to female death rates from neoplasms
for periods during the 20th century for groups aged 55-9 to 80
and over. This ratio rose until the mid century in younger groups,
until 1971-75 in the 70-74 year old age group and until the late
1980s in the 75-79 year old age group but then began to fall as
cohorts with less sex divergent smoking exposures reach the relevant
age group. This pattern is clearly indicative of an interaction
between gender and cohort effects. Similarly, Manton (2000) has
demonstrated that differences in male and female cohort mortality
rates underlie many of the period changes for heart disease, stroke
and lung cancer mortality evident in the USA between 1962 and
1995.
5. FUTURE PROSPECTS
IN THE
UK
As seen here, the extent of the gender gap in
life expectancy, both at birth and at age 65, has been diminishing
in recent decades in countries of the UK. It would seem very probable
that this trend will continue, reflecting the much smaller differences
between women and men in smoking behaviour in cohorts now entering
age groups most at risk of smoking related diseases; the reduction
in the proportion of men with health decrements consequent on
work related risks (eg coal mining; exposure to asbestos etc)
and the general lessening of differences in the lifestyles of
women and men. This assumption—of a continuation of the
narrowing of gender differentials immortality—is reflected
in official projections. However, it would seem very unlikely
that the female advantage will disappear, both because there appears
to be biological advantages associated with being female and because
there are reasons to suppose that female mortality may continue
to decline. Firstly the mortality of older women in the UK is
rather high in comparison with that of women in France and Italy,
for example, suggesting a room for improvement, and secondly current
cohorts of older women include large proportions with characteristics
generally associated with high mortality risk (eg low education).
June 2004
Table 1: Survival to age 80 by birth
cohort, England and Wales
|
| Year of birth | % surviving to age 80
|
|
| Men
| Women |
| 1861 | 10 |
16 |
| 1881 | 14 |
25 |
| 1901 | 17 |
34 |
| 1911 | 23 |
41 |
| 1921 | 29 |
47 |
|
| Source: Data from Government Actuary's Department.
|
Table 2: Trends in life expectancy at birth and
at age 65 by gender, selected developed countries
|
| Year and country
| Life expectancy (years)
|
| | At birth
| At age 65
|
| | M
| F | Difference
| M | F
| Difference |
|
| 1900-01 | England & Wales
| 44.8 | 48.7
| 3.9 | 10.1
| 11.1 | 1.1
|
| France |
43.2 | 46.9
| 3.8 | 10.0
| 10.9 | 0.9
|
| Japan |
44.0 | 44.9
| 0.9 | 10.1
| 11.4 | 1.2
|
| Sweden |
50.8 | 53.6
| 2.9 | 12.1
| 13.0 | 0.9
|
| USA | 46.4
| 49.0 | 2.6
| 11.4 | 12.0
| 0.7 |
| 1950-01 | England & Wales
| 65.3 | 70.3
| 5.0 | 10.8
| 13.4 | 2.6
|
| France |
63.4 | 69.2
| 5.8 | 12.2
| 14.6 | 2.4
|
| Japan |
57.6 | 60.9
| 3.3 | 10.9
| 13.0 | 2.1
|
| Sweden |
69.8 | 72.4
| 2.6 | 13.5
| 14.3 | 0.8
|
| USA | 65.6
| 71.1 | 5.5
| 12.8 | 15.1
| 2.3 |
| 1970-01 | England & Wales
| 68.8 | 75.0
| 6.2 | 11.9
| 15.8 | 3.9
|
| France |
68.4 | 75.8
| 7.4 | 13.0
| 16.8 | 3.7
|
| Japan |
69.3 | 74.7
| 5.4 | 12.5
| 15.4 | 2.8
|
| Sweden |
72.2 | 77.2
| 5.0 | 14.3
| 16.9 | 2.6
|
| USA | 67.2
| 74.9 | 7.7
| 13.1 | 17.1
| 4.0 |
| 1995 | England & Wales
| 74.4 | 79.6
| 5.2 | 14.8
| 18.4 | 3.6
|
| France |
73.9 | 81.9
| 8.0 | 16.1
| 20.6 | 4.5
|
| Japan |
76.4 | 82.8
| 6.4 | 16.5
| 20.9 | 4.4
|
| Sweden |
76.2 | 81.5
| 5.3 | 16.0
| 19.7 | 3.7
|
| USA | 72.4
| 79.3 | 6.9
| 15.3 | 19.2
| 3.8 |
| 2000-2/5 | England & Wales
| 75.9 | 80.6
| 4.7 | 16.0
| 19.1 | 3.1
|
| France |
75.2 | 82.8
| 7.6 | 16.3
| 21.2 | 4.9
|
| Japan |
77.8 | 85.0
| 7.2 | 17.4
| 22.6 | 5.2
|
| Sweden |
77.6 | 82.6
| 5.0 | 16.7
| 20.5 | 3.8
|
| USA | 74.6
| 80.4 | 5.8
| 15.6 | 19.2
| 3.6 |
|
Sources: Data from Government Actuary's Department (E&W),
Berkeley Mortality Data Base (http://demog.berkeley.edu/wilmoth/mortality);
Ministry of Health and Welfare (Japan), Statistics and Information
Department, 18th life tables, Tokyo, 1998; United Nations 2002.
Figure 1: Sex differentials in life expectancy at birth,
selected European countries, 1995-2000
REFERENCES:
Lawlor, D A, Ebrahim, S and Davey Smith, G (2001) "Sex
matters: secular and geographical trends in sex differences in
coronary heart disease mortality". British Medical Journal,
323, 541-545.
Manton, K G (2000). "Gender differences in the cross-sectional
and cohort age dependence of cause-specific mortality: the United
States, 1962 to 1995". Journal of Gender Specific Medicine,
3, 47-54.
Velkoff, V and Kinsella, K (1993), Aging in Eastern Europe
and the Former Soviet Union, US Bureau of the Census, Washington
DC.
Verbrugge, L M (1979) "Marital status and health".
Journal of Marriage and the Family, 41, 267-285.
Verbrugge, L M (1989) "Gender, aging and health",
in K S Markides (ed), Aging and health, Sage Publications,
Newbury Park (Ca), pp 23-78.
Waldron, I (1986) "What do we know about sex differences
in mortality?" Population Bulletin, no 18. New York:
United Nations.
Waldron, I (1993) "Recent trends in sex mortality ratios
for adults in developed, countries". Social Science &
Medicine, 36, 451-62.
|
