Health
statistics 01
Spring 1999
Quarterly
IN THIS ISSUE
Page
In brief
2
Health indicators
4
Socio-economic differentials in health: illustrations from the Office for National
Statistics Longitudinal Study
Reports the latest findings on social differences in health using the ONS Longitudinal Study
Seeromanie Harding, Joanna Brown, Michael Rosato and Lin Hattersley
5
Prescribing for patients with asthma by general practitioners in England and
Wales, 1994–96
Studies trends in the management of asthma in general practice, particularly in the use of
inhaled steroids and inhaled bronchodilators
Azeem Majeed and Kath Moser
16
Death certification and the epidemiologist
Describes the various processes by which deaths are certified in England and Wales and
illustrates the importance of accurately recording causes of deaths
Tim Devis and Cleo Rooney
21
Trends in mortality of young adults aged 15 to 44 in England and Wales
Updates a previous analysis investigating the reasons for the levelling of mortality in young
adults since the mid-1980s
Paul Aylin, Karen Dunnell and Frances Drever
34
Weekly deaths in England and Wales
Describes the method used by ONS to estimate weekly death registrations between 1995 and
1998 and the alternative methods investigated during 1996–98
Sue Kelly and Helen Lawes
40
Annual update: 1997 Mortality statistics: Cause (England and Wales)
44
Tables
List of tables
Tables 1.1–6.3
Notes to tables
47
48
70
Recent ONS publications
71
Government Office Regions and counties in England
71
London: The Stationery Office
A publication of the
Government
Statistical Service
Health Statistics Quar terly 01
Spring 1999
in brief
1991 Censuses with information from birth
and death registration through to the late 1990s
for a 1 per cent sample of the population.
Launched
In autumn 1997 ONS announced that it was
conducting a review of the case for funding the
addition of 2001 Census data and subsequent
events up to 2011. The review was conducted
by a team comprising Richard Bartholomew
(Department for Education and Employment),
Professor Angela Dale (Manchester
University), John Pullinger (ONS), Dr Gabriel
Scally (NHS Executive), Tim Skinner
(Australian Bureau of Statistics), and Marian
Storkey (London Research Centre), with
support from Jillian Smith (ONS) and
Professor Heather Joshi (City University).
Welcome to the first issue of Health
Statistics Quarterly. It has taken a year to
plan this new publication – starting with
consultation with customers, then
planning the programme of articles,
revising the tables, agreeing a timetable
for the inclusion of monitor material
throughout 1999 and, not least, ensuring
that this publication and Population Trends
form a coherent and comprehensive
package of the work we do in ONS on
demography and health.
I trust that customers, old and new, will find Health Statistics Quarterly informative and
useful in their work. We intend to be dynamic, reflecting our customers’ needs, and any
comments and contributions on our new quarterly reporting strategy will be most
welcome.
Karen Dunnell
Population and Health Monitors
A key part of the development of our
population and health publications is to
discontinue the Population and Health Monitor
Series and to incorporate the information in
Health Statistics Quarterly and Population
Trends. Between them these two publications
will contain most of the data previously issued
in Population and Health Monitors but with the
new title of ‘Reports’. Alternative
arrangements are being made for the local
population estimates and electoral statistics.
Data for these will be published separately in
order to meet required publication dates.
All abortions monitor data are replaced by a
regular table (4.2) in Health Statistics
Quarterly. There will also be a regular graph
showing quarterly abortions rates for residents
of England and Wales from 1981 onwards.
Office for National Statistics
2
Figures for the September quarter of 1998
show a provisional abortion rate of 13.7 per
1,000 women aged 14–49. This is an increase
of 4 per cent compared with the same quarter
in 1997. A rise is shown in all age groups
except the 45 and over group. Since the March
quarter of 1996 there has been an upward trend
in the abortion rate as shown by Figure D on
page 4.
2001 Census information to be added to
the ONS Longitudinal Study
Following a review, plans are now being made
to include information from the 2001 Census
and subsequent events in the ONS
Longitudinal Study.
The ONS Longitudinal Study has brought
together information from the 1971, 1981 and
The review team consulted widely and
received many detailed comments and
thoughtful submissions. It reported in June
1998 and recommended that a strong case
exists for adding information from the 2001
Census and subsequent events up to 2011.
ONS has accepted these recommendations.
The review team’s report will be published on
22 February. The executive summary appears
below.
Executive Summary of the 1998 Review
of the ONS Longitudinal Study
Every ten years, prior to the decennial census
of population, a review of the Office for
National Statistics Longitudinal Study (LS) is
completed to make the case for linking the next
LS/Census sample into the study and for the
next decade’s data capture of vital event
information. A review board of six people, five
of whom are external to ONS, with a range of
relevant experience, was convened to assess
the LS in the context of a wider strategy for
longitudinal data, to review the framework in
which the LS has operated and past and
planned uses up to 2001, to consider what
would be gained by adding the 2001 Census
and subsequent event data to 2011 and to make
recommendations.
Health Statistics Quar terly 01
In the course of the review, comments on the
study were invited from many organisations
and individuals, including government
departments, research councils, centres and
societies, independent foundations and
institutes, regional health authorities and a
wide range of social statistics interests in the
GSS. Comments were also invited through
newsletters, user groups and committees.
The Longitudinal Study (LS) is a dynamic
document of the social fabric of England and
Wales. Its combination of health, geography,
household and occupational information has
made it a valuable tool to monitor a broadly
conceived policy of public health as well as to
support a range of other policy areas from
health inequalities to housing. Research
commissions have come from a large number
of academic institutions as well as the Health
and Safety Executive, Department of Health,
Inland Revenue, Department of the
Environment, Transport and the Regions, the
Lord Chancellor’s Department, Department
for Education and Employment, Welsh Office,
Higher Education Funding Council and the
Arts Council and from within ONS itself.
Quality is an asset in the LS due to the design
of the study and the care with which it is
maintained. The quality checking completed in
preparation of the LS provides a significant
saving in terms of the skilled weighting and
imputation work which would otherwise be
needed. While it is important not to be
complacent about possible biases, the overall
high representativeness of the data greatly
enhances robustness and confidence in
analysis. The study is often used as a validator
for new classifications and other data sets. It
also provides a focus for skills and methods for
the management and analysis of complex
longitudinal data.
A range of other countries have linked datasets
which are similar to some aspects of the
England and Wales LS. These include
Denmark, Finland, France, Israel, the
Netherlands, Norway, Sweden and the USA.
The Scandinavian registration systems give
these countries an opportunity to produce 100
per cent linkage of data sources for analysis.
Clearly other countries find linked datasets
safe and worthy of investment. However very
few countries maintain a supported,
dynamically linked research database suitable
for a range of analyses with the level of access
offered by the LS. Other countries, such as
Australia, New Zealand, Scotland and
Northern Ireland are looking at the
opportunities of initiating something with
features akin to the LS.
The LS is surrounded by a very comprehensive
set of security measures to provide a safe
setting. It has also had the benefit of an aware
guardianship, sensitive to both real and
perceived confidentiality issues. Nevertheless
the evolution of the access arrangements
shows where possible opportunities to help
users have been taken. This should continue,
but the need for maintenance of confidentiality
must remain paramount.
The main options for potential new
developments of the LS database are
increasing access, increasing the sample size
and additional linkage. Of these the first raises
issues of confidentiality which must be
considered before any change, the second is
very costly and the third has significant policy
interest but must be approached cautiously to
ensure that public confidence and privacy are
safeguarded.
Costing estimates are at an early stage, but the
possibility of making the 2001 LS/Census link
through automatic matching of records, would
seem to offer substantial efficiency savings. We
believe that such an approach would be
feasible and would not compromise
confidentiality. The research value of the
timely arrival of a completed, 2001 inclusive
LS database means that business as usual data
capture, quality maintenance and
documentation should be maintained through
the link period.
The steady state management of the LS
occupies twenty staff. The activities relate to
seven different functions across three ONS
locations (London, Southport and Titchfield).
Given the nature of the work it would not be
cost effective to co-locate these activities and
their separation has always been regarded as
an additional confidentiality assurance. The
use of record linkage is a cost-effective way of
providing the information in the LS database.
If it were to be gathered by bespoke surveys,
the costs would have to be augmented by the
costs of fieldwork and following half a million
individuals and it would take thirty years to
build to the current time span.
Spring 1999
will be undertaken to establish the level of
these potential savings.
We conclude that the LS is a powerful and
flexible database which fills a unique position
in the range of social and demographic
sources for statistics in the UK. Though
confidentiality restrictions are comprehensive
they are necessary and have not prevented a
wide range of organisations, including
government departments, from making
extensive use of the data. Arrangements for
management and processing of the LS are
satisfactory, but some efficiency developments
are possible, especially automatic linkage of
the 2001 LS/Census sample, if confidentiality
considerations allow. Such a link would be of
great value as would the next ten years of event
data capture into the study. We recommend that
these take place and that the LS continues to
develop its access arrangements and further
considers linkage to other data sources.
The full report, Report of the ONS
Longitudinal Study Review, 1998, is available
from:
Sales Office
Room B1/6
Office for National Statistics
1 Drummond Gate
London SW1V 2QQ
Tel: 0171-533 5678
Opportunities to make efficiency savings have
been taken over the years and remaining
labour intensive tasks are important in
maintaining the quality of the LS data. Further
efficiency savings which are feasible are in the
area of automation of event processes and,
particularly, in automating the 2001 LS/
Census link. These should be thoroughly
explored. In line with ONS policy the LS
database is provided free to the point of
analysis. From this point on a general policy
of recovery of full economic cost should be
applied. For the LS this would apply to
academic computing costs.
The yearly cost of the LS, including linkage of
census data, is £850,000. Early estimates
indicate that automation might reduce this
amount to £700,000 per year. Further work
3
Office for National Statistics
Health Statistics Quar terly 01
Spring 1999
Health indicators
Figure A
England and Wales
Population change (mid-year to mid-year)
Thousands
200
100
0
Net migration and
other changes
Natural change
Total change
8
–9
7
97
6
–9
5
–9
96
95
4
–9
94
–9
3
93
2
–9
92
1
–9
91
0
–9
90
9
–9
89
8
–8
88
7
–8
87
6
–8
86
5
–8
–8
84
85
4
3
–8
2
–8
83
82
1
–8
0
–8
81
80
9
–8
79
8
–7
78
7
–7
77
6
–7
76
5
75
–7
–7
74
3
–7
–7
73
72
2
–7
71
19
4
-100
Mid-year
Figure B
Age-standardised mortality rate
Rate per million population
20,000
15,000
10,000
5,000
0
1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997
Year
Figure C
Infant mortality (under 1 year)
Rate per 1,000 live births
20
15
10
5
0
1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997
Year
Figure D
Quarterly abortion rates – residents
Rate per 1,000 women 14–49
15
14
13
12
11
10
Abortion rate
9
8
1981
1982
1983
1984
1985
Office for National Statistics
1986
4
1987
1988
1989
1990
Year
Moving average rate
1991
1992
1993
1994
Provisional rate
1995
1996
1997
1998
Socio-economic differentials
in health: illustrations from the
Office for National Statistics
Longitudinal Study
Seeromanie Harding,
Joanna Brown, Michael Rosato
and Lin Hattersley,
ONS
BACKGROUND
Monitoring the health of the
nation is central to the
government’s initiative to reduce
inequalities in health. Using the
ONS Longitudinal Study, the
latest findings on health
inequalities at various stages of
the life course are reported.
Some of the areas covered in
this study are: trends in class
mortality and the influence of
unemployment; the impact of
social deprivation in childhood
and young adulthood on health
in later life; and the health of
children born to Irish migrants
in England and Wales.
The advantage of studies using longitudinal rather than cross-sectional
data to examine the impact of social deprivation on health is that by
tracking the same individuals over time, accurate monitoring of health
differentials can be made. In the recent Government Green Paper, Our
Healthier Nation: a contract for health, 1 Longitudinal Study research
was used to show that social gradients in mortality widened during the
1980s.2, 3, 4, 5, 6 Studies have also shown the importance of early life
factors on adult health. The notion of accumulation of risk of adverse
events over time has direct relevance to policies that aim to arrest or
reverse social deprivation. As part of a study of the impact of social
exclusion on health in adult life, children living in children’s homes and
young men in places of detention were followed up for 25 years.7
Another study of the health of children born to Irish migrants examined
the effects of socio-economic position on health. 8, 9 This paper has a
dual purpose of presenting updated findings from these studies and
demonstrating the usefulness of longitudinal data in monitoring the
health of the nation.
METHOD
The Longitudinal Study is a record linkage study that contains
information on individuals and households from censuses and routinely
registered vital events. Details of the study can be found elsewhere. 10 It
began in 1971 and is based on a representative 1 per cent sample (about
550,000 people) of the population of England and Wales.
There is a range of information from the censuses that can be used to
classify people into different social groups. In the following analyses,
established indicators of socio-economic status such as occupation
based social class, housing tenure and access to cars were used.6, 11
5
Office f or National Statistics
Health Statistics Quar terl y 01
Spring 1999
Three outcome measures were used: death from all causes during the
period 1976–95, the incidence of major cancers 1976–90, and the
prevalence of limiting long-term illness in 1991. Most of the analyses
presented in this paper were based on individuals present at the 1971
Census, the cohort with the longest follow-up (Box One). As is now
conventional for occupation-based social class analysis of longitudinal
data, the first five years of follow-up (1971–75) were excluded from the
analysis to allow for the effects of health selection (Box One).12
A range of statistical approaches was used. Death and cancer incidence
rates were standardised by the direct method, using all Longitudinal
Study women and men as standard populations. Life tables were used to
derive life expectancy at birth for each social class. A summary
measure of class differences, the index of inequality 13 (Box One), was
calculated to avoid the problem of comparing socio-economic
categories of different sizes. Smaller categories at the extremes are
likely to lead to larger ratios when comparing the bottom with the top
of the hierarchy. This indicator of social position was then related to
death rates and cancer incidence rates using Cox regression analysis.
Cox regression models were also used to examine joint influences, such
as Irish origin and social class, on health outcomes.
MONITORING TRENDS IN OCCUPATION- BASED SOCIAL
Box one
CLASS DIFFERENTIALS IN MORTALITY
THE 1971 AND 1981 COHORTS
The 1971 Cohort comprises all Longitudinal Study (LS)
members present at the 1971 Census and traced in the
National Health Service Central Register (NHSCR). LS
members were classified by their socio-economic
details at the 1971 Census and, for those who sur vived
the first ten years, at the 1981 Census and followed-up
to the end of 1995. The 1981 Cohort comprises all LS
members present at the 1981 Census and traced in the
NHSCR.
HEALTH SELECTION
In a longitudinal study, health selection may affect
mortality differentials in the early period of follow-up.
Health selection refers to the process whereby
individuals are included or excluded from a group
because of their health status. Employed individuals are
generally assigned to a social class. They also tend to be
healthier than those out of work. The effects of health
selection on mortality differences by social class reduce
with increased follow-up. Most of the effect disappears
after the first five years.
THE INDEX OF INEQUALITY
Comparisons of rate ratios between socio-economic
categories are affected by the difference in sizes of the
groups. Smaller groups at the extremes are likely to
lead to larger ratios when comparing the bottom with
the top of the hierarchy. The index of inequality was
constructed to avoid this problem as adjustments were
made for the different group sizes. 13 Social class, based
on all six levels within each five-year age-band was
assigned a value between zero and one according to
the proportion of subjects above the midpoint of each
class. For example, men aged 40–44 in social class I
comprised 6 per cent and were assigned a value of 0.03,
which was the proportion of subjects above the
midpoint of that class; those in the next class
comprised an additional 21 per cent and were assigned
a value of 0.06 + (0.21 / 2) = 0.17. This indicator of
social position was then related to mortality rates
using Cox regression. The relative risks were adjusted
for age at entry to study. A value of 1.00 indicates
equality between the groups. Above 1.00 represents
higher mortality among the less advantaged.
Office for National Statistics
6
It is increasingly recognised that longitudinal data is required for
accurate measurement of socio-economic differentials. Issues related to
numerator/denominator biases and mobility between classes can
obscure real trends when cross-sectional data is used. There has been
consistent evidence of social class patterns in mortality in the UK.
Recently the emphasis has been on measuring trends over time, the
influence of other factors, such as employment status, and gender
differences in these differentials.
Examining mortality of women by social class is usually problematic in
England and Wales because of intermittent labour market participation.
Details of partner’s occupation from the censuses were linked to
members’ records so that women who could not be classified to an
occupation were classified by their partner’s. When this method was
used, the proportion of classified women increased from about half to
three-quarters.
Figure 1 shows trends in life expectancy at birth by social class for
women and men. Children were classified by parental social class.
Among women, life expectancy at birth varied systematically with
social class in the 1980s. In the 1990s, life expectancy of women in
professional or managerial jobs was 81 compared with 78 years for
women in partly skilled or unskilled jobs. Between the late 1970s and
early 1990s, life expectancy increased for all classes but the rate of
increase differed by class. This accounts for the narrowing of class
differences between the late 1970s and early 1980s followed by an
increase. Between the 1970s and early 1990s, the largest gain in life
expectancy was made by women in professional and managerial jobs
(three years) and the smallest by those in partly skilled and unskilled
jobs (two years).
Among men, life expectancy varied systematically over the entire
period of follow-up. In the 1990s, life expectancy at birth of men in
professional or managerial jobs was 76 compared with 71 years for men
in partly skilled or unskilled jobs. Between the 1970s and early 1990s,
the largest gain in life expectancy was made by men in non-manual
skilled jobs (four years) and the smallest by those in partly skilled and
unskilled jobs (three years).
The calculation of life expectancy included individuals who joined the
study after it began in 1971. Changes in the classification of
occupations to a social class over time could have had some influence
on these trends. In the following analysis (Figures 2 and 3), only men
and women who were present at the start of the study were included in
the analysis, and as they were classified to a social class at the start of
study, only one classification of occupations was used. Occupation
details were recorded for those aged 15 and over but those aged 15–34
were censored out of the analyses to avoid the bias from the ageing of
the cohort over time. Those aged 65 and over were excluded because of
the poorer quality of occupation information at post retirement ages.
Health Statistics Quar terly 01
Figure 1
Trends in life expectancy at birth by occupation-based social class, England and Wales
Women
Years
82
Years
82
80
80
78
78
76
76
74
74
72
Manual skilled
Non-manual
skilled
Partly skilled
and Unskilled
Manual skilled
Non-manual
skilled
Partly skilled
and Unskilled
68
66
66
1977–81
1982–86
1987–91
1992–95
Using the index of inequality, trends similar to those observed with life
expectancy were seen for women (Figure 2). Figure 3 shows that death
rates fell in each class grouping. The narrowing of relative differences
in the early 1980s was due to a larger fall in the death rates in manual
than in the non-manual class grouping. This was followed by larger
falls in death rates in the non-manual classes, which resulted in a
widening of the class differential.
Among men there was a trend of increasing relative class differences
over the last two decades (Figure 2). The increase in class differences
was progressive compared with those from the life expectancy
approach and this was probably due to the differences in approach
Figure 2
Trends in occupation-based social class
differences in mortality among women and
men, aged 35–64: relative risk based on index
of inequality. 1971 cohort, England and Wales
1977–81
1987–91
1992–95
The influence of employment status on class
differentials in mortality
The relationship between unemployment and ill health is well
established. 5, 14, 15, 16 Recent work has shown that people experiencing
more than one period of unemployment have higher mortality than
those in continuous employment.15 Unemployment rose between the
1970s and 1980s and in this section we examine the impact of
unemployment on class mortality.
Trends in death rates by occupation-based
social class among women and men aged
35–64. 1971 cohort. England and Wales
Death rates/100,000 persons
4500
Women
Men
2.8
4000
2.6
3500
2.4
B
2.2
B
1982–86
mentioned earlier. The increase in class differentials was due to the
greater fall in the death rates of men in professional/managerial jobs
than in that of those who were in partly/unskilled jobs (Figure 3).
Figure 3
Relative risk
3.0
1.6
Professional and
managerial
70
68
1.8
Men
72
Professional and
managerial
70
2.0
Spring 1999
B
B
B
B
2500
2000
B
1500
B
1.4
3000
1000
500
1.2
1.0
1976–81c+
1981c+–85
1986–91c+
1991c+–95
+1981/91 refers to day before census day in the first time period and census day
onwards in the following time period.
0
1976–81c+
Male manual
Male non-manual
Female manual
Female non-manual
1981c+–85
1986–91c+
1991c+–95
+ 1981/91 refers to day before census day in the first time period and census day onwards
in the following time period.
7
Office f or National Statistics
Health Statistics Quar terl y 01
Table 1
Spring 1999
Trends in death rates/100,000 persons by
employment status and social class among men
aged 36–64. 1971 cohort, England and Wales
Social class
Follow-up period
1976–81c +
1981c+–85
1986–91c +
1991c +–95
Employed
Non-manual
Manual
3361
4105
2518
3281
2178
3281
1642
2455
Unemployed
Non-manual
Manual
4689
5917
3980
4505
2500
6059
2889
4113
+ 1981/91 refers to day before census day in the first time period and census day onwards
in the following time period.
Table 2
Relative risk of mortality 1986–95, of men aged
35–64 by employment status, in 1971 and 1981.
1971 cohort, England and Wales
Employment status in 1971 and 1981
Relative risk
Employed 1971 and 1981
Employed 1971, unemployed 1981
Employed 1981, unemployed 1971
Unemployed 1971 and 1981
1.00
1.63*
1.66*
2.63*
* p<0.05.
Table 3
Social class differentials adjusted for employment
status among men aged 35–64: relative risk,
1986–95. 1971 cohort, England and Wales
1971†
Employment status
Social class
I/II
IIIN
IIIM
IV/V
1981**
1.32*
1.17*
1.00
1.01
1.42
1.64*
1.00
1.30*
1.28*
1.54*
Social class and employment status differentials in
mortality among women aged 16–59 at the 1981
census: relative risk, 1986–95. 1981 cohort,
England and Wales
Adjusted for age only
Employment status^
Employed
Unemployed
Social class~
I/II
IIIN
IIIM
IV/V
*
^
~
†
Too few women were unemployed in 1971 (2 per cent) to allow a
similar study of trends. In 1981, 4 per cent of women were unemployed
and this increase allowed some examination of the influence of
unemployment and social class on mortality of women. Women present
at the 1981 Census, the 1981 cohort (Box One), aged 20–59 in 1986–
95, were classified by their employment status at the 1981 Census.
Table 4 shows that mortality was 45 per cent higher among women who
were unemployed than employed and after adjusting for class it was 32
per cent higher. Consistent class gradients were also evident before and
after adjustment for employment status. Adjusted for employment
status mortality of women in Social Class IV/V was 39 per cent higher
than that of those in Social Class I/II. As with men, both employment
and social class remained important predictors of the mortality of
women.
Mortality gives a limited measure of the prevalence of disease in a
society and in the next section we examine social differentials in the
level of morbidity from three main cancers.
MONITORING SOCIO- ECONOMIC DIFFERENTIALS IN THE
INCIDENCE OF AND SURVIVAL FROM BREAST, OVARIAN AND
PROSTATE CANCERS
* p<0.05.
† Employment status and social class in 1971.
** Employment status and social class in 1981.
Table 4
Table 1 shows that over the last two decades mortality was consistently
higher among men in a manual than in a non-manual class regardless of
employment status. In every time period, mortality was higher in each
class grouping among those who were unemployed than those
employed. Table 2 shows that men who were employed at both
censuses had the lowest mortality and those who were unemployed at
both censuses the highest. The higher mortality of the unemployed,
however, could not explain class differences. Table 3 shows that for the
most recent period of follow-up, 1986–95, large class differences
persist after adjusting for employment status. Although unemployment
increased between 1971 and 1981, those unemployed contributed less
than 10 per cent of all deaths, so that though unemployment is an
important predictor of mortality, it could not have been the main
attributable factor for the trends in class differentials.
Adjusted for social class and
employment status†
1.00
1.45*
1.00
1.32*
1.00
1.05
1.34*
1.40*
1.00
1.05
1.32*
1.39*
p<0.05.
Own employment status.
Based on own occupation, then partner’s.
Women who were classified by both employment status and social class were included in
this model.
Office for National Statistics
8
Currently, there is a great emphasis on reducing deaths from cancers
through prevention and secondary intervention strategies. Monitoring
trends in social differentials in incidence and survival is important for
the planning and provision of both preventative and curative services in
public health, and also for aetiological reasons. In a longitudinal study,
length of survival after the registration of a cancer can be measured.17
Survival rates are more accurate than death rates as differences in death
rates may be biased by differences in the length of survival from
diagnosis to death. In the ONS Longitudinal Study, the cause of death is
known so that assumptions do not need to be made about whether death
was from the primary tumour or another cause. In the following
analyses, women present at the start of study in 1971 were followed up
until 1990 for cancer incidence and until 1995 for survival. The
difference in follow-up was because cancer registrations lag behind
death registrations. Housing tenure at the start of study was used as a
measure of socio-economic status; owner occupied housing being a
more advantaged status than rented housing.
Breast and ovarian cancers
From earlier studies we know that breast cancer incidence was higher in
women in a higher socio-economic status.18 This was still evident
among older women (Table 5). At younger ages, however, the pattern of
breast cancer incidence appeared to be changing. In 1986–90, there was
little difference in incidence rates among those in owner occupied and
rented housing. This was due to a larger rise in incidence among those
in rented housing. Similar trends were seen with occupational social
Health Statistics Quar terly 01
Table 5
Age in follow-up
period
Housing tenure
Trends in socio-economic differentials in breast
cancer incidence, by age 1976–90, incidence rates/
100,000 persons and percentage change. 1971
cohort, England and Wales
1976–81c+
1981c+–85
Percentage
change
30–64 years
Owner occupied
Rented
Rate ratio
572
543
0.95
508
378
0.74
-11
-30
65 and over
Owner occupied
Rented
Rate ratio
860
792
0.92
869
597
0.87
-20
-25
1986–90
Percentage
change
523
546
1.04
3
44
705
484
0.69
2
-19
Table 6
Housing
tenure
Owner occupied
Rented
Rate ratio
Spring 1999
Trends in socio-economic differentials in ovarian
cancer incidence among women aged 30 and over,
1976–90: incidence rates/100,000 persons and
percentage change. 1971 cohort, England and Wales
1976–81c+
1981c +–85
134
118
0.88
112
126
1.13
Percentage
change
-16
7
1986–90
Percentage
change
131
86
0.66
+ 1981 refers to day before census day in the first time period and census day onwards in
the following time period.
+ 1981 refers to day before census day in the first time period and census day onwards in the
following time period.
Figure 4a
Socio-economic differentials in survival from breast cancer among women by age at registration, 1971 cohort,
England and Wales
30–64years
65 years and over
Sur vival probability
Survival probability
Years of follow-up
Figure 4b
Years of follow-up
Socio-economic differentials in survival from ovarian cancer among women by age at registration, 1971 cohort,
England and Wales
65 years and over
30–64years
Sur vival probability
Survival probability
Years of follow-up
17
-32
Years of follow-up
9
Office f or National Statistics
Health Statistics Quar terl y 01
Spring 1999
class.19 Recent studies in the US have also shown a change in the
direction of the class differential at younger ages.20 The aetiology of
breast cancer is still largely unknown but such international trends
support the influence of environmental factors. The change in childbearing patterns may have contributed to these changes. Later childbearing was established earlier among women in non-manual classes
than among those in manual classes.21
Figure 4a shows survival from breast cancer at ages 30–64 and 65 and
over at registration. In the younger age group, survival was poorer
among those in rented compared with those in owner occupied housing.
Median survival time was about 10 years among those in rented
housing compared with about 14 years for those in owner occupied
housing. From other work we know that although stage at presentation
influences these survival differences, issues related to access to
healthcare is the more important predictor.22 At ages 65 and over,
differences in survival were not apparent.
Among women aged 30 and over, the direction of the housing
differential for ovarian cancer was not consistent over all three time
periods (Table 6). In 1986–90, incidence was higher in women in owner
occupied housing but in 1981–85, it was higher in tenants. This pattern
was seen at both younger (30–64 years) and older ages (not shown).
The striking feature of Figure 4b is the poor survival (under two years)
from this cancer in both housing categories. At younger ages, survival
was also significantly poorer among those in rented than in owner
occupied housing.
Prostate cancer
Table 7 shows that the incidence of prostate cancer remained higher
among men who were in owner occupied housing over the 20 years of
follow-up period. This pattern was consistent by age (not shown).
Among those aged 30–64 at registration, survival appeared poorer
among those in rented than owner occupied housing but the difference
was not statistically significant (Figure 5). Median survival time for
those in owner occupied housing was about six years compared with
four for those in rented housing. At older ages, there were no differences in survival, median survival being about three years for both
those in rented housing and those in owner occupied housing. The
prognosis of prostate cancer is good if detected early and the poor
survival in both of these age groups highlights the need to both promote
awareness of risks and to ensure equitable access to services.
Figure 5
Table 7
Trends in socio-economic differentials in prostate
cancer incidence among men aged 30 and over,
1976–90: incidence rates/100,000 persons and
percentage change. 1971 cohort, England and
Wales
Age in
1976–81c+
follow-up period
Housing tenure
Owner occupied
Rented
Rate ratio
311
303
0.97
1981c+–85
Percentage
change
1986–90
324
266
0.82
4
-12
332
277
0.83
MONITORING THE HEALTH OF SOCIALLY EXCLUDED
GROUPS
Children living in children’s homes, young men in
places of detention and in psychiatric homes
In a longitudinal study, the influence of factors in early life can be
examined in relation to adult health and socio-economic status.
Accumulation of factors over the life course has direct relevance to
intervention strategies, which aim to arrest or reverse the impact of
social deprivation. The following analyses illustrate the use of the
Longitudinal Study in the examination of the impact of social exclusion
on life chances.7
Children aged under 18 living in children’s homes, and men aged 15–29
years in places of detention and in psychiatric institutions, were
identified at the 1971 Census and their subsequent social position and
health over 25 years was examined. Their socio-economic position at
subsequent censuses was measured using a range of indices. The health
outcome measures used were death from any cause and the prevalence
of a limiting long-term illness 20 years after the start of follow-up.
65 years and over
30–64years
Sur vival probability
Years of follow-up
Office for National Statistics
10
2
4
+ 1981 refers to day before census day in the first time period and census day onwards in the
following time period.
Socio-economic differentials in survival from prostate cancer among men by age at registration. 1971 cohort,
England and Wales
Survival probability
Percentage
change
Years of follow-up
Health Statistics Quar terly 01
Figure 6
Spring 1999
Children aged under 18 in homes, and men aged 15–29 in places of detention and in psychiatric institutions in 1971:
socio-economic status in 1981 and 1991. 1971 cohort, England and Wales
(a) Children, aged under 18 years, living in children’s homes in the 1971 Census
Percentage
100
1981
80
1991
60
40
20
0
car access
no car access
NPH*
non-manual
manual
unclassified
manual
unclassified
All other children, aged under 18 years, in the 1971 Census
Percentage
100
80
60
40
20
0
car access
no car access
NPH*
non-manual
(b) Men, aged 15–19, in places of detention in the 1971 Census
Percentage
100
80
60
40
20
0
car access
no car access
non-manual
manual
unclassified
employed
unemployed
employed
unemployed
Men, aged 15–29 years, living in psychiatric homes in the 1971 Census
Percentage
100
80
60
40
20
0
car access
Percentage
no car access
non-manual
manual
unclassified
All other men, aged 15–29 years, in the 1971 Census
100
80
60
40
20
0
car access
no car access
* NPH refers to non-private households.
non-manual
manual
employed
11
unemployed
Office f or National Statistics
Health Statistics Quar terl y 01
Spring 1999
The high loss to follow-up of these two groups, about a quarter,
highlighted the difficulties in researching socially deprived groups.
Among those who remained in the study, the picture was clear. Figure 6
shows that both of these groups moved into and remained at a lower
socio-economic status than all other comparable study members. For
example, 16 per cent of those who were in children’s homes in 1971
were in a non-manual class in 1981 and 38 per cent in 1991. This
compared with 37 per cent and 56 per cent for all other children.
Among young men who were in places of detention in 1971, 43 per
cent were unemployed in 1981 and 38 per cent in 1991. This compared
with 7 per cent and 6 per cent for all other LS members of the same ages.
Table 8 shows that the risk of morbidity and of premature mortality among these groups was considerably higher than that of all
others of comparable ages. Prevalence of a limiting long-term
illness was more than three times higher; mortality risk more than
twice among children in homes and men in places in detention, and
men who were in psychiatric institutions, six times higher.
Table 8
Health
outcome
Mortality
Relative risk
Teenage conception rates are higher in the UK than the rest of Western
Europe and the findings from these studies imply that strategies to
lower these rates are likely to bode well for teenagers and their
children.
MONITORING THE HEALTH OF CHILDREN BORN TO
MIGRANTS – MORTALITY AND CANCER INCIDENCE AMONG
CHILDREN BORN TO IRISH MIGRANTS
There is a dearth of data on the health of migrants and their children.
Studies have examined mortality patterns of first generation migrants
living in England and Wales, and recently the emphasis has been on
examining the influence of socio-economic status on these patterns.3, 8, 9, 28, 29
Over the last two decades, death rates of people born in Ireland and
living in England and Wales have been higher than those of all people
living in England and Wales. Their death rates were also higher than the
rates of those remaining in Ireland.30 Some of this excess mortality was
attributed to the selective migration of those who were disadvantaged in
health and socio-economic circumstances. This study examined
whether this mortality disadvantage persisted in the children born to
Irish migrants (second generation Irish) and living in England and
Wales.8, 9
Office for National Statistics
12
1.00
Children
in homes
All
other men
Men in
detention
Men in
psychiatric
institutions
15
85
172
7
93
42,173
26
74
147
27
73
95
2.50*
1.00
2.83*
6.10*
* p<0.05.
Family structure and social background of those who
became teenage mothers
Table 9 shows clear associations between indicators of both parental
socio-economic position and family structure and the risk of becoming
a teenage mother in the 1980s. Furthermore, by both indicators of
socio-economic position, there were stepwise increases in becoming a
teenage mother. Family structure was also an important predictor.
Daughters of teenage mothers were more likely to become teenage
mothers themselves. After adjustment for the associations between
family structure and socio-economic position, absence of a normally
resident father was no longer a statistically significant factor.
All other
children
Limiting long-term
illness
Yes
4
No
96
All=100%
112,325
Table 9
Teenage conception rates have become the focus of recent policy
initiatives on social exclusion and gender-related issues.23 Studies using
other longitudinal data on teenagers in the 1960s and 1970s showed that
lower parental socio-economic position, living in a lone parent family
and coming from a larger family were associated with a higher risk of
becoming a teenage mother. 24, 25 The Longitudinal Study allowed a
detailed examination of the influence of these factors on teenage
motherhood in the 1980s.26, 27 Women aged 15–19 years who had a birth
between 1981 and 1990, were classified by their parental socioeconomic position and family structure at the 1981 Census.
Children aged under 18 living in childrens’ homes,
and men aged 15–29 in places of detention and
psychiatric institutions in 1971: prevalence of
limiting long-term illness at the 1981 Census, and
mortality, 1971–95. 1971 cohort, England and Wales
Relative risk of teenage motherhood, 1981–90, by
indicators of socio-economic position and family
circumstances in 1981. 1981 cohort, England and
Wales
Unadjusted
Adjusted**
Family social class
I
II
IIIN
IIIM
IV
V
Unclassified
1.00
2.36*
3.59*
6.41*
7.10*
9.96*
9.16*
1.00
2.18*
2.91*
3.87*
3.88*
4.39*
4.16*
Housing tenure
Owner occupied
Privately rented
Local authority
1.00
1.64*
3.54*
1.00
1.44*
2.40*
Age of mother
20 years and over
Under 20 years
1.00
2.31*
1.00
1.52*
Presence of normally resident father
Yes
1.00
No
1.77*
1.00
1.16
Number of siblings
None (lone child)
One
Two
Three or more
1.00
0.94
1.30*
1.67*
1.00
0.69
1.15*
1.81*
* p<0.05.
** Adjusted for all variables in table.
Table 10
Mortality of second generation Irish: relative risk
by age, 1971–95. 1971 cohort, England and Wales
Age
15–44
45–64
65+
15+
Women
All other
Second generation Irish
1.00
1.27*
1.00
1.07
1.00
1.12*
1.00
1.11*
Men
All other
Second generation Irish
1.00
1.29*
1.00
1.16*
1.00
1.00
1.00
1.12*
* p<0.05.
+ 1981/91 refers to day before census day in the first time period and census day onwards
in the following time period.
Health Statistics Quar terly 01
Table 11
ovarian cancer was not. Further studies are needed to understand why
the incidence of ovarian cancer changed so dramatically between the
two generations.
Mortality of second generation Irish adjusted for
indicators of socio-economic status: relative risk,
1971–95. 1971 cohort, England and Wales
Women
Men
Model 1: Irish~
All other
Second generation Irish
1.00
1.28*
1.00
1.21*
Model 2: Housing tenure
Owner occupied housing
Privately rented
Local authority
1.00
1.53*
1.31*
1.00
1.36*
1.26*
Model 3: Social class
Non-manual
Manual
Unclassified
1.00
1.48*
1.59*
1.00
1.20*
1.30
Model 4: Car access
Access to 2 or more
Access to 1
No access
1.00
1.26
1.72*
1.00
1.04
1.56*
Model 5: Adjusted for all indicators,
of socio-economic position
Second generation Irish v other
1.12*
1.18*
Poorer health in second generation Irish may be the result of the
persisting effect of parental socio-economic disadvantage. Although
socio-economic status was an important influence, being of Irish origin
was an independent predictor of poorer health in these studies. Another
longitudinal study of second and subsequent generation of Irish people
living in Scotland showed similar findings.31 These studies demonstrate
that differential health experience need not disappear rapidly following
migration and can persist across several generations.
SUMMARY
Narrowing inequalities in health between social groups is one of the
key aims of the Government Green Paper, Our Healthier Nation. 1 This
report presents updated findings from a range of studies on health
inequalities which are useful not only for public health monitoring
purposes but also for aetiological reasons. Other work which has not
been reported in this paper includes assessing the relative importance of
personal disadvantage and area disadvantage on health inequalities, the
contribution of social mobility to social gradients in health, and the
relationship between the birthweights of babies and coronary heart
disease mortality of their mothers. 32, 33, 34, 35, 36, 37
* p<0.05.
~ Excludes persons who could not be classified by all three indicators.
The shift from describing to explaining patterns and trends in health
differentials necessitates the collection of Longitudinal data. Longitudinal studies are expensive and the impact of a small sample size on the
power of studies is always problematic. They can be used, however, to
explore hypotheses that can then be tested in more suitable surveys.
They are also not suitable for the study of rare diseases, but in terms of
public health the study of common causes of death are of greater
importance. The ONS Longitudinal Study, albeit based on 1 per cent of
the population, has many advantages including a range of data from
censuses on key social indicators, long follow-up and a very low level
of loss to follow-up of its members. It remains one of the key tools for
monitoring socio-economic differentials in health in England and
Wales.
Table 10 shows that mortality was higher among second generation
Irish people compared with all other Longitudinal Study members. At
ages 15 to 44, mortality was almost 30 per cent higher among both men
and women. Table 11 shows that socio-economic status was an
important predictor of their mortality experience; those who were less
advantaged experienced higher mortality than those more advantaged.
After adjusting for differences in the distributions of socio-economic
circumstances, some, but not all, of the mortality excess was explained.
In another study of cancers in this population, the incidence of all
cancers including lung, ovarian and cervical cancers in women was
shown to be higher than all other women (Table 12) and the incidence
of prostate cancers in men higher than all other men (not shown).29
Socio-economic position was again clearly an important predictor of
these patterns. Adjusting for socio-economic status did not explain the
higher incidence of these cancers. The incidence of lung cancer was
also higher in first generation Irish (not shown) but the incidence of
Table 12
Spring 1999
Incidence of main cancers among second generation Irish women: relative risk, 1971–89. 1971 cohort, England and Wales
Breast
Lung
Colorectal
Ovary
Cervix
All cancers^
Model 1
Irish, adjusted for age
Second generation Irish v. other women
1.15
1.62*
1.06
1.75*
1.84*
1.21*
Model 2
Socio-economic status adjusted for age
Owner occupier
Local authority
Privately rented
1.00
0.58
0.82
1.00
3.73*
1.70
1.00
2.68*
0.67
1.00
0.70
0.37
1.00
3.15*
1.85
1.00
1.30*
0.91
Model 3
Irish, adjusted for age and socio-economic status
Socio-economic status
Second generation Irish v. other women
0.92*
1.18
1.22*
1.60*
0.99
1.07
0.87*
1.74*
1.33*
1.86*
1.02
1.22*
* p<0.05.
^ Exluding non-melanoma skin cancer (ICD9 140-208 x 173).
13
Office f or National Statistics
Health Statistics Quar terl y 01
Spring 1999
REFERENCES
Key findings
1
2
Updated results on key areas of health inequalities are
presented using the most recent data in the
Longitudinal Study.
SOCIAL CLASS AND EMPLOYMENT DIFFERENTIALS IN
3
4
MORTALITY
●
●
●
In 1990–95, life expectancy at birth of women in
Social Class IV/V was 78 years and that of women in
Social Class I/II was 81 years. The comparable figures
for men were 71 and 76 years.
Between the 1970s and 1990s, death rates declined
among all groups but the fall was greater among
those in Social Class I/II than among those in Social
Class IV/V.
In 1986–95, mortality of women and men who were
unemployed remained higher compared with that of
the employed. Mortality of unemployed women was
45 per cent higher than those employed.
SOCIO- ECONOMIC DIFFERENTIALS IN INCIDENCE
AND SURVIVAL FROM BREAST, OVARIAN AND
OF
5
6
7
8
9
PROSTATE CANCERS
●
●
●
At ages 65 and over, incidence of breast cancer
remained higher among women in advantaged socioeconomic circumstances than in less advantaged
ones. At younger ages, however, incidence was higher
among the less advantaged and their survival was
also poorer.
In 1986–90, incidence of ovarian cancer was higher
among those in advantaged socio-economic
circumstances than among those in less advantaged
ones. Survival from ovarian cancer was poor, about
two years, regardless of socio-economic
circumstances.
Incidence of prostate cancer remained higher among
more advantaged men. There were no significant
differences in survival, under six years, between the
social groups.
LIFE
●
●
●
CHANCES OF SOCIALLY EXCLUDED GROUPS
Children living in children’s homes, and young men in
prisons and in psychiatric institutions, were more
likely to move into a disadvantaged socio-economic
position compared with other people of comparable
ages.
Their risk of morbidity and premature mortality was
also considerably higher.
Having a teenage mother and disadvantaged socioeconomic circumstances were independent
predictors for daughters becoming teenage mothers.
HEALTH OF CHILDREN OF IRISH MIGRANTS LIVING
ENGLAND AND WALES
IN
●
Mortality of children born to Irish migrants and
living in England and Wales was higher than that of
all other people living in England and Wales. The
incidence of ovarian, cervical and lung cancers among
women and prostate cancer among men was also
higher.
Office for National Statistics
14
10
11
12
13
14
15
16
17
18
19
20
21
Department of Health (1998). Our Healthier Nation: a contract for
health. Consultation paper. The Stationery Office, London.
Hattersley L (1997). ‘Expectation of life by social class’ in (ed)
Drever F and Whitehead M, Health Inequalities, pp. 73–82. The
Stationery Office, London.
Harding S and Maxwell R (1997). ‘Differences in the mortality of
migrants’ in (ed) Drever F and Whitehead M, Health Inequalities,
pp. 108–21. The Stationery Office, London.
Harding S, Bethune A, Maxwell R and Brown J (1997). ‘Mortality
trends using the Longitudinal Study’ in (ed) Drever F and
Whitehead M, Health Inequalities, pp. 143–55. The Stationery
Office, London.
Bethune A (1997). ‘Unemployment and mortality’ in (ed) Drever F
and Whitehead M, Health Inequalities, pp. 156–67. The Stationery
Office, London.
Smith J and Harding S (1997). ‘Mortality of women and men using
alternative social classifications’ in (ed) Drever F and Whitehead
M, Health Inequalities, pp. 168–85. The Stationery Office, London.
Harding S, Rosato M and Dunnell K (1998). ‘ONS Longitudinal
Study’, Population Trends 93. The Stationery Office, London.
Harding S and Balarajan R (1996). ‘Patterns of mortality in second
generation Irish living in England and Wales: Longitudinal Study’.
British Medical Journal: 312, pp. 1389–92.
Harding S (1996). ‘The health of second generation Irish living in
England and Wales’, in The health of the Irish in Britain, a
community conference, pp. 11–20. Federation of Irish Societies,
London.
Hattersley L and Creeser R (1995). Longitudinal Study 1971–1991.
History, organisation and quality of data, LS 7. The Stationery
Office, London.
Goldblatt P (1990). ‘Mortality and alternative social classifications’
in (ed) Goldblatt P, Longitudinal Study: mortality and social
organistion, LS 6, pp. 164–92. HMSO, London.
Fox J, Goldblatt P and Jones D (1990). ‘Social class mortality
differentials: artifact, selection or life circumstances?’ in (ed)
Goldblatt P, Longitudinal Study: mortality and social organistion,
LS 6, pp. 100–108. HMSO, London.
Kunst A and Mackenbach J (1994). ‘The size of mortality
differences associated with educational level in nine industrialised
countries’. American Journal of Public Health 84, pp. 932–7.
Moser K, Goldblatt P, Fox J and Jones D (1990). ‘Unemployment
and mortality’ in (ed) Goldblatt P, Longitudinal Study: mortality
and social organistion, LS 6, pp. 82–97. HMSO, London.
Bartley M (1994). ‘Unemployment and ill health: understanding the
relationship’. Journal of Epidemiology and Community Health 48,
pp. 333–7.
Lewis G and Sloggett A (1998). ‘Suicide, deprivation, and
unemployment: record linkage study’. British Medical Journal,
317, pp. 1283–6.
Kogevinas E (1990). Longitudinal Study: socio-demographic
differences in cancer survival, LS 5. HMSO, London.
Leon D (1988). Longitudinal Study: Social distribution of cancer,
LS 3. HMSO, London.
Brown J, Harding S, Bethune A and Rosato M (1997). ‘Incidence of
Health of the Nation cancers by social class’. Population Trends 90,
pp. 40–47. The Stationery Office, London.
Krieger N (1997). ‘Race, ethnicity, social class and incidence of
cancer’. National Centre for Health Statistics 1997 Joint Meeting of
the Public Health Conference on Records and Statistics and the
Data Users’ Conference. Washington DC.
Jones C (1992). ‘Fertility of the over thirties’. Population Trends
67, pp. 10–16. HMSO, London.
Health Statistics Quar terly 01
22
23
24
25
26
27
28
29
Schrijvers C, Mackenbach J, Lutz J, Quinn M and Coleman M
(1995). ‘Deprivation, stage at diagnosis and cancer survival’.
International Journal of Cancer 63, pp. 324–9.
The Times (10 November 1998). Ad urges children to report violent
fathers. Government campaign will back women’s issues.
Kiernan K (1980). ‘Teenage motherhood – associated factors and
consequences – the experiences of a British birth cohort’. Journal
of Biosocial Science 12, pp. 393–405.
Manlove J (1997). ‘Early motherhood in an intergenerational
perspective: the experiences of a British Cohort’. Journal of
marriage and the family 59, pp. 263–79.
Rosato M (1997). ‘Teenage fertility in England and Wales: trends in
socio-economic circumstances between the 1971 and 1981
censuses’. Dissertation submitted in partial fulfillment of the
requirements of the MSc in medical Demography, London School of
Hygiene and Tropical Medicine.
Botting B, Rosato M and Woods R (1998). ‘Teenage mothers and
the health of their children’. Population Trends 93. The Stationary
Office, London.
Harding S, Davey Smith G, Chaturvedi N, Nazroo J and Williams
R (1997). ‘Ethnicity and Health: epidemiological approaches’.
Paper presented at conference ‘Making a difference for black and
minority ethnic communities’ hosted by Health Education
Authority/Kings Fund/Afiya/Policy Studies Institute, London.
Harding S (1998). ‘The incidence of cancers among second
generation Irish living in England and Wales’. British Journal of
Cancer 78 (7), pp. 958–61.
30
31
32
33
34
35
36
37
Spring 1999
Wild S and McKeigue P (1997). ‘Cross sectional analysis of
mortality by country of birth in England and Wales, 1970–92’.
British Medical Journal 314, pp. 705–9.
Abbott J, Williams R and Davey Smith G (1998). ‘Mortality in men
of Irish Heritage in West Scotland’. Journal of Public Health 112,
pp. 229–32.
Sloggett A and Joshi H (1994). ‘Higher mortality in deprived areas:
community or personal disadvantage?’ British Medical Journal
309, pp. 1470–74.
Sloggett A and Joshi H (1998). ‘Deprivation Indicators as
Predictors of Life Events, 1981–1992’. Journal of Epidemiology
and Community Health 52 (4), pp. 228–33.
Rosato M, Harding S, McVey E and Brown J (1998). ‘Research
implications of improvements in access to the ONS Longitudinal
Study’. Population Trends 91, pp. 35–42. The Stationery Office,
London.
Bartley M and Plewis I (1997). ‘Does health-selective mobility
account for socio-economic differences in health? Evidence from
England and Wales, 1971 to 1991’. Journal of Health and Social
Behaviour 38, pp. 376–86.
Gleave S, Bartley M and Wiggins R (1998). Limiting long-term
illness: a question of where you live or who you are? LS Working
Paper 77.
Blane D, Harding S and Rosato M (1999). ‘Social class mobility
and mortality among middle-aged men in England and Wales’.
Journal of Royal Statistics Society A 162, Part 1, pp. 59–70.
15
Office f or National Statistics
Health Statistics Quar terly 01
Spring 1999
Prescribing for patients
with asthma by general
practitioners in England
and Wales 1994–96
Azeem Majeed and
Kath Moser,
ONS
The objective of this study was to
examine trends in the
management of asthma in
general practice, and in
particular, to examine trends in
the use of inhaled steroids and
inhaled bronchodilators between
1994 and 1996. The data for the
study came from 288 general
practices, total list size 2.1
million, about 4 per cent of the
population in England and Wales,
on the General Practice Research
Database. Between 1994 and
1996, the percentage of
asthmatics being prescribed
inhaled steroids, either alone or
in combination with
bronchodilators, increased in all
age groups. The largest increase
in the use of combination
treatment was seen in children
under five years of age.The use of
bronchodilators alone in patients
with asthma fell in all age groups
but particularly in children. The
results suggest that the
management of asthma in
primary care is changing, with an
increase in the percentage of
asthmatics who are being
prescribed combination
treatment with bronchodilators
and either inhaled steroids or
inhaled cromoglycate.
Office for National Statistics
INTRODUCTION
Asthma remains an important cause of ill health. During the 1980s,
both hospital admission rates and the number of deaths from asthma
increased substantially. Admission rates for asthma increased from 1.1
to 1.6 per 1,000 between 1981 and 1985, an average annual increase of
nearly 10 per cent. The rate of increase then slowed but there was still a
further 1.7 per cent annual increase in admission rates during the
second half of the 1980s. The annual number of deaths from asthma
increased by 477 between 1980 and 1989, from 1,480 to 1,957.1,2
Consultation rates for asthma with general practitioners increased
nearly fivefold between 1976 and 1994.3 These increases in morbidity
and mortality occurred at a time when several effective treatments for
asthma were available and when new delivery systems for anti-asthma
medication were being introduced. Despite the availability of effective
treatments, previous surveys have shown that many patients with
asthma suffer from exacerbations of their condition, even in practices
that take a special interest in asthma care.4 Many patients and doctors
were over-reliant on bronchodilator medication (used to treat the
symptoms of asthma) and were reluctant to use prophylactic
medication such as inhaled steroids or inhaled cromoglycate (used to
prevent attacks of asthma from occurring).
Concerns about the growing morbidity from asthma led to the
publication in 1990 of British Thoracic Society guidelines for the
management of chronic asthma.5 These guidelines, which were revised
in 1993 and 1997,6,7 emphasised the importance of using prophylactic
medication such as inhaled steroids and cromoglycate in addition to
inhaled bronchodilators to help reduce the morbidity caused by asthma.
The impact of these guidelines on prescribing for asthma in general
practice is difficult to assess because routine general practice
prescribing data, derived from the Prescribing Analysis and Cost
16
Health Statistics Quar terly 01
(PACT) system, are only available at general practice level.8 The data
that are available show that there has been a substantial increase in the
use of bronchodilators and inhaled steroids by general practitioners
since the early 1980s.1 However, PACT data cannot be used to examine
the use of these drugs in different age groups. Furthermore, because
many prescriptions for respiratory drugs are for chronic airways
disease, routine prescribing data also cannot be used to examine trends
in the management of asthma as it is not possible to separate
prescribing for asthma from prescribing for chronic respiratory
disorders.
The objective of this study was to examine trends in the management of
asthma in general practice, and in particular, to examine trends in the
use of inhaled steroids and inhaled bronchodilators between 1994 and
1996, using data from the General Practice Research Database.
Figure 1
Spring 1999
Treatment for asthma by age group and sex:
rates per 1,000 in England and Wales, 1996
Treatment rate per 1,000
140
Males
Females
120
100
80
METHODS
The data for this study came from 288 general practices, total list size
2.1 million, in England and Wales contributing data to the United
Kingdom General Practice Research Database. The General Practice
Research Database (GPRD) was originally set up in 1987 by a
commercial company, VAMP Ltd (now Reuters Health Information
Ltd). It is currently owned by the Department of Health and has been
managed by the Office for National Statistics (formerly the Office of
Population Censuses and Surveys) since 1994. General practices
participating in the GPRD follow agreed guidelines for the recording of
clinical and prescribing data, and submit anonymised patient-based
clinical records to the database at regular intervals. Consequently, the
database contains longitudinal information on diagnoses, prescriptions
and hospital referrals. The availability of these data offer opportunities
for research on drug safety, the use of health services, and the
epidemiology and natural history of many diseases. 9
The accuracy and comprehensiveness of the data recorded in the GPRD
has been documented previously.10 All of the 288 practices included in
this analysis contributed data throughout the period 1994–96 and
passed regular quality checks. The combined population of the practices
had a very similar age-sex composition to the population of England
and Wales.11,12
For the most recent years for which prescribing data were available
(1994–96), we identified all patients with a diagnosis of asthma who
were being prescribed anti-asthmatic medication. These data were used
to calculate the number of people currently receiving treatment for
asthma per 1,000 population. We also calculated the percentage of
patients being treated for asthma who were being treated with
bronchodilators alone, inhaled corticosteroids or inhaled cromoglycate
alone, or with both bronchodilators and prophylactic medication.
Finally, we calculated prescribing rates for each of the 288 practices
and examined the association between the prevalence of asthma and the
use of combination treatment (defined as treatment with beta-2 agonists
in combination with either inhaled steroids or inhaled cromoglycate).
RESULTS
In 1996, 6.6 per cent of males and 6.8 per cent of females were
receiving treatment for asthma. Treatment rates were highest in boys
aged 5–15 years (123 per 1,000), followed by girls aged 5–15 years and
then boys aged 0–4 years. Among adults, treatment rates were highest
in people aged 65 years and over. Among children, substantially more
boys than girls were being treated for asthma. Among people aged
16–64 years, this pattern was reversed with a higher proportion of
women than men receiving treatment. Among the elderly, treatment
rates were similar in men and women (Figure 1). A greater percentage
of the patients registered with the practices were receiving treatment for
60
40
20
0
0–4
5–15
16–24
25–44
Age group
45–64
65+
asthma in 1996 than in 1994 (6.6 per cent versus 6.3 per cent in males
and 6.8 per cent versus 6.3 per cent in females). This increase was
almost entirely confined to adults; treatment rates in children showed
little change between 1994 and 1996.
Between 1994 and 1996, the percentage of patients being treated for
asthma who were prescribed bronchodilators in combination with either
inhaled steroids or inhaled cromoglycate increased from 62 per cent to
65 per cent in males and from 64 per cent to 67 per cent in females.
Most of the remaining patients were prescribed bronchodilators alone
with less than 10 per cent of patients being prescribed prophylactic
medication only (inhaled steroids or cromoglycate). This increase in the
use of combination treatment was greatest in children under five years
of age (58 per cent of boys treated with combination treatment in 1996
compared with 47 per cent in 1994 and 55 per cent of girls treated with
combination treatment in 1996 compared with 44 per cent in 1994).
However, despite this increase in young children, the use of
combination treatment with bronchodilators and either inhaled steroids
or cromoglycate in patients with a diagnosis of asthma remained
highest in the elderly (Table 1). The use of bronchodilators alone in
patients with asthma fell in all age groups, and particularly in children,
but even in 1996, 28 per cent of males and 26 per cent of females
receiving treatment for asthma were prescribed bronchodilators alone.
Inter-practice variation
The number of people undergoing treatment for asthma in 288 general
practices in this study varied from 19 to 135 per 1,000 (Table 2). The
percentage of patients with asthma being treated with bronchodilators
alone varied from 5 per cent to 60 per cent and the percentage on
inhaled corticosteroids alone varied from 0 per cent to 15 per cent. The
percentage on inhaled steroids, either used alone or in combination with
a bronchodilator, varied from 39 per cent to 95 per cent. The percentage
on inhaled cromoglycate alone varied from 0 per cent to 5 per cent;
17
Office for National Statistics
Health Statistics Quar terly 01
Table 1
Spring 1999
Percentage of patients prescribed treatment for asthma in 1994 and 1996 who were on combination treatment or who
were prescribed bronchodilators, inhaled corticosteroids or inhaled cromoglycate alone
Males
Combination treatment (%)
Bronchodilators alone (%)
Corticosteroids alone (%)
Cromoglycate alone (%)
1994
1996
1994
1996
1994
1996
1994
1996
0–4
5–15
16–24
25–44
45–64
65+
46.8
62.3
53.0
56.8
70.7
75.5
57.8
63.8
56.6
59.0
71.8
77.3
50.1
30.4
42.5
36.1
20.9
17.9
37.0
28.5
38.6
33.9
19.7
15.9
2.1
5.1
3.7
5.8
7.4
6.2
4.4
6.3
4.2
6.5
7.8
6.6
0.8
1.9
0.8
1.2
0.8
0.2
0.6
1.2
0.5
0.6
0.6
0.2
All ages
61.8
64.7
31.5
28.3
5.3
6.2
1.1
0.7
Age group
Females
Combination treatment (%)
Bronchodilators alone (%)
Corticosteroids alone (%)
Cromoglycate alone (%)
1994
1996
1994
1996
1994
1996
1994
1996
0–4
5–15
16–24
25–44
45–64
65+
44.3
60.4
58.8
62.4
71.6
72.8
54.9
63.2
60.0
64.1
73.0
75.2
52.2
33.0
35.7
30.3
19.7
20.4
39.6
29.5
34.7
28.7
17.9
17.1
1.9
4.6
5.0
6.9
8.0
6.4
4.6
5.8
5.0
6.9
8.6
7.4
1.4
1.8
0.4
0.4
0.5
0.3
0.6
1.3
0.3
0.2
0.4
0.3
All ages
64.2
66.9
28.9
25.7
6.0
6.8
0.7
0.5
Age group
Table 2
Inter-practice variation in prevalence of asthma
and in percentage of patients receiving treatment
for asthma who were on combination treatment
or who were prescribed bronchodilators, inhaled
corticosteroids or inhaled cromoglycate alone
Mean (standard deviation)
Prevalence per 1,000
Combination treatment (%)
Bronchodilators alone (%)
Inhaled corticosteroids alone (%)
Steroids alone or in combination (%)
Cromoglycate alone (%)
66.0 (16.3)
65.7 (6.4)
27.2 (6.8)
6.4 (3.0)
72.0 (6.9)
0.7 (0.8)
Range
19.1
37.5
4.6
0.0
39.3
0.0
to
to
to
to
to
to
134.9
95.1
59.8
15.3
95.1
4.9
20 per cent of the 288 practices prescribed no cromoglycate and a
further 60 per cent prescribed cromoglycate for less than 1 per cent of
their patients with asthma. The use of combination treatment with
bronchodilators and either inhaled steroids or cromoglycate varied from
38 per cent to 95 per cent. There was a weak positive association
between the prevalence of asthma and the use of combination treatment
(correlation coefficient 0.21, 95 per cent confidence intervals 0.10 to
0.32). Hence, although practices with a higher prevalence of asthma
tended to use more combination treatment, this association was not very
strong, and explained only about 4 per cent of the variation in the use of
combination treatment.
DISCUSSION
This study is among the largest to examine the management of asthma
in primary care. We found that the management of asthma in primary
care is changing, with an increase between 1994 and 1996 in the
percentage of asthmatics who are being prescribed combination
Office for National Statistics
18
treatment of bronchodilators and either inhaled steroids or inhaled
cromoglycate. The use of inhaled corticosteroids alone also increased
during the same period. Hence, it appears that general practitioners are
changing their management of asthma to reflect the recommendations
of the British Thoracic Society guidelines published in 1990 and
revised in 1993. The introduction of these guidelines has coincided with
a fall in death rates from asthma and a stabilisation in admission rates.
Whether these changes are due in part to improved management of
asthma in primary care (in particular, to greater use of inhaled steroids)
is a question that cannot be answered by this study.
The use of cromoglycate fell between 1994 and 1996, continuing a
long-term decline in the use of this agent.13 Most practices now make
little use of cromoglycate and now mainly use inhaled steroids for the
prophylaxis of asthma. We found only a very weak association between
the prevalence of asthma and the use of combination treatment. This
implies that practices in which a high proportion of patients are being
treated for asthma are not very much more likely to use combination
treatment. Hence, the management of asthma appears to be one area of
medical treatment where a higher number of patients being treated does
not necessarily seem to result in a major improvement in prescribing.
A major problem with studies of the epidemiology and management of
asthma is the absence of an agreed case definition. This leads to
problems comparing the results of different studies. 14 In previous
epidemiological studies, three different methods have generally been
used to diagnose asthma: a doctor’s diagnosis; the presence of
symptoms of asthma; and the presence of bronchial hyper-reactivity. In
this study, we defined asthmatic patients as those who had received a
diagnosis of asthma from a doctor and who had in the previous 12
months been prescribed medication for the treatment of asthma. The
main problem with using a doctor’s diagnosis to define asthma is that
Health Statistics Quar terly 01
there is great variation between doctors in their propensity to diagnose
asthma. The additional condition that the patients had to be on
medication for the treatment of asthma to be classified as asthmatic was
added to try to reduce the extent of this variation by ensuring only
patients being actively treated were included in the study. Another
disadvantage of using a medical diagnosis to define asthma is that some
asthmatic patients will not have had their asthma diagnosed by a
doctor. Hence, studies such as this which rely on a doctor's diagnosis,
will include some patients who do not have asthma and exclude other
patients who do have asthma. Finally, most previous studies of the
epidemiology of asthma have focused almost entirely on children,
hence it is really only possible to compare the results for children in
this study with the results of previous studies. The prevalence of
childhood asthma in this study was in line with those in an
epidemiological overview of asthma published in 1994,16 providing
some validation of our results.
Key findings
●
●
●
We were unable to investigate other aspects of asthma morbidity such
as acute asthma attacks, attendance at accident and emergency
departments, and referral to hospital outpatient departments.17,18
Investigating the association between such measures and prescribing for
asthma in primary care will be an important area for future research.
Finally, revised guidance on the management of asthma recommending
earlier use of inhaled steroids, was issued by the British Thoracic
Society in 1997 (Box One). Encouragingly, we found that over 70 per
cent of asthmatic patients in primary care were being prescribed inhaled
steroids in 1996, either alone or in combination with a bronchodilator.
The further impact of the revised British Thoracic Society guidelines on
prescribing in general practice will be examined in future analyses of
data from the General Practice Research Database.
●
●
Box one
CURRENT BRITISH THORACIC SOCIETY GUIDELINES
Spring 1999
In asthmatic patients, inhaled steroids are used
to reduce the severity of symptoms and the
risk of patients suffering from sudden attacks
of asthma.
British Thoracic Society guidelines published in
1990 and revised in 1993 recommended that
greater use of inhaled steroids should be made
in the management of asthma in general
practice.
In 1996, about 6.6 per cent of males and 6.8
per cent of females were receiving treatment
for asthma. Treatment rates for asthma were
highest in children aged 5–15 years. Among
adults, rates were highest in people aged 65
years and over.
Between 1994 and 1996, the percentage of
patients being treated for asthma who were
prescribed bronchodilators in combination
with inhaled steroids and/or inhaled
cromoglycate increased from 62 per cent to 65
per cent in males and from 64 per cent to 67
per cent in females. The increase was greatest
in children under five years of age (58 per cent
of boys treated with both in 1996 compared
with 47 per cent in 1994 and 55 per cent of
girls treated with both in 1996 compared
with 44 per cent in 1994).
Over 70 per cent of asthmatic patients are
currently being prescribed inhaled steroids by
their general practitioners.
REFERENCES
ON THE MANAGEMENT OF CHRONIC ASTHMA IN
ADULTS AND SCHOOL CHILDREN
PUBLISHED IN
– GUIDANCE
1997
1
2
Step 1 Occasional use of bronchodilators
Step 2 Regular use of inhaled anti-inflammatory
agents (steroids)
Step 3 High dose inhaled steroids or low dose
inhaled steroids plus long acting
bronchodilator
Step 4 High dose inhaled steroids and regular
bronchodilators
3
4
5
6
Step 5 Addition of regular steroid tablets
7
British Thoracic Guidelines published in 1990 and 1993
emphasised the importance of the regular use of
inhaled steroids in patients with mild or moderate
asthma. The most recent revision of these guidelines
published in 1997 further emphasised the importance
of the early use of inhaled steroids. A step-wise
approach to treatment is recommended with the
objective of abolishing the symptoms of asthma as
quickly as possible. Once control is achieved, treatment
can be stepped down.
8
9
10
11
Department of Health (1995). Asthma. An epidemiological
overview. Department of Health, London.
Majeed A and Kaur B (1998). ‘The changing nature of asthma’, in
Hilton S and Levy M (eds), Asthma in practice. Royal College of
General Practitioners, London. (Forthcoming in 1999.)
Drever F (1994). ‘Asthma, the changing scene?’ Population Trends
78, pp. 44–6.
Keeley D (1993). ‘How to achieve better outcome in treatment of
asthma in general practice’. BMJ 307, pp. 1261–3.
British Thoracic Society (1990). ‘Guidelines for management of
asthma in adults: I chronic persistent asthma’. BMJ 301, pp. 651–3.
British Thoracic Society (1993). ‘Guidelines on the management of
asthma’. Thorax 48, pp. S1–24.
British Thoracic Society (1997). ‘The British guidelines on asthma
management. 1995 review and position statement’. Thorax 52, pp.
S1–21.
Majeed A, Evans N and Head P (1997). ‘What can PACT tell us
about prescribing in general practice?’ BMJ 315, pp. 1515–9.
Walley T and Mantgani A (1997). ‘The UK General Practice
Research Database’. Lancet 350, pp. 1097–9.
Hollowell J (1997). ‘The General Practice Research Database:
quality of morbidity data’. Population Trends 87, pp. 36–40.
Office for National Statistics (1998). Key Health Statistics from
General Practice 1996 (Series MB6 No. 1). Office for National
Statistics, London.
19
Office for National Statistics
Health Statistics Quar terly 01
12
13
14
15
Spring 1999
Lawson D H, Sherman V and Hollowell J (1998). ‘The General
Practice Research Database’. Quarterly Journal of Medicine 91,
pp. 445–52.
Warner J O (1995). ‘Review of prescribed treatment for children
with asthma in 1990’. BMJ 311, pp. 663–6.
Strachan D P (1995). ‘Epidemiology’, in Silverman A (ed),
Childhood asthma and other wheezing disorders. Chapman and
Hall, London.
Speight A N, Lee D A and Hey E N (1983). ‘Underdiagnosis and
undertreatment of asthma in childhood’. BMJ 286, pp. 1253–6.
Office for National Statistics
20
16
17
18
Anderson H R, Esmail A, Hollowell J, Littlejohns P and Strachan D
P (1994). ‘Lower respiratory disease’, in Stevens A and Raftery J
(eds), Health care needs assessment. Radcliffe Medical Press, Oxford.
Griffiths C, Sturdy P, Naish J, Omar R, Dolan S and Feder G
(1997). ‘Hospital admissions for asthma in east London:
associations with characteristics of local general practices,
prescribing, and population’. BMJ 314, pp. 482–6.
Shelley M, Croft P, Chapman S and Pantin C (1996). ‘Is the ratio of
corticosteroid to bronchodilator a good indicator of the quality of
asthma prescribing? Cross-sectional study linking prescribing data
to data on admissions’. BMJ 312, pp. 1124–6.
Death certification and the
epidemiologist
Tim Devis and Cleo Rooney,
ONS
The various processes by which
deaths are certified and
registered by doctors and
coroners, are of considerable
interest to the epidemiologist
studying mortality trends and
patterns. This paper describes
and discusses these processes in
England and Wales, some dating
from the nineteenth century,
and illustrates in particular the
importance of accurately
recording the causes of deaths.
It also updates an earlier article
on this subject1 and takes
account of the many changes
affecting registration and
certification in recent years.
I NTRODUCTION
In England and Wales the modern system of birth and death registration
has its origins in an act of parliament which came into force in 1837. 2
This provided the basis for a national registration system, directed by a
General Register Office headed by the Registrar General. However, as
originally presented to parliament in 1836 the Registration Bill did not
include a provision to record the cause of death when a death was
entered in the register. Following pressure from the reformer Edwin
Chadwick and others for its inclusion, the Bill was amended and later
passed with this requirement included. Chadwick’s interests extended
beyond the legal and administrative advantages of a uniform system to
the insight such information could give into social conditions and
public health problems. As a result the Registrar General has been able
to collect and publish mortality information by cause of death since the
Act came into operation.3,4
The first Registrar General, Thomas Lister, was quick to enlist the
co–operation of the medical profession in making the statements of
cause of death as accurate as possible. At his invitation the heads of
various medical colleges pledged themselves and their members to
give an authentic name to the conditions leading to death, when
completing the registration. Lister entrusted the work of analysing
and developing these new data sources to his medical statistician
William Farr, who in the next 40 years was outstanding in exploiting
and expanding the use of the data, particularly in providing evidence
of the effects of insanitary and unhealthy conditions. Farr was also a
prime mover in securing recognition of the importance of scientific
nomenclature and scientific classification in medical statistics,
initiating a nosology which culminated in an internationally agreed
classification of diseases, injuries and causes of death. The general
21
Office for National Statistics
Health Statistics Quar terly 01
Spring 1999
arrangement of diseases by anatomical site was proposed by him in
1855, and has survived as the basis of the International
Classification of Diseases (ICD), first adopted in 1900. The Ninth
Revision of ICD 5 (ICD9) is currently used by ONS to classify causes
of death. The Tenth Revision is already in use in several other
countries, and will be introduced in 2001 in England and Wales.
However, it was introduced for morbidity coding in the National
Health Service in 1995.
The subsequent Births and Deaths Registration Act of 1874, which
made death registration compulsory, also placed a specific duty on the
medical practitioner who attended the deceased during the last illness
to provide a statement of the cause of death, unless there was an
inquest – this act involved coroners for the first time.
A further advance followed in the Birth and Death Registration Act
1926, which compelled medical practitioners to use a standard
printed form for certifying the causes of death. Up to this time the
form in use distinguished ‘primary’ and ‘secondary’ causes of
death, often making it difficult to say which cause was a
consequence of another. 6 From 1927 certifiers were required to
give in sequential order the medical conditions leading to death, as
discussed later in this paper. Another act of 1926 made provision
for coroners to distinguish between post-mortem examinations
which accompanied inquests and those where no inquest was held.
The power to hold a post-mortem7 and then dispense with an
inquest was also introduced. Statistics on this aspect of registration
have thus been available only since 1928.
More recently, the Registration Service Act of 1953 consolidated earlier
provisions covering the organisation of the registration service, while
the Births and Deaths Registration Act of the same year covered the
registration of births, stillbirths and deaths. Two Population Statistics
Acts, of 1938 and 1960, made provision inter alia for certain
information to be collected in confidence and not entered in the public
record; these details may be used by the Registrar General only for
statistical purposes, and may not be released under any circumstances.8
For deaths, this information at present includes the marital status of the
deceased, and the age of the surviving spouse (if any) of the deceased.
RECENT DEVELOPMENTS
Earlier this decade ONS (then the Office of Population Censuses and
Surveys (OPCS)) carried out an extensive redevelopment of its
collection and processing systems for population, health, and
registration data – in particular, for births and deaths. For deaths this
included: the progressive computerisation of registration in local
offices; the move to a large deaths database to hold all mortality data
from 1993; and the introduction of automated coding of cause of death.9
BIRTHS AND DEATHS REGISTRATION ACT 1953
(Form prescribed by Registration of Births and Deaths Regulations 1987)
Registrar to enter
No. of Death Entry
MEDICAL CERTIFICATE OF CAUSE OF DEATH
…………………
For use only by a Registered Medical Practitioner WHO HAS BEEN IN ATTENDANCE during the deceased’s last illness,
and to be delivered by him forthwith to the Registrar of Births and Deaths.
Name of deceased ...............................................................................................................................................................................
Date of death as stated to me .................................................................. day of ...............................
............................... Age as stated to me ..........
Place of death .....................................................................................................................................................................................
Last seen alive by me ............................................................................. day of ...............................
.........................................
} {
1 The certified cause of death takes account of information
obtained from post-mortem.
2 Information from post-mortem may be available later
3 Post mortem not being held.
4 I have reported this death to the Coroner for further action.
(See overleaf)
Please ring
appropriate
digit(s) and letter
a
b
c
CAUSE OF DEATH
Seen after death by me.
Seen after death by another medical practitioner
but not by me
Not seen after death by a medical practitioner.
IM
C
E
The condition thought to be the ‘Underlying Cause of Death’ should
appear in the lowest completed line of Part I.
N
E
These particulars not to be
entered in death register
Approximate interval
between onset and death
I (a)Disease or condition directly
leading to death† .............................................................................................................................................................................................
...................................
(b)Other disease or condition, if any,
leading to: I(a) ................................................................................................................................................................................................
...................................
(c)
Other disease or condition, if any,
leading to: I(b) ...............................................................................................................................................................................................
...................................
II
P
S
Other significant conditions
CONTRIBUTING TO THE DEATH but not related to the disease or condition
causing it .........................................................................................................................................................................................................
The death might have been due to or contributed to by the employment followed at some time by the deceased
...................................
Please tick
where applicable
✝ This does not mean the mode of dying, such as heart failure, asphyxia, asthenia, etc: it means the disease, injury, or complication which caused death.
I hereby certify that I was in medical attendance during
the above named deceased’s last illness, and that the
particulars and cause of death above written are true
to the best of my knowledge and belief.
For deaths in hospital:
Qualifications as registered
Signature ....................................................................... by General Medical Council ........................................
Residence ................................................................................... Date ..................................................................
Please give the name of the consultant responsible for the above- named as a patient ........................................................................................................................
Office for National Statistics
22
Health Statistics Quar terly 01
Box one
GROUNDS FOR REFERRING A DEATH TO THE CORONER
ONS encourages the prevailing practice of voluntar y
referral to the coroner by the certifying doctor who
should consider :
• Whether the death was
– an accident (whenever it occurred);
– a suicide;
– related to the deceased’s employment.
• Whether the death occurred during or shortly after
detention in police or prison custody.
• Whether s/he or another doctor is legally qualified
to certify the death.
A registrar is legally obliged to refer a death to the
coroner (unless it has already been reported) if it falls,
or appears from the doctor’s death certificate to fall,
into one of the following categories:
• The deceased was not attended during his or her
last illness by a doctor.
• The registrar has been unable to obtain a duly
completed death certificate, or else it appears that
the deceased was not seen by the certifying doctor
either after death or during the 14 days before death.
• The cause of death appears to be unknown.
• The registrar has reason to believe the death was
unnatural, or caused by violence or neglect, or by
abortion, or was in any way suspicious.
• The death appears to have occurred during an
operation or before recover y from the eff ect of
an anaesthetic.
• The death certificate suggests that death was due to
industrial disease or industrial poisoning.
Source: Letter to all doctors in England and Wales from the Deputy Chief Medical
Statistician, ONS, dated l July l996.
The main effect of these changes on registration practice has been in the
way deaths have been handled by local registrars. When a death occurs,
the attending doctor will usually complete a medical certificate of cause
of death (MCCD) (see left) which is taken to the local registrar, who
generally produces a draft of the details about the death. Until recently
the registrar carried out the registration by filling out a form by hand, but
this practice is now uncommon. With computerisation of the registration
service the details from the MCCD, and other particulars supplied by the
informant (or, if there was an inquest, by the coroner), are entered into a
PC by the registrar. Draft details about the death are then printed
automatically, and the information stored and sent weekly on floppy disk
to ONS for processing.
At present (November 1998) 97 per cent of death registrations are
handled in computerised registration offices. Registrations in non–
computerised offices are notified to ONS through paper drafts, from
which information is keyed into the processing systems. Redevelopment
in this area has meant that information about nearly all deaths can be
handled more consistently and efficiently than before.
Spring 1999
A new database was also introduced to store information on deaths.
The information sent from registration offices is now loaded on to the
database, and then processed and edited. Prior to 1993 ONS (then OPCS)
produced an annual computer file containing details of all registrations in
a particular year. In the new system there are in practice two deaths
databases: one contains textual information corresponding to the public
record – these are the details supplied by informants when registering a
death, and to applicants requesting a copy of the death certificate – and
the other is a statistical database, which contains only coded details of
each death. Outputs are obtained by accessing the database to supply
information, whether on individual deaths or as datasets to produce
tabulations. Information on cause of death is held in coded form in the
statistical database, and as text in the other (registration) database.
Information sent to ONS includes causes of death, which are coded to
ICD9, and an underlying cause, identified by the Automated Cause
Coding system (ACCS). About 80 per cent of deaths are now coded this
way. The cause coding of deaths certified after inquest is still carried
out clerically to ensure consistent handling of these cases.10 The rules
and procedures used in ACCS ensure more consistency than the clerical
system, and have several advantages – improved consistency, better
international comparability, and the automatic coding of all causes
mentioned on most death certificates.
A revised reporting form for coroners introduced in 1993 changed the
data on deaths from injury and poisoning. Less specific detail about
the nature of injury is now available for external cause deaths,
compared with previous years.11
CERTIFICATION
When a person dies the attending doctor completes the MCCD for the
local registrar of births and deaths. The informant, often a relative of
the deceased, delivers the doctor’s certificate when s/he goes to register
the death. The majority of deaths are handled in this way, and unless the
registrar considers it necessary to refer the case to the coroner – which
is unusual as only 1 per cent are so referred – the death is registered
without further ado.
The registrar collects from the informant all the details for registration,
except the cause of death. Other details required for statistical purposes
are collected partly from the MCCD and partly from the informant. All
these details appear in the draft entry prepared by the registrar, a copy
of which is sent to ONS.
The MCCD is in a form prescribed by law, 12 and should be completed
only by a doctor who has been in attendance during the deceased’s last
illness. Much of the epidemiological interest centres round the medical
causes of death, and this part of the certificate is discussed below in
more detail. Although the certifier gives the date and place of death,
these details are entered in the register from the statement of the
informant. Other information given by the certifier does not appear in
the register, a copy of which is supplied to the informant, but is entered
by the registrar for statistical use within ONS.
Most of these residual statistical details, such as when the deceased was
last seen alive, whether the body was seen after death by the certifier,
whether death was related to employment, and duration of illness for
each medical condition mentioned, are analysed in annual published
tables. Of the remainder, information on whether or not a post-mortem
has taken place, and whether the death has been reported to a coroner, is
valuable in analysing the treatment of more unusual deaths, and is
discussed below. In addition, Box B (on the rear of the MCCD) is used
by the certifier to indicate that more information may be available later.
23
Office for National Statistics
Health Statistics Quar terly 01
Spring 1999
Certification and registration of deaths in
England and Wales
Figure 1
All Deaths
➝
➝
➝
Originally certified
by doctor on
MCCD*
Not properly
certified and not
referred to coroner
No MCCD* – sent
direct to coroner
➝
➝
by doctor
Yes
➝
No
Deaths for which there was no doctor in attendance, such as sudden or
unexpected or violent deaths, may be referred directly to the coroner by the
police. At this point the coroner has a number of possible courses of action.
Where he is satisfied that the death is due to natural causes and the cause is
correctly certified, he will instruct the local registrar (on form 100A) to
register the death, using the cause of death given on the original medical
certificate. If this certificate was not completed in the first place, the
registrar will use the cause of death stated by the coroner on form 100A.
Alternatively, the coroner may order a post-mortem examination,
particularly where the death was sudden and the cause unknown. If this
shows unequivocally that the death was due to natural causes he may
dispense with the requirement to hold an inquest. He will then certify
the cause of death from the pathologist’s report on the post-mortem
(form 100B). The registrar will use this certificate to register the death.
Yes
Referred to
coroner?
For some deaths the doctor may certify the cause and report the case to
the coroner, or the registrar may report it. Deaths which should be
referred to the coroner by the registrar are listed in Box One.
by registrar
➝
➝
➝
➝
Coroner not
consulted
Coroner orders No Not legally
➝ certified by
post–mortem
coroner
and/or inquest?
No
Yes
Yes
➝
➝
➝
➝
➝
Registrar
Certified by
coroner
➝
Certified by
doctor
Inquest with or
without post–
mortem
Uncertified
➝
➝
➝
➝
Post–mortem
only
Deaths by certifier, and by whether inquest and post-mortem carried out, England and Wales, 1879–1996
Certification
Doctor
with post-mortem
without post-mortem
Coroner
Post-mortem only
Inquest and post-mortem
Inquest only
}
}
Certified with post-mortem
Uncertified
Total
Total deaths†
*
†
**
If it appears that someone is to be charged with an offence in relation to
the death, the coroner must adjourn the inquest until legal proceedings
are completed. Since 1978 it has been possible to register these deaths
at the time of adjournment, when the coroner issues form 120. This
form includes details of injuries which led to the death, but no verdict.
In the case of motor vehicle incidents, this will provide enough
information to code the cause of death. Other deaths, such as possible
homicides, are given a temporary code for underlying cause until final
information becomes available.13 This is supplied by the coroner to the
registrar on form 121.
A very small proportion of deaths remains legally ‘uncertified’. ONS
receives copies of at least one medical certificate of cause of death for
* Medical certificate of cause of death.
Table 1
➝
If an inquest is necessary, the death can be registered only after the
inquest. In nearly all cases the inquest follows a post-mortem by a
pathologist. In most cases the inquest concludes the investigation and
the death is then certified by the coroner (form 99(REV)). This provides
the registrar with details of the deceased as well as the inquest findings
as to cause of death.
1879
(%)
1928
(%)
1953
(%)
1963
(%)
1973
(%)
90
90
8
83
91
5
1
7
1987
(%)
1992
(%)
1996
(%)
9
77
86
10
73
83
6
72
78
2
74
76
2
76
77
*
*
78
12
4
1
17
18
3
1
22
20
4
0
24
19
4
0
23
18
4
0
22
5
8
9
2
3
14
n/k
9
20
26
27
26
24
*
5
1
0
0
0
0
0
0
100
100
100
100
100
100
100
100
526,255
460,389
572,868
587,478
566,994
558,313
560,135
}
114,642 **
Doctor’s post-mortems for 1996 could not be separately identified with accuracy.
Numbers of deaths are registrations in each year except 1996, which are occurrences.
Deaths in second quarter of 1953.
Source: RGs Annual Reports for years 1879 to 1973; Annual Reference Volumes in DH1 series for 1987 to 1996.
Office for National Statistics
24
Health Statistics Quar terly 01
these cases, which are registered and coded as normal. This group
includes deaths for which the doctor who completed the medical
certificate did not fulfil all the legal requirements for doing so. For
instance, the doctor was not in attendance on the deceased during the
last illness and did not see the body, and the coroner did not order a
post-mortem but issued form 100A. It also includes deaths of foreign
military personnel (and their dependants) in England and Wales, where
the certifying doctor was not a medical practitioner registered in
England and Wales for the purpose of issuing certificates.
A copy of the MCCD is shown on page 22, while copies of the
coroner’s forms may be found at the back of any recent volume in the
DH series, published by ONS. The processes of certification and
registration in England and Wales are illustrated in Figure 1.
Figure 2
Spring 1999
Deaths in 1996 by place of death and per cent
certified by doctors and coroners, with or
without inquest, in England and Wales
Coroner with post-mortem
Inquest
Doctor
Percentage
100
75
50
Table 1 shows the decreasing role of the doctor in death certification,
and the increasing roles, particularly in tandem, of the coroner and
pathologist who carry out post-mortems. These have been influenced on
the one hand by changes in statutory responsibilities, e.g. 1926 Acts,
and on the other by changing patterns in causes of death – for instance,
a decline in deaths due to infectious diseases and a rise and fall in
accidental deaths. Over the last three decades the pattern has stabilised,
with coroners now certifying about one death in every four. The
contribution of the pathologist is now mainly to coroners’ cases.
DOCTORS
Medical practitioners in England and Wales have been legally
required to certify the cause of death for patients under their care
since 1874. Current legislation requires that the doctor ‘shall sign a
certificate in the prescribed form stating to the best of his/her
knowledge and belief the cause of death’ for the death of any person
attended during their last illness. The doctor is also required to deliver
the certificate to the registrar. 14
The legal responsibility for referring deaths to the coroner rests with
the registrar of births, marriages and deaths, rather than the doctor.
In practice 96 per cent of deaths (in 1996) which involved both a
doctor and coroner were referred by the doctor, 15 often before
completing the MCCD. In fact, if the coroner accepts jurisdiction,
the doctor may not complete a MCCD at all since the registrar must
then use the coroner’s certificate to register the death. The Registrar
General does not normally enforce completion of a redundant
document in these cases.
To be able to complete the MCCD a medical practitioner must, by
law, be fully registered with the General Medical Council. House
officers in hospitals who are only provisionally registered may
complete MCCDs only for patients who die in hospital, and are
under the care of a consultant who is also supervising their training.
The consultant should supervise the certification, having a
responsibility to see that it is done correctly and in accordance with
the law. Since 1986 the name of the consultant responsible for care
of the deceased, as well as that of the certifier, must be entered on
the certificate for all deaths in hospital. Any further enquiries are
addressed to the consultant, who is more likely to be still available
in the same post than are junior staff.
A high proportion of certificates are signed by doctors, normally
without the benefit of post-mortem information. How experienced are
these certifiers? Some insight into this may be gained from an analysis
of doctor certified deaths by place of occurrence.
The place where a death occurred must be recorded by the registrar, and
is usually based on details supplied by the informant. In 1996, 54 per
25
0
Hospice
All nonNHS
hospitals
Other
communal
establishments
All NHS Own home Elsewhere
hospitals
Total number of deaths by place
number
20,152
per cent 3.6
58,050
44,721
305,552
115,148
16,512
10.4
8.0
54.5
20.6
2.9
cent of deaths took place in NHS hospitals and places for the care of the
sick, which includes NHS nursing homes, while 10 per cent were in
similar communal establishments outside the NHS; 21 per cent
occurred in the deceased’s own home, and 8 per cent in other
communal establishments, which includes old people’s homes.16
Figure 2 shows the percentage of deaths occurring in each of these
places which are certified by doctors, by coroners after post-mortem
only, and after inquest. Nearly all deaths in hospices are certified by
doctors, as would be expected in communal establishments which
care specifically for patients known to be terminally ill. At the
opposite extreme, deaths most likely to be certified by coroners are
those which occur ‘elsewhere’. These include deaths at the scene of
an accident as well as others outside the home or hospital. While 83
per cent of deaths in non–psychiatric hospitals are certified by
doctors, the proportion of those in psychiatric hospitals is 90 per cent.
Deaths in general hospitals will include some caused by accidents or
violence where death is not immediate.
Of the 560 thousand deaths in 1996, 436 thousand were certified by
doctors rather than by coroners. Sixty two per cent of these occurred in
hospitals (270 thousand). Thus almost one half of all deaths are
registered by hospital doctors – who are mostly junior.
CORONERS
Coroners date from 1194, when they were created to fill the need for an
official whose main duty was to protect the financial interest of the
Crown in criminal proceedings.17 This official was elected as a ‘keeper
of the pleas of the Crown’ or custos placitorum coronas, from which
the present name derives. 18 Over time the duties changed, so that by
1500 the main function performed by the coroner was the holding of
inquests into violent deaths. The 1887 Coroners Act established that
coroners were concerned not so much with safeguarding financial
25
Office for National Statistics
Health Statistics Quar terly 01
Spring 1999
Table 2
Box two
DEATHS REFERRED FOR FURTHER INVESTIGATION IN
S COTLAND AND NORTHERN IRELAND
Deaths referred to coroners by source of referral,
England and Wales, 1987 to 1996
Source of referral
1987
(%)
1992
(%)
1996
(%)
Formally referred
In both Scotland and Northern Ireland the procedures for
certifying and registering most deaths are similar to those
in England and Wales. However, in Scotland the registrar
of deaths is obliged to report to the procurator fiscal any
death which falls into a broad range of some 20
categories. Deaths may also be reported to the
procurator fiscal by the police or by attending doctors.
When a death has been referred to him, the fiscal must
decide whether any further enquiry is necessary, for
instance whether to request an autopsy. The death
meanwhile will have been registered in the normal way,
and the information passed to the General Register Office
for Scotland (GRO(S)), where it will be used to code the
cause of death.
The procurator fiscal is responsible for prosecution of
criminal offences in his district, as well as for
investigating any sudden, violent, suspicious or accidental
death, or death from an unknown cause, which is
reported to him. 25 The main aim for the fiscal is to
establish whether or not there has been any criminality
or possible negligence involved in a death. In such cases
he must repor t the results of his investigation to the
Crown Office. There a decision is made as to whether
criminal proceedings are to be instituted, or a public
enquiry held, or no further action taken. He is not
obliged to establish the precise cause of death in a
medical sense, once the possibility of criminal
proceedings has been ruled out. His enquiries are
conducted informally and in private.
When the procurator fiscal has completed his
examination of the case he informs GRO(S) of any
changes to the information originally recorded on the
death certificate, including clarification of the cause of
death. GRO(S) will amend their records accordingly and,
if necessary, change the code assigned to the death.
In 1996 registrars notified procurators fiscal of the
particulars of 12,516 deaths where the circumstances
indicated that the fiscal might wish to enquire into the
cause of death. 26 The number notified constituted 21 per
cent of all deaths. In 1,443 (12 per cent) of these cases,
reports were subsequently received from the
procurators fiscal, of which 1,264 (10 per cent)
warranted amendments to the particulars given in the
relevant death entries. There is no equivalent to the
inquest in Scotland, but there are a small number of fatal
accident enquiries, covering usually less than 100 deaths
annually.
In Nor thern Ireland cases are referred to a coroner in
much the same way as in England and Wales, although
the list of formal grounds for referral is more general
than in Scotland or England and Wales, including for
example deaths ‘as a result of violence or misadventure,
or by unfair means’. Referrals to a coroner in Northern
Ireland are relatively less common – 2,799 deaths were
referred in 1996 (18 per cent of all deaths). For 1,047
of these deaths the coroner indicated that there was no
reason to dispute the certified cause of death, and for a
further 1,268 the coroner stated that a post-mortem
had been carried out but, again, there was no reason to
dispute the certified cause. Inquests were conducted on
the remaining 484 deaths in that year.
Office for National Statistics
26
Doctor
Registrar
17
1
19
1
24
1
11
72
0
8
72
0
5
70
0
100
100
100
566,994
558,313
560,135
Not formally referred
Certified by coroner, not referred
Certified by doctor, not referred
Uncertified and not referred
Number of deaths
Source: ARV DH1 1987 (table 9), 1992 (table 9), 1996 (table 21).
interests, but in providing a service for the investigation of both the
cause of and the circumstances surrounding deaths. Thereafter the
coroner’s interest in medical causes of death grew with the need for
more precise information on mortality. A later act, of 1926, empowered
the coroner to order an autopsy without having to proceed to an inquest,
and to adjourn inquests where someone had been charged with an
offence related to the death.19 A review of the law and practice on
medical certificates of cause of death, and of coronial practice and the
reporting of deaths to coroners, produced some wide ranging
recommendations in 1971 20 but few of these have been brought into
operation, although a recent act consolidated other changes in the
intervening years. 21 Comparable procedures for reporting deaths
requiring further investigation for Scotland and for Northern Ireland are
shown in Box Two.
The main duty of a coroner is to enquire into certain deaths occurring
in his district. 22 His only other significant function is to conduct
inquests into treasure trove. Each coroner is appointed by a local
authority, such as a county council, metropolitan district or London
borough.23 A coroner will be either an experienced barrister or
solicitor, or a legally qualified medical practitioner of at least five
years’ standing. 24
Although there are formal requirements for referral to coroners, doctors
are also encouraged to seek informal advice from coroners in cases
where there is some doubt about the need to refer. There are no reliable
statistics on the number of informal referrals. In 1996, 30 per cent of
deaths were formally referred to coroners, of which nearly all were
from doctors (Table 2). The legal responsibility for referral which is
placed on registrars is usually only a backstop, generating about 4 per
cent of all referrals. A further 18 per cent come from other sources,
such as the police. Referrals by the doctor have increased since 1987,
from 60 per cent of referrals to 80 per cent in 1996.
In many cases, referral to a coroner does not mean that he will certify
the death (Table 3). For 27 per cent of formal referrals in 1996 the
coroner saw no need to conduct any further investigation, and accepted
the cause of death given by the attending doctor. The proportion so
accepted has nearly doubled in recent years. The remaining 73 per cent
of referred deaths almost always involved a post-mortem investigation.
Post-mortems may be carried out by pathologists for coroner enquiries,
or to assist doctors in certifying cause of death. Most (93 per cent in
1992) are for coroners – see Table 4. The remainder form only 1 in 50
of all deaths certified by doctors; figures later than 1992 are not
available at present.
Health Statistics Quar terly 01
Doctors can certify deaths only where they know the cause. They can
request a post-mortem with the next of kin’s consent. Figure 3 shows
that in 1992 post-mortems for deaths certified by doctors were most
common for infant deaths.
I NVESTIGATION OF CAUSE OF DEATH BY POST-MORTEM
AND INQUEST
The proportion of deaths certified by doctors which have been subject
to autopsy has been extremely low for many years (Table 4). This may
be explained in part by higher levels of diagnostic certainty related to
modern ante–mortem investigative techniques. At the same time this
means that there is decreasing pathological audit of the validity of
clinical death certification. A recent study of certificates completed by
various grades of hospital clinicians, GPs and pathologists found that
senior hospital doctors make more errors than their juniors, while GPs
and pathologists make fewest errors.27 This agreed with other studies
showing that inaccuracies in death certification arise from inadequate
formulation of cause of death and failure to record relevant information.
Table 3
Spring 1999
Coroners certify nearly a quarter of all deaths in England and Wales but
most of these cases involve neither an inquest nor any suspicion of
violence. They are referred to the coroner because they were sudden
and unexpected, because there was no doctor in attendance during the
deceased’s last illness, or because the doctor who did attend is not
available to give a certificate – for example, the patient’s general
practitioner may be on holiday and s/he has not consulted another
doctor. In these circumstances, if post-mortem examination establishes
a clear natural cause of death the coroner need not hold an inquest. S/he
Deaths referred to a coroner, and subsequent action, England and Wales, 1987 to 1996
Total deaths
Certification and action after referral
1987
(%)
1992
(%)
Referred deaths
1996
(%)
1987
(%)
1992
(%)
18
4
71
14
0
67
13
0
1996
(%)
Referred to coroner
Certified by coroner
No inquest, with post-mortem
Inquest, with post-mortem
Inquest, without post-mortem
Certified by doctor
Total referred to coroner
Not referred to coroner
Number of deaths
20
4
0
19
4
0
4
6
8
15
20
27
28
28
30
100
100
100
72
100
72
100
70
100
566,994
558,313
569,683
158,463
156,899
169,944
}
}
61
12
Source: ARV DH1 1987 (table 9), 1992 (table 9), 1996 (table 21).
Table 4
Certification of deaths by whether post-mortem carried out, England and Wales, 1987 and 1992
1987
Certification
Total
deaths
(%)
By doctor, with post-mortem
By coroner, with post-mortem
Certified with post-mortem
2
24
26
By doctor, without post-mortem
Total certified
74
Other (uncertified)
Total deaths
Number of deaths
Deaths with
postmortem
(%)
9
91
100
1992
Deaths
certified by
doctors
(%)
Total
deaths
(%)
3
2
22
24
97
100
76
0
0
100
100
566,994
148,643
431,783
558,313
Deaths with
postmortem
(%)
Deaths
certified by
doctors
(%)
7
93
100
2
98
100
134,826
432,212
Source: ARV DH1 1987 (table 9) and 1992 (table 9).
27
Office for National Statistics
Health Statistics Quar terly 01
Figure 3
Spring 1999
Percentage of deaths by underlying cause which are certified by doctors with post-mortem examination,
England and Wales, 1987 and 1992
ICD9 cause chapter
Deaths in
1992
I
Infectious and parasitic diseases
(2,091)
II
Neoplasms
III
Endocrine, nutritional and metabolic
diseases, and immunity disorders
1992
IV
(137,822)
1987
(9,797)
Diseases of the blood-forming organs (2,244)
V
Mental disorders
(12,103)
VI
Diseases of the nervous system
and sense organs
(10,267)
VII
Diseases of the circulatory system (176,414)
VIII
Diseases of the respiratory system
(51,037)
IX
Diseases of the digestive system
(11,903)
X
Diseases of the genitourinary system (4,460)
XI
Complications of pregnancy, childbirth
and the puerperium
XIl
Diseases of the skin and
subcutaneous tissue
XIlI
Diseases of the musculoskeletal
system and connective tissue
(829)
(4,077)
XIV Congenital anomalies
XV
(1,019)
Certain conditions originating in
the perinatal period
(225)
XVI Symptoms, signs and ill-defined
conditions
(4,341)
EXVIl External causes of injury
and poisoning
All deaths
(14)
(881)
(432,212)
0
10
20
30
Percentage of deaths certified by doctors, with post-mortem
Source: ARV DH1 1987 (table 10), 1992 (table 10).
may certify the cause of death based on the pathologist’s post-mortem
findings. About 40 per cent of deaths from ischaemic heart disease,
which may occur suddenly in people without previous symptoms, are
certified by coroners in this way (Figure 4). In contrast only about 5 per
cent of cancer deaths, where the course of the disease is often long and
diagnosis is usually confirmed by biopsy or other tests before death, are
certified by coroners. Only 2 per cent of cancer deaths are subject to
inquest, usually to investigate possible industrial causes.
The causes of death which are most often certified after post-mortem
and/or by coroners reflect the necessity for investigation for legal
reasons or because the cause is unknown (Figure 4). Thus, 90 per cent
of deaths from external causes in 1996 were subject to post-mortem and
inquest before certification by coroners. Virtually the only exceptions
were deaths due to falls and fractures, mainly of the elderly (60 per cent
with coroners inquest, 9 per cent with coroners post-mortem only and
31 per cent by doctors).
In contrast, only 10 per cent of deaths due to mental disorders (largely
senile dementia) have had post-mortems. This partly reflects the long
course of such illnesses, but also what causes can be found at autopsy.
Office for National Statistics
28
Thus, though other research has shown that in 90 per cent of deaths
from suicide a history of mental illness can be established,28 only 12 per
cent of suicide certificates in 1996 mentioned mental disorders. All
these suicide deaths are subject to post-mortem as well as inquest.
Pathologists carrying out post-mortems for coroners may not have
access to any information on the medical history of the deceased.29 Any
history they are given is more likely to come from the police than a
doctor. The primary purpose of such examinations is legal – to rule out
accidents, violence or unnatural causes. It is not to test the validity of
clinical diagnosis nor to produce accurate statistics of causes of death.
The proportion of deaths due to signs, symptoms and ill defined causes
(ICD9 chapter XVI 780–779) which are certified by coroners has fallen
considerably since 1987 (see Figure 4). Instructions in books of
certificates were changed in 1986 and now state that ‘in deaths in the
elderly when no specific condition is identified as the patient gradually
deteriorates and dies, “old age” or “senility” is acceptable as the sole
cause of death for persons aged 70 and over’. The number of deaths
certified as due to ‘old age’ (ICD9 797) has risen steadily since 1986 to
more than 8,500 in 1996. Most of these are at ages over 85, and are
Health Statistics Quar terly 01
Figure 4
Spring 1999
Percentage of deaths by underlying cause which are certified by coroners with post-mortem examination,
England and Wales, 1987, 1992 and 1996
ICD9 cause chapter
Deaths in
1996
I
Infectious and parasitic diseases
II
Neoplasms
III
Endocrine, nutritional and metabolic
diseases, and immunity disorders
(7,502)
1992
Diseases of the blood-forming
organs
(1,986)
1987
V
Mental disorders
(9,296)
VI
Diseases of the nervous system
and sense organs
(9,772)
IV
(3,636)
(139,459)
VII
Diseases of the circulatory system (237,669)
VIII
Diseases of the respiratory system
(88,630)
IX
Diseases of the digestive system
(19,846)
X
Diseases of the genitourinary system (6,752)
XI
Complications of pregnancy, childbirth
and the puerperium
(41)
XIl
Diseases of the skin and
subcutaneous tissue
(1,075)
XIlI
Diseases of the musculoskeletal
system and connective tissue
(3,517)
XIV Congenital anomalies
XV
Certain conditions originating in
the perinatal period
(1,227)
(149)
XVI Symptoms, signs and ill-defined
conditions
(10,772)
EXVIl External causes of injury
and poisoning
(16,061)
All deaths
1996
(560,135)
0
20
Source: ARV DH1 1987 (table 10), 1992 (table 10), 1996 (table 22).
certified by doctors without post-mortem. Nearly all deaths from other
causes in this ICD chapter, such as sudden infant death syndrome (SIDS),
or sudden adult death, or where no cause could be ascertained, are still
certified by coroners. Only in the case of SIDS are these deaths usually
certified without holding an inquest in addition to post-mortem.30
The type of death referred to a coroner may alter over time with a
changing perception of what is natural or unnatural, and with whether
some legal responsibility for the death may fall on a third party. For
instance, deaths from Creutzfeldt-Jakob disease (CJD) were rarely
certified by coroners before 1990 unless there was a history of treatment
with hormones derived from pooled human pituitaries. However, in the
past few years CJD deaths have been increasingly subject to inquest,
which may lead to long delays in their registration.31 All such cases
should have been referred to the United Kingdom CJD surveillance unit
in Edinburgh, usually before death, irrespective of how the death is
eventually certified and registered. Up-to-date figures on presumed and
confirmed cases and deaths from sporadic, iatrogenic and new-variant
CJD are now available from the UK CJD surveillance unit.
40
60
80
100
Percentage of deaths certified by coroner
SELECTING THE UNDERLYING CAUSE OF DEATH
Mortality statistics, including those referred to above, are usually based on a
single cause per death. This is the ‘underlying cause of death’, defined by the
World Health Organisation (WHO) as ‘(a) the disease or injury which
initiated the train of events directly leading to death, or (b) the circumstances
of the accident or violence which produced the fatal injury’.32 This is
generally the most useful single cause for public health purposes.
The medical certificate of cause of death in England and Wales has been in
the format recommended by WHO since 1927. In part I, which has three
lines, the certifier is asked to state the conditions leading directly to death,
starting with the immediate cause on line Ia and going back through the
sequence on subsequent lines. Part II is for other conditions which
contributed to the death but were not part of the direct causal sequence.33 If
the death certificate has been properly completed, with only one condition
on each line, and the conditions in part I forming an acceptable sequence,
the general rule can normally be used to select the condition entered in the
lowest completed line of part I as the underlying cause (UCD).
29
Office for National Statistics
Health Statistics Quar terly 01
Table 5
Spring 1999
Deaths by type of certificate, number of causes per
line, and determination of underlying cause,
England and Wales, 1986 and 1996
Type of certificate
Percentage of total deaths Percentage where underlying
cause is derived by
general rule
Simple certificate
(one cause per line)
Line Ia only
Lines Ia and II
Lines Ia and Ib
Lines Ia, Ib and II
Lines Ia, Ib and Ic
Lines Ia, Ib, Ic and II
Complex certificate
(more than one cause
on at least one of the lines)
Total
1986
1996
1986
1996
20
9
32
8
8
2
25
10
28
10
6
2
100
54
74
62
56
51
100
91
77
72
56
57
22
20
n/k
n/k
100
100
Source: 1986 figures from Ashley and Devis paper, table 7; 1996 figures from unpublished tables.
If the death certificate has not been completed correctly, it becomes
necessary to apply one or more of the three selection rules in the Ninth
Revision of the International Classification of Diseases (ICD9).34 In
addition, there are nine modification rules which apply to particular
conditions, combinations or circumstances. For example, two or more
mentioned conditions may be linked to derive a composite underlying
cause, so that renal failure (ICD9 code 586) due to hypertension (401)
becomes hypertensive renal disease (403). Later stages of the same
disease process are preferred to earlier ones – acute myocardial
infarction (410) rather than ischaemic heart disease (414); and trivial or
ill defined conditions, or senility, may be ignored when this allows the
coder to select a more specific or lethal condition. The purpose of these
rules is to derive the most useful information from the death certificate,
even when it has been badly completed, and to do this uniformly so that
data will be comparable between places and times.
Table 5 shows clearly that the proportion of deaths for which the
underlying cause was selected using the general rule is higher in 1996
than it was in 1986 – that is, ONS is now following international practice
more closely, and overturning the opinion of the certifier less often than
ten years ago. The greatest increase, in both relative and absolute terms,
in the use of the general rule is on certificates which have only a single
condition in part I on line a, and one or more conditions in part II. This
probably reflects changes in the application of ICD9 rule 3 over time.
Table 6
ICD9 Rule 3
Rule 3 as published by WHO states that ‘if the condition selected by the
general rule, or rules 1 or 2, can be considered a direct sequel of
another reported condition, whether in part I or part II, select this
primary condition. . .’35 For example, if the certifier has written ‘renal
failure due to obstructive uropathy’ in part I and ‘benign prostatic
hypertrophy (BPH)’ in part II, the coder would assume the BPH to be
the condition which initiated the sequence leading to the renal failure.
In 1984 OPCS introduced a much broader interpretation of rule 3. This
happened because an increasing proportion of certificates were noted to
have in part I only a condition which was considered likely to be the
terminal event in a person made immobile, disabled or ill by other
conditions present for a longer time, which the certifier placed in part
II. In particular, there had been very large increases in the proportion of
deaths ascribed to bronchopneumonia, especially in the elderly.
The 11 conditions considered terminal included pneumonia (of any
type), pulmonary embolism, venous thrombosis and embolism, cardiac
or hepatic failure, and cardiac arrest.36 In the broader OPCS
interpretation of 1984, when any of these conditions would have been
the underlying cause, and any other major condition was recorded in
part II, rule 3 was to be used to select that major condition. No evidence
for an aetiolgical or pathological sequence was required: the
presumption was simply that these 11 conditions could be the ‘terminal
event’ in any disease process.
This meant, for instance, that schizophrenia or rheumatoid arthritis in
part II could be selected as the underlying cause of renal failure or
pneumonia in part I. This change led to an abrupt fall in deaths
ascribed to pneumonia and the ten other less common conditions
regarded as terminal, and a corresponding apparent rise in deaths from
many predominantly chronic conditions, including mental disorders.
Deaths registered in 1984 were independently coded according to both
the old international, and the OPCS only rules. The results of this
bridge coding were compared to calculate conversion ratios for every
cause affected, so that time trends could be interpreted across the
change. The broader interpretation of rule 3 continued for all deaths
registered in England and Wales from 1984 until the automated
system for coding cause of death (ACCS) was introduced in OPCS in
1993.37 This incorporates software developed in the USA, and so
applies the American interpretation of the WHO rules – which are
probably closer to those used internationally. A description of the
effects of moving back to the internationally accepted interpretation
of rule 3 has been published by ONS.38
Multiple cause of death data, England and Wales, 1986 and 1996
1996 deaths
ICD9
code
Medical condition
162
174
250
290
295
410
428
436
485
785
malignant neoplasm of lung
malignant neoplasm of female breast
diabetes
organic psychoses
schizophrenia
acute myocardial infarction
heart failure
acute cerebrovascular disease
bronchopneumonia
cardiovascular symptoms
Source: ARV DH2 1985; unpublished 1996 data.
Office for National Statistics
30
Underlying cause
Mentioned causes
30,810
12,179
5,994
6,592
36
68,356
9,753
39,273
45,644
40
32,807
15,084
23,682
15,191
534
73,852
73,748
52,279
99,115
6,121
1986 deaths
Mentions/underlying cause
1.06
1.24
3.95
2.30
14.8
1.08
7.56
1.33
2.17
153
Mentions/underlying cause
1.05
1.15
2.90
1.60
3.07
1.07
13.5
1.20
5.34
169
Health Statistics Quar terly 01
MULTIPLE CAUSES
OF DEATH
Multiple cause data, in which all the conditions mentioned by the
certifier are coded, are available for 1985 and 1986 and all years
since 1993. The data can be used to elucidate the effects of changes
in selecting the underlying cause of death (UCD) on routine
mortality statistics. Analyses of multiple cause data help to show the
contribution of particular diseases in the death process, as well as
allowing some measurement of associations between diseases. Table
6 shows that some conditions are nearly always selected as the
underlying cause of death whenever they appear on the certificate.
These include major cancers and acute cardiovascular events,
including myocardial infarction and stroke. In contrast, there are
nearly four times as many deaths with diabetes mentioned as have it
selected as the underlying cause. Symptoms, signs and ill defined
conditions such as cardiorespiratory arrest are often given as the
immediate cause, but hardly ever selected as the underlying cause,
unless there is no other choice.
The ratio of mentions to UCD gives an indication of how much the
UCD statistics could be inflated by changes in the selection rules. There
is comparatively little scope for artefactual increases in rates of death
from acute myocardial infarction, which is nearly always selected as the
underlying cause of death whenever it appears. The same is true of
malignant neoplasm of the lung. This largely reflects the views of
certifiers that these conditions are directly lethal. The ratio of mentions
to UCD is a little higher for breast cancer, reflecting the better survival,
so that this condition may appear in part II of the certificate in elderly
women dying of other conditions.
Much higher ratios are seen in conditions such as heart failure, which
may be caused by a variety of pathologies; diabetes, which may be
placed in part II when it complicates treatment of other conditions and
so compromises survival; chronic degenerative conditions such as
organic psychoses (these deaths are mostly senile dementia), which
certifiers appear not to regard as lethal; terminal conditions such as
bronchopneumonia; and ill defined descriptions of symptoms and signs.
Changes in the proportion of any of these conditions selected from the
large pool of ‘mentions’ can have very large effects on their apparent
underlying cause mortality rates.
Differences in the ratios of mentions to underlying cause between
the mid–1980s and the 1990s shed further light on the changes in
UCD selection when automated coding was introduced in 1993. The
higher ratio of mentions/UCD for diabetes and organic psychoses in
the 1990s shows that the reversion to the international interpretation
of rule 3 has made selection of these conditions as the underlying
cause less likely now. In contrast, bronchopneumonia is two and a
half times as likely to be selected as UCD now as it was ten years
ago. Heart failure and cardiovascular symptoms are also more likely
to be selected now.
ENQUIRIES
As noted above, the conditions mentioned on the death certificate are
used to derive an underlying cause of death. However, in some cases
more information on causes of death may become available later, such
that the underlying cause may be subsequently amended. At present, the
ways in which this may happen are:
•
•
•
by the certifying doctor indicating on the front of the medical
certificate that information from a post-mortem may be available later;
by the certifying doctor indicating on the back of the MCCD (in
Box B) that more information may be available later (for example,
results of bacteriology tests taken before death);
by coroners certifying deaths after post-mortem but without
•
•
Spring 1999
inquest, indicating that results of histology or bacteriology tests
may be available later;
following an inquest, the second page of the coroners certificate –
which provides details of how a fatal accident occurred – is posted
to ONS by the registrar, and may arrive later than the electronic
registration;
where an inquest has been adjourned and an accelerated registration
carried out, the coroner may later provide a final underlying cause
and verdict.
In the first three instances, the registrar is obliged to send a request for this
further information at the same time as registering the death. In the last two
instances, the deaths processing branch in ONS may send an enquiry if the
expected information has not arrived after a suitable interval.
In 1996 there were some 14 thousand deaths where Box B on the
MCCD was ticked, and from subsequent enquiry the underlying cause
was amended for about 20 per cent of these. These enquiries are open–
ended and their effect is limited – few medical conditions are affected,
and many of the amendments result in changes only within the same
three–digit ICD code.
As well as these five sources of further information ‘medical enquiries’
(MEs) have been made by the Registrar General’s Office in most years
since 1881.39 These enquiries were sent to certifiers if insufficient details
had been provided on the certificate for coding the cause of death precisely.
Coders used a standard list of ICD codes which would generate a medical
enquiry if selected as the original underlying cause on a death certificate.
No follow–up was sent if the ME produced no reply, and enquiries were
not normally sent for deaths at age 75 or over. An example of an enquiry is
where the death certificate was due to ‘malignant melanoma, site
unspecified’ – the certifier would be asked for the exact site of the primary
growth, if known.40 An enquiry form was also used for neonatal deaths.
OPCS used this system up to 1992, when it was abandoned as it was not
possible to deal with these deaths in a timely way, nor to generate enquiries
easily from the (new) automated coding system. In any case, studies of the
effects of enquiries on mortality statistics in the early 1990s suggested that
the system needed review. This was one of many changes which took place
in 1993 at the time of redevelopment. Not using medical enquiries meant
that from 1993 the numbers of deaths assigned to less specific causes
increased, while the number from more specific causes decreased. Many
conditions where medical enquiries produced this ‘loss’ of numbers are
those with less well-defined descriptions. Examples are: malignant
neoplasm of other and ill-defined sites (ICD 195), malignant neoplasm
without specification of site (ICD 199), and other diseases of respiratory
system (ICD 519). Complementing these are conditions where numbers
show a marked net increase, such as malignant neoplasm of small intestine
(ICD 152) and bacterial meningitis (ICD 320). For most conditions the
effects were small, but and these have been analysed in some detail.41
CONCLUSIONS
The process of death registration in England and Wales is a very complex
one. It is governed by a variety of laws and regulations which have
accumulated over more than a century and a half. Information gathered
through these processes is used for public health planning and evaluation,
resource allocation, epidemiological and clinical research and a variety of
other purposes. These purposes are not all necessarily well served by the
current system for investigating and certifying cause of death.
Doctors have little training in how to complete medical certificates of
cause of death, and uncertain knowledge of their legal obligations. Many
are unaware of the uses to which the information they give will be put,
including the policy decisions which may ultimately affect their own
practice. The key issues for certifiers are the identification of a sequence
31
Office for National Statistics
Health Statistics Quar terly 01
Spring 1999
of events (where present), together with a clear understanding of the
nature of the underlying cause of death. These could be addressed by
better training in certification of medical students and junior doctors, and
in the uses of cause of death data. With this appreciation many doctors
might take more care in filling out MCCDs, in particular providing as
much relevant information as possible, on the standard certificate. A
video was produced by ONS in 1995 to assist in training.
Investigation of the cause of a quarter of all deaths, the vast majority of
which are due to disease, is left to coroners, whose purpose is to
prevent homicide going undetected, and to provide a final check on
other legislation affecting public safety such as industrial health and
safety. This is not a system designed to provide reliable, uniform, good
quality information on deaths from diseases for public health purposes
or for medical and epidemiological research. In addition, coroners
certify many deaths from natural causes for purely administrative
reasons: there is no doctor in attendance available to give a certificate.
Alternatives are possible. For example, investigation of sudden and
unexpected deaths in Sweden is the responsibility of a local public
health official, who may consult medical personnel and records before
deciding whether autopsy or other investigation is required.42
Many data users have little knowledge of the processes involved in
producing mortality data, or of the effects these have on the accuracy,
comparability and reliability of the end information. This paper
provides some information, and reference to other sources, to help
interpret routine statistics and research. Further analyses of multiple
cause data will appear in forthcoming ONS publications. These data
will be made available to researchers, on request.
ACKNOWLEDGEMENTS
We thank those colleagues in ONS who assisted in preparing this paper,
and acknowledge also the help and useful comments of Graham
Jackson (GRO(S)), Lawrence McKeag and John Gordon (both of
GRO(NI)) and Dr John Ashley (formerly of OPCS).
REFERENCES
1
2
3
4
5
6
7
8
9
There are also possible improvements for ONS to consider. A revised
system for medical enquiries which makes a significant improvement to
information on cause of death, is being considered for implementation
in the near future, in conjunction with the move to ICD10 coding. More
fundamental would be an overhaul of the legislation governing death
certification, investigation and registration, and of restrictions
surrounding release for medical and epidemiological research of data
collected at death registration.
10
11
12
13
14
15
Key findings
16
17
• Certification and registration is straightforward for
70 per cent of deaths, requiring a doctor to certify
the cause(s) and a registrar to enter these and
other details in a register.
• In about 30 per cent of cases the death is referred
to a coroner, and in about 4 per cent an inquest is
subsequently held. Most of the inquest deaths are
from accidental or violent causes.
• From 1993 ONS has used cause coding rules more
in line with international practice, enabling greater
consistency and easier comparison.
• It is now possible to analyse all conditions mentioned
on death certificates, leading to a better understanding
of particular diseases in the death process.
18
19
20
21
22
23
24
25
26
27
28
Office for National Statistics
32
Ashley J and Devis T (1992). ‘Death certification from the point of
view of the epidemiologist’. Population Trends 67, pp. 22–28.
TSO, London.
Births and Deaths Registration Act 1836.
OPCS (1980). Population and Health Statistics in England and
Wales.
Ashley J, Cole S K and Kilbane M P J (1991). ‘Health information
resources: UK – health and social factors’, in Holland W W (ed),
Oxford Textbook of Public Health, volume 2, pp. 29–53. Oxford
University Press, Oxford.
World Health Organisation (1977). Manual of the International
Statistical Classification of Diseases, Injuries and Causes of Death,
Ninth Revision, volume 1. WHO, Geneva.
General Register Office (1929). Registrar General’s Statistical
Review 1927, p. 145. HMSO, London.
Report of the Committee on Death Certification and Coroners
(Brodrick Report) (1971), p 367. Cmnd 4810 . HMSO, London.
Schedules to the Population (Statistics) Act 1938, and the
Population (Statistics) Act 1960.
OPCS (1996). Mortality statistics: cause 1993 (revised) and 1994,
series DH2 no 21, section 3. HMSO, London.
Rooney C and Devis T (1996). ‘Mortality trends by cause of death
in England and Wales 1980–94: the impact of introducing
automated cause coding and related changes in 1993’. Population
Trends 86, pp. 29–35. TSO, London.
ONS (1998). Mortality statistics: injury and poisoning 1996, series
DH4 no 21, pp. xxii–xxiii. TSO, London.
Registration of Births and Deaths Regulations 1987, schedule 2.
DH4 no 21 op cit, section 2.10.
Birth and Death Registration Act 1953, section 22.
DH4 no 21 op cit, p. xiv.
ONS (1998). Mortality statistics: general 1996, series DH1 no 29.
TSO, London.
Brodrick Report op cit, p. 107.
Matthews P and Foreman J (1993). Jervis on the Office and Duties
of Coroners, eleventh edition, p. 3. Sweet and Maxwell, London.
Coroners (Amendment) Act 1926.
Brodrick Report op cit.
Coroners Act 1988.
Jervis on Coroners op cit, p 37.
ibid, pp. 14–15.
Jervis on Coroners op cit, p. 18.
Brodrick Report op cit, p. 90 et seq.
Registrar General for Scotland. Annual Report 1996, p. 67.
GRO(Scotland), Edinburgh (1997).
James D S and Bull A D (September 1995). ‘Death certification: is
correct formulation of cause of death related to seniority or
experience?’ J R Coll Physicians Lond 29(5), pp. 424–428.
Barraclough B et al (1974). ‘A hundred cases of suicide: clinical
aspects’. British Journal of Psychiatry 125(0), pp. 355–373.
Health Statistics Quar terly 01
29
30
31
32
33
34
35
36
37
Personal communication from Dr G Mandsley (1998).
ONS (1997). Mortality statistics: general 1993–95, series DH1 no
28, table 45. HMSO, London.
Devis T and Rooney C (1997). ‘The time taken to register a death’.
Population Trends 88, pp. 48–55. TSO, London.
WHO Manual op cit, p. 763.
ibid p. 700.
ibid pp. 703–712.
ibid p. 705.
OPCS (1985). Mortality statistics: cause 1984, series DH2 no 11,
pp. v–ix. HMSO, London.
Rooney C and Devis T. Population Trends 86 op cit.
38
39
40
41
42
Spring 1999
DH2 no 21 op cit, p. xxv–xxvii.
Swerdlow A J (1989). ‘Interpretation of England and Wales cancer
mortality data: the effect of enquiries to certifiers for further
information’. British Journal of Cancer 59, pp. 787–791.
DH2 no 21 op cit, p. xii–xxiv.
Peters K D et al (in press). ‘Results of the ICE questionnaire on
registration and coding practices’. In Proceedings of the
International Collaborative Effort on Automating Morality Statistics,
eds Peters K D and Rosenberg H M, Hyattsville.
ONS (1998). Mortality statistics: cause 1997, series DH2 no 24.
TSO, London.
33
Office for National Statistics
Health Statistics Quar terl y 01
Spring 1999
Trends in mortality of young
adults aged 15 to 44 in
England and Wales
Paul Aylin, Imperial College
School of Medicine at St. Mary’s,
Karen Dunnell and
Frances Drever, ONS
This article updates a previous
analysis investigating the
reasons for the levelling of
mortality in young adults since
the mid–1980s. Among men
increases in deaths from AIDS,
suicides and deaths related to
drugs are the main reasons.
Teenagers’ mortality has
improved consistently whereas
that for men aged 40–44 has
ceased to improve. The
situation for women is more
complex with increases from
infections, mental disorders
and digestive diseases.
INTRODUCTION
Dunnell1 observed that death rates for most groups of adults have
been declining throughout the century, but that between the middle
and end of the 1980s, death rates among young men and women aged
15 to 44 stopped declining. This was still the case in the middle of the
1990s. Although mortality in the age group 15 to 44 accounted for
only 3 per cent of all deaths in 1996, these deaths are important
because of their contribution to years of potential life lost. Over one
third of deaths in this age range are from external causes of injury and
poisoning and so could be considered to be avoidable. This article
looks at trends in mortality in the 15 to 44 age group between 1979
and 1996. For five-year age groups of men and women, the actual
numbers of deaths in 1994–96 from the main causes are compared
with the expected numbers based on the 1986–88 rates.2 Patterns in
various causes of death are described.
METHODS
Mortality statistics for England and Wales are compiled by the Office
for National Statistics from death registrations. Analyses reported here
use data from the CD-ROM, Twentieth Century Mortality,3 an
aggregated dataset containing year, age, sex, populations and
underlying cause of death coded to the ninth revision of the
International Classification of Diseases (ICD9).4 Data for the year 1996
were obtained directly from the databases at ONS. Deaths and
populations for the years 1979 to 1996 were used. Mortality data for
years 1979 to 1992 are based on date of registration. Those for years
1993 to 1996 are based on date of occurrence.5 Age-specific mortality
rates were calculated using three-year moving averages. Figures in this
article have these rates plotted against the middle year, for example the
rate for 1979–81 is plotted at 1980. Similarly, in the text the three-year
Office for National Statistics
34
Health Statistics Quar terly 01
death rates are assigned to the middle year, for example the statement ‘a
death rate of 325 per million in 1980’ really means the three-year death
rate over the period 1979 to 1981 was 325 per million.
RESULTS
All cause mortality
Mortality has been decreasing steadily from 1979. The age-standardised
rate for males in 1996 was 72 per cent of the 1979 rate. For females the
decrease was less – 76 per cent.
Figure 1
‘Index’ of age-specific mortality rates, England
and Wales, males (3- year moving averages)
Spring 1999
However, the mortality rates in different age groups have not all seen
this steady fall.
Figures 1 and 2 show the changes for different age groups for males
and females separately. The age-specific rates of 1980 are taken to be
100. Childhood mortality (ages under 15) has improved the most. The
mortality rate for boys and girls in the mid-1990s was about one half
the rate in 1980. For women, older adult age groups show similar
patterns of decrease. Those aged 15 to 44 show faster declines in the
early 1980s but there now seems to be a plateau in mortality rates for
Figure 2
‘Index’ of age-specific mortality rates, England
and Wales, females (3- year moving averages)
Index value
(1979–81 =100)
Index value
(1979–81 =100)
100
100
15–44
90
90
75+
75+
15–44
80
80
45–74
45–74
70
70
60
60
Under 15
Under 15
50
50
40
40
1979
Table 1
1981
1983
1985
1987
Year
1989
1991
1993
1979
1995
1981
1983
1985
1987
Year
1989
1991
1993
1995
Age-specific death rates per million population, men and women aged 15 to 44, England and Wales, 1979–96
Age group
Years
15–19
20–24
25–29
30–34
35–39
40–44
15–44
Rate per million
Men
1979–81
1982–84
1985–87
1988–90
1991–93
1994–96
833
746
697
707
640
572
873
840
825
869
838
839
861
825
778
839
835
879
962
955
957
976
989
1,030
1,405
1,293
1,298
1,351
1,395
1,315
2,451
2,193
2,122
2,013
2,014
2,018
1,231
1,148
1,113
1,126
1,118
1,109
Women
1979–81
1982–84
1985–87
1988–90
1991–93
1994–96
325
298
286
297
283
265
367
324
314
317
323
303
449
419
372
364
364
374
609
582
576
542
519
521
969
894
871
865
804
799
1,658
1,480
1,421
1,373
1,310
1,317
730
666
640
626
601
596
34
Office f or National Statistics
Health Statistics Quar terl y 01
Figure 3
Spring 1999
Percentage changes in death rates among
men aged 15–44, England and Wales, 1979–96
Figure 4
Percentage changes in death rates among
women aged 15–44, England and Wales,
1979–96
30–34
25–29
100
100
20–24
35–39
90
90
15–19
30–34
20–24
35–39
40–44
80
80
25–29
40–44
70
70
15–19
60
1979–81
Table 2
1982–84
1985–87
1988–90
1991–93
1994–96
60
1979–81
1982–84
1985–87
1988–90
1991–93
1994–96
Percentage distribution of main causes of death for men aged 15–44, England and Wales, 1994–96
Age group
Cause of death
Infectious diseases
including AIDS/HIV
Neoplasms
Lung
Leukaemia, etc .
Mental disorder
Nervous system
Circulatory diseases
Ischaemic heart disease
Stroke
Respiratory diseases
Digestive diseases
External causes
Accidents
Suicides, etc.
Other causes
ICD code
15–19
(%)
001-139
140-239
162
200-208
290-319
320-389
390-459
410-414
430-438
460-519
520-579
E800-E999
E800-E949
E950-E959,
E980-E989
except E988.8
001-999
Number of deaths = 100%
- 0 per cent.
* Less than 0.5 per cent.
Office for National Statistics
36
20–24
(%)
25–29
(%)
30–34
(%)
35–39
(%)
40–44
(%)
2
2
4
8
7
4
9
4
5
3
5
*
1
3
1
61
44
8
3
6
3
5
1
1
3
1
66
37
10
*
4
5
4
7
2
2
4
2
57
27
12
1
4
4
6
11
4
2
6
4
45
20
17
2
4
3
6
19
11
3
6
7
32
14
24
4
3
2
6
27
19
4
6
8
22
9
13
12
25
6
26
6
22
4
16
3
11
2
2,678
4,492
5,524
6,648
7,287
10,205
Health Statistics Quar terly 01
this group of women. Except for the 15 to 44 age group, each age group
of men shows a steady decrease in mortality rates. The 15 to 44 group
had declining mortality rates between 1980 and 1986. Between 1987
and 1994, there was a small increase – the age-specific rate in 1995
being roughly the same as in 1986.
As the earlier analysis 1 concluded, ‘the age group 15–44 is not a
homogeneous group when it comes to explaining changes in death
rates’. Table 1 therefore shows the age-specific death rates for five-year
age groups of men and women. Figures 3 and 4 use the rates in Table 1
expressed as percentage changes, taking 1979–81 as 100 per cent.
Table 3
Spring 1999
Among women, there has been a consistent decrease in rates among
those aged 35–39 and 40–44. The improvement in mortality has been
greatest for the oldest women in the group, those aged 40–44, where
the mortality rate in 1995 was only 79 per cent of that in 1980. This is
similar to the improvement among women in the 45–74 year age
group as shown in Figure 2. Mortality has improved least among
women aged 30–34. They and their sisters aged 25–29 experienced a
rise in mortality during the 1990s. The youngest women in the group,
those aged 15–19 and 20–24, however, experienced continuing
improvement in the 1990s following a period of little change in the
late 1980s (Figure 4).
Percentage distribution of main causes of death for women aged 15–44, 1994–96, England and Wales
Age group
Cause of death
ICD code
Infectious diseases
Neoplasms
Breast
Cervix
Lung
Lymphomas
Leukaemia, etc.
Mental disorder
Nervous system
Circulatory diseases
Ishaemic heart disease
Stroke
Respiratory diseases
Digestive diseases
External causes
Accidents
Suicides, etc .
Other causes
15–19
20–24
25–29
30–34
35–39
40–44
(%)
(%)
(%)
(%)
(%)
(%)
001-139
140-239
174
180
162
6
16
*
4
17
*
1
*
5
22
4
3
*
4
34
11
5
1
3
44
16
6
2
2
51
19
4
4
200-208
290-319
320-389
390-459
410-414
430-438
460-519
520-579
E800-E999
E800-E949
E950-E959,
E980-E989
except E988.8
001-999
6
3
8
6
*
2
5
1
43
30
9
3
8
9
*
3
4
2
41
22
9
3
6
12
1
4
6
3
32
15
7
2
6
13
2
4
5
5
22
10
6
1
4
14
4
6
5
6
16
7
5
1
4
16
5
6
4
7
11
5
10
8
14
8
14
8
10
7
7
5
5
4
1,169
1,544
2,250
3,232
4,347
6,624
Number of deaths = 100%
- 0 per cent.
* Less than 0.5 per cent.
Table 4
Changes in numbers of deaths, men, England and Wales
Age group
Cause of death
ICD code
15–19
20–24
25–29
30–34
35–39
40–44
1994–96 deaths compared with expecteds using 1986–88 rates
Infectious diseases
Neoplasms
Lung
Lymphomas, leukaemia, etc.
Mental disorders
Nervous system
Circulatory diseases
Ischaemic heart disease
Stroke
Respiratory diseases
Digestive diseases
External causes
Accidents
Suicides, etc.
001-139
140-239
162
200-208
290-319
320-389
390-459
410-414
430-438
460-519
520-579
E800-E999
E800-E949
E950-E959,E980-E989
except E988.8
14
-67
-2
-42
23
-10
-19
3
-23
-13
4
-497
-482
32
-49
-5
-24
151
-35
41
4
6
17
2
-160
-332
88
-112
2
-40
164
32
43
0
-15
52
9
212
-120
277
-220
-17
-79
109
13
-227
-224
-59
193
-126
29
-96
265
-233
-75
-26
87
3
-398
-419
-52
171
-224
-28
-60
216
-201
-109
-53
79
35
-939
-927
-19
183
-397
-10
-60
-31
-12
138
30
313
81
105
237
-13
474
57
791
001-999
-578
29
569
285
118
-243
Other causes
All causes
36
Office f or National Statistics
Health Statistics Quar terl y 01
Table 5
Spring 1999
Changes in numbers of deaths, women, England and Wales
Age group
Cause of death
ICD code
15–19
20–24
25–29
30–34
35–39
40–44
1994–96 deaths compared with expecteds using 1986–88 rates
Infectious diseases
Neoplasms
Lung
Breast
Cervix
Lymphomas, leukaemia, etc.
Mental disorders
Nervous system
Circulatory diseases
Ischaemic heart disease
Stroke
Respiratory diseases
Digestive diseases
External causes
Accidents
Suicides, etc.
001-139
140-239
162
174
180
200-208
290-319
320-389
390-459
410-414
430-438
460-519
520-579
E800-E999
E800-E949
E950-E959,E980-E989
except E988.8
Other causes
All causes
001-999
18
1
1
-2
-1
0
5
1
-8
2
-4
-16
-9
-67
-54
26
-19
-1
-6
-4
39
11
19
5
-3
-9
-20
-9
-55
-54
62
-94
-9
-20
-46
71
27
2
43
2
-8
31
-2
-51
-31
102
-345
-16
-60
-130
51
7
2
-33
-17
-55
14
22
-116
-42
73
-448
-32
-112
-138
66
10
-20
-42
-58
-28
42
53
-84
-3
45
-556
-37
-218
-113
74
21
-29
-59
-73
-47
27
140
-66
-44
-4
-47
-12
-29
2
14
-49
1
-78
-282
-19
3
-122
-70
32
-348
-698
-472
Among men, there has been much greater variability between the
five- year age groups. But similar to women, the youngest and
oldest have the most improving death rates. This is particularly
obvious for teenagers, whose mortality has improved throughout the
period. The previous analysis singled out the 40–44 age group as
having improving mortality compared with the increasing rates for
those aged 20–39 during the 1980s. This has now changed and there has
been no improvement in mortality among 40–44 year olds since 1989.
Each five-year age group of men aged 20–39 had an equal or higher
mortality rate in 1995 compared with 1983 (Figure 3).
Causes of death 1994 to 1996
Having looked in some detail at the age specific changes in death
rates as a whole we now turn to specific causes of death among each
five- year age group.
ICD94 divides causes of death up into chapters, then groupings within
chapters and then individual diseases. A list of causes was identified,
being chapters of the ICD, and groupings within chapters, where these
accounted for 5 per cent or more of the overall mortality for at least one
of the age groups within the 15 to 44 band. The percentage distributions
of deaths from these causes are given in Tables 2 and 3 for 1994–96.
Among men aged 15–19, 44 per cent of deaths were due to accidents.
This proportion decreased with age to 9 per cent of men aged 40–44.
For men in their twenties one in four deaths was a suicide. The
proportions were lower for younger and older men. Circulatory and
cancer deaths claimed higher proportions of the total deaths at each
subsequent older age group. So, for example, 24 per cent and 28 per
cent of deaths in 40–44-year-old men were due to cancer and
circulatory disease respectively.
For women the picture is somewhat different. Cancer accounts for 16 per
cent of deaths in the 15–19 age group – rising to 51 per cent in the 40–44
age group. Consequently a smaller proportion of deaths than among men
are accounted for by accidents, suicides and circulatory diseases.
Nevertheless among the youngest age groups, 15–19 and 20–24, 30 per
cent and 22 per cent of deaths respectively were due to accidents.
Office for National Statistics
38
Changes in the causes of death over time
The previous detailed analysis compared deaths in 1987–89 with those
in 1984–86 – a period when mortality in the 15–44 age group appeared
to have levelled. It concluded that in men, decreases in deaths from
accidents, circulatory disease and cancer had been more than offset by
increases in deaths from suicides and AIDS/HIV. Among women the
decreases had been offset by a wide variety of increases from breast and
cervix cancer, digestive and nervous system deaths and suicides.
The previous analysis was repeated to compare deaths in 1994–96 with
those in 1986–88. The age and cause specific rates in 1986–88 and the
populations in 1994–96 were used to calculate expected numbers of deaths
in 1994–96. Tables 4 and 5 summarise the analysis for men and women.
Over the period, there was a change in the coding of certain deaths.5 Where
this is relevant, it is mentioned in the following commentary.
Among men the youngest and the oldest groups, 15–19 and 40–44, both
had fewer than expected deaths in 1994–96. The other age groups all
had higher than expected numbers of deaths. Nevertheless the pattern of
changes was similar for each of the age groups. As in the previous
analysis there were reductions in deaths from cancer, circulatory disease
and accidents. But again these are more than offset for men aged 20–39
by increases in deaths from infectious and parasitic diseases. These
include tuberculosis, viral hepatitis, mumps, and HIV infection. At the
beginning of the 1980s, HIV had not been discovered. It was first
included in the ICD in code 279.1, which is outside this chapter, before
being moved to codes 042–044 in 1993. To be consistent, all deaths in
code 279.1 have been transferred to be included in this chapter in the
analyses here. The major contributor to these extra deaths from
infections is the number of deaths coded to HIV infection. The change
in rules for coding causes of death is the main reason for the increase in
deaths from respiratory disease.
For all age groups of men there was an increase in deaths from mental
disorders. The major contributors to the excesses are those deaths
coded to drug dependence, non-dependent abuse of morphine and
non-dependent alcohol abuse. Not all deaths which involve drugs are
coded to this chapter of the ICD. Trends in these codes are only part of
Health Statistics Quar terly 01
the complete picture of deaths to drug addicts and should be interpreted
with care. ‘For example, where a drug addict dies as a result of an
overdose this may be classified under accidental poisoning (E850-858)
rather than drug dependence (304) if there is no specific mention of
dependence or addiction on the coroner’s certificate’.6 Among men aged
30–44 there has been a substantial decrease in deaths from digestive
disease – these are mainly chronic liver disease due to alcohol. For most
age groups there were increases in suicides and other causes. ‘Other
causes’ require further investigation to identify whether any diseases in
particular are contributing to this increase in mortality.
Table 5 shows the comparable picture for women. Unlike for men,
numbers of deaths in 1994–96 were lower than expected for each age
group except 25–29. As for men, accidents were fewer as were heart
disease, stroke and all cancers. However unlike the 1980s, breast and
cervix cancer mortality decreased in all age groups. But deaths from
lymphomas, leukaemia, etc. increased in all age groups. Infectious
disease deaths were also higher than expected, reflecting the
continuation of HIV/AIDS as a major problem. Digestive diseases
showed above expected deaths among women aged 30–44. These were
mostly related to alcohol. Unlike men, there is no evidence of increases
in suicides or drug-related deaths.
Key findings
●
●
●
●
Mortality among young adults first became an area of concern in the
late 1980s when death rates failed to continue the long-term decline.
Children and older adult death rates continued to decrease. This analysis
in the mid-1990s shows clearly that the pattern has prevailed. Among
women there have been small decreases in death rates between 1992 and
1995 for those aged 15–24, but no decreases for those aged 25–44.
Among men there have been significant improvements for 15–19-year-olds
because of reductions in deaths from accidents. But for older groups there
have been no improvements and some general deterioration since 1986.
In addition the improvement in death rates in the 40–44 age group of
men that continued until 1989 now appears to have halted. There have
been large increases in deaths from ‘other causes’ which need further
investigation. In general among men decreases in deaths from cancer,
circulatory disease, digestive disease and accidents are being offset by
increases in infectious disease deaths, mostly HIV related, mental
disorders, suicides, respiratory diseases and ‘other causes’. For women
there has been a turn around in breast and cervix cancer deaths which
are now decreasing. These are partly offset by increases in deaths from
lymphomas and leukaemias. Other causes which are increasing are
infections, mental disorders, respiratory and digestive diseases.
For most age and sex groups death rates have been
declining year on year.
Those for people aged 15–44 stopped declining or
slowed in the late 1980s.
For some five-year age groups of men, death rates
were higher in 1994–96 than they were in 1988–90.
Decreases in deaths from some cancers, heart disease
and accidents are being offset by increases in HIVrelated deaths, suicides and a range of other causes.
REFERENCES
1
2
3
S UMMARY AND CONCLUSION
Spring 1999
4
5
6
7
8
9
10
Dunnell K (1991). ‘Deaths among 15–44 year olds’. Population
Trends 64, pp. 38–43. HMSO, London.
Department of Health (1993). ‘Vital Statistics’. On the State of the
Public Health 1992, pp. 36–7. HMSO, London.
Office for National Statistics (1997). Twentieth Century Mortality
CD-ROM. ONS, London.
World Health Organisation (1977). Manual of the International
Statistical Classification of Diseases, Injuries, and Causes of
Death, volume 1 (ninth revision). WHO, Geneva.
Office for National Statistics (1996). Mortality statistics: cause
1993 (revised) and 1994, series DH2 no 21. HMSO, London.
Christophersen O, Rooney C and Kelly S (1998). ‘Drug-related
mortality: methods and trends’. Population Trends 93, pp. 29–37.
TSO, London.
Blane D and Drever F (1998). ‘Inequality among men in
standardised years of potential life lost, 1970–93’. British Medical
Journal 317, p. 255. BMJ, London.
Tickle L (1991). ‘Mortality trends in the United Kingdom, 1982 to
1992’. Population Trends 86, pp. 21–8. HMSO, London
Kelly S and Bunting J (1998). ‘Trends in suicide in England and
Wales, 1982–96’. Population Trends 92, pp. 29–41. TSO, London.
Public Health Laboratory Service (1997). ‘The Epidemiology of
HIV infection and AIDS’ – a collection of articles that focus on the
UK and present data to the end of 1996. Communicable Disease
Report, Vol. 7, Review Number 9. PHLS, London.
It is important to continue to monitor mortality in young adults. First,
because although death rates are relatively low, they make a
considerable contribution to years of life lost 7 and to life expectancy
measure. 8 Second, it is among this age group that new public health
issues emerge, such as the changing epidemiology of suicide,9 the
increase in drug-related deaths6 and the continuation of the HIV
epidemic. 10
38
Office f or National Statistics
Health Statistics Quar terly 01
Spring 1999
Weekly deaths in England
and Wales
Sue Kelly, ONS, and
Helen Lawes, formerly
ONS
INTRODUCTION
The primary use of (provisional)
weekly death counts continues
to be rapid surveillance for
early warning signs of an upturn
in the number of deaths due, for
example, to a particularly cold
snap or an influenza epidemic.
Subsequently, weekly/daily
death counts are used for
epidemiological analysis. This
short article describes the
method used by ONS to
estimate weekly death
registrations between 1995 and
1998, and the alternative
methods investigated during
1996–98. The new method, in
use since October 1998, is both
more accurate and more
robust. The article also discusses
plans by ONS for future work in
this area.
The publication of weekly deaths data has a long history, dating back to
the 1850s when figures first appeared in the Registrar General’s Weekly
Return. Subsequently, weekly deaths tables have involved the
development of methods concerned with the weighting of available
figures to counter late registrations of death. In 1993 the handling of
death registrations in England and Wales changed from a manual to a
near fully-automated system,1 increasing the speed by which
information could be both entered onto the ONS deaths database and
extracted from it.
This article describes the method used by ONS to estimate weekly
death registrations between 1995 and 1998, the alternative methods
investigated during 1996–98, and the new method in use since October
1998. The article also discusses ONS’ plans for future work in this area.
The primary use of (provisional) weekly death counts continues to be
rapid surveillance for early warning signs of an upturn in the number of
deaths due, for example, to a particularly cold snap or an influenza
epidemic. Subsequently, weekly/daily death counts are used for
epidemiological analysis, including investigating the size of an
epidemic, who in the population was affected and what they died of. In
recent years ONS has carried out detailed analyses of the 1989/90
influenza epidemic2 and the 1996/97 winter when there was a very cold
snap at the end of December/beginning of January. 3
METHOD USED FROM MID-1995 TO SEPTEMBER 1998
Estimates of the number of death registrations in a particular week
(ending on a Friday) were produced on the Wednesday of the following
week. The method was based on the number of deaths submitted by
registration districts. The database submissions file was interrogated to
find out how many districts had sent information to ONS by
Wednesday. For the districts which had submitted, the program looked
at the submission file for the previous year, and counted their
submissions for the corresponding period, one year earlier. This
produced a figure of total registrations submitted by these districts for
the previous year. The actual death registrations for the previous week
received by Wednesday morning were then multiplied by a weighting
factor, based on these figures (see Box One).
This approach was regarded as more robust than simply grossing up on
the basis of the percentage of districts from which weekly returns had
been received. The calculation also took into account the size of
districts based on previous submission history. However, the program
was unable to cope with the splitting of registration districts, as
occurred in April 1996. Districts which had undergone changes could
Office for National Statistics
40
Health Statistics Quar terly 01
Box one
EXPLANATION OF ESTIMATION METHODS
Method used mid-1995 to September 1998
This method multiplies the number of deaths registered
in the week of interest and received by the following
Wednesday by a weighting factor.
For week of interest, the estimated number of death
registrations = (e/d1) x d2, where
d1 = deaths received last year from districts which
have already submitted in current year,
d2 = deaths received in current year from districts
who have submitted by Wednesday, and
e = total deaths received last year from all districts.
Method 1
This method adapts the previous method to allow for
records from districts in existence in the current year
which were not in existence in the previous year. Such
changes usually come about when a district splits into
two or more districts. The formula is the same but with
d1 now representing the sum of deaths received last
year from districts which have already submitted in the
current year plus deaths received in the current year
from districts which did not exist in the previous year.
Method 2
This method involves grossing up the number of
records for the week of interest received by the
following Wednesday by the percentage of districts yet
to submit.
For example, if 7,000 records received represents a 75
per cent sample of all districts, then the estimate is
7,000 x (100/75).
Method 3
This method is similar to the one used from mid-1995
to September 1998, but compares current figures with
registrations received six weeks ago, rather than one
year ago.
not be matched between years in order to obtain figures for the
calculation. Consequently, this method overestimated ‘true’ numbers of
weekly death registrations by an average of 588 (or 5.4 per cent) over
the period September 1996 to July 1998.
ALTERNATIVE METHODS FOR WEEKLY DEATH REGISTRATIONS
DATA
Between 1996 and 1998 alternative methods of estimating weekly death
registrations were investigated. One estimation method, to avoid the
problems mentioned above, was to manually add the actual numbers of
registrations for districts which had changed and had therefore been
‘missed’ by the program (Method 1 in Figure 1) (see Box One).
Spring 1999
Method 2 weighted the number of records received so far by the
percentage of districts yet to submit (see Box One). However, this
estimate was distorted by the fact that records received to date were
disproportionately composed of records from computerised register
offices. Conversely, the districts yet to submit were represented to a
disproportionate extent by non-computerised register offices. This
group tends to include relatively large numbers of small rural districts
with low numbers of weekly deaths, thereby distorting the picture still
further. This method produced a poor estimate and is not shown in
Figure 1.
A third method involved looking at the registrations received six weeks
ago, rather than registrations one year previously (Method 3 in Figure 1).
This ensured that changes to registration district boundaries were
obviated. However, it would not take account of seasonal variations in
mortality levels.
These three alternative methods of estimating weekly death
registrations were run in parallel with the ‘main’ method from
September 1996 to July 1998. Additionally, from November 1997 to
July 1998, we also recorded the actual number of death registrations for
each week received by ONS and entered on the database by the
following week. Figure 1 shows data for Methods 1 and 3, the method
in use from mid-1995 to September 1998, and the actual number of
registrations (that is the Wednesday count unadjusted), together with
‘final’ data (as at October 1998), for the period November 1997 to July
1998. On comparing the various methods we concluded that both
Methods 1 and 3 produced a more accurate estimate than the method in
use at the time, but that Method 3 was slightly more robust.
During the November 1997-July 1998 period, in addition to recording
data for the various methods on the Wednesday following the week of
interest, we also produced data on the Thursday and Friday. These extra
couple of days brought the actual figure closer to the ‘final’ figure, but
had little, if any, effect on the estimated figures.
It should be noted that Figure 1 does not include any estimated figures
for weeks 51, 52 and 1, which cover the Christmas and New Year
period. In practice these figures are produced later than usual due to
both the closure of register offices and the shutdown of the ONS
computer over the holiday period.
A decision was taken in September 1998 to change the method of
producing weekly death figures to estimation Method 3. This new
method was first implemented for deaths registered in the week ending
2 October 1998.
DEATHS
BY DATE OF OCCURRENCE
Published weekly mortality data has traditionally used death
registrations. Problems associated with using death registrations rather
than occurrences, due to the fact that the length of time between
occurrence and registration may vary considerably, 4 have long been
recognised. In particular, there is an uneven pattern of registrations
resulting from the influence of Bank Holidays when register offices are
closed.
When running all the alternative methods for weekly deaths data over
the period November 1997 to July 1998, we also produced a count of
the actual number of death occurrences to date, on the database, for the
previous week. As for the registrations, we took these counts on a
Thursday and Friday as well as on the Wednesday following the week
of interest. Figure 2 compares these counts and the ‘final’ number of
death occurrences (as at October 1998).
41
Office for National Statistics
Health Statistics Quar terly 01
Figure 1
Spring 1999
Weekly death registrations using different methods, 1997/98
Number of deaths
14,000
12,000
10,000
8,000
6,000
4,000
Method 1995-98
Method 3
Method 1
Actual
Final
2,000
0
45 46 47 48 49 50 51 52 1
Figure 2
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Week number
Weekly death occurrences: different counts compared to final figures, 1997/98
Number of deaths
14,000
12,000
FJ
J F
J
J
10,000
8,000
F
J
J
F
F J
FJ
J J
J
J J J J
F
FJ
F F
F
J FJ F F JJ J FJ FJ F
J J J F
J
J
J
J J J
J J
F
J
J
J
J
J
6,000
F F
J J F
J JJ
FJ
J
J
F J FJ J FJ FJ
J
J FJ
J J J J J J J
JJ
4,000
J
Wed
J
Thurs
F
Fri
Final
2,000
0
45 46 47 48 49 50 51 52 1
Office for National Statistics
2
3
42
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Week number
Health Statistics Quar terly 01
Not surprisingly counts extracted on a Friday were closest to the ‘final’
figures, but still substantially below by an average of 2,616 deaths (or
24.1 per cent) per week. The trend lines for the actual data were very
similar to the trend line for the ‘final’ figures, except at Christmas/New
Year and Easter. Indeed the peak in the ‘final’ figures in week 1 was
accompanied by a trough in the actual figures. This is due to the fact
that, although we have plotted occurrence data, there is still a
registration artifact caused by the closure of register offices over the
Christmas and New Year Bank Holidays. Also, the actual counts for
weeks 51 and 52 were higher than would be expected because these
counts were taken later than usual (beginning of January), due to both
the closure of register offices and the shutdown of the ONS computer
over the holiday period, allowing more death registrations to reach
ONS before the occurrence counts were extracted.
We concluded that actual occurrence counts, while being the ideal
measure, could not be used in their ‘raw’ form. ONS plans to look into
the possibility of modelling the actual occurrence data. In particular,
this will need to take into account the exact days of the week on which
Christmas and New Year Bank Holidays fall. This work is planned to be
carried out in the second half of 1999. Any new count using
occurrences would continue to be produced to the current timetable, in
order that rapid surveillance can continue.
weekly estimates of death registrations. Long series of these data
should not be used for epidemiological analysis. Instead users should
contact ONS for actual data. For recent data (deaths within the last nine
months) we can provide deaths by date of registration, whilst for earlier
data we can provide deaths by date of occurrence.
Key points
●
●
●
The estimation method used from mid–1995 to
September 1998 consistently overestimated weekly
death registrations.
A new estimation method was introduced in October
1998, which is both more accurate and more robust.
Weekly occurrence figures are the ideal measure, and
ONS plans to look into the possibility of modelling
actual weekly occurrence data.
REFERENCES
1
S UMMARY
The study of weekly deaths in relation to temperature or climate should
use deaths by date of occurrence, avoiding the problem of late
registrations and the unnatural peaks and troughs this creates in the
data. We will be carrying out further work on weekly occurrence data
later in the year. In the meantime, a new weekly estimate of death
registrations has been in use since October 1998. This estimate is both
more accurate and more robust than the estimate in use from mid-1995
to September 1998.
Spring 1999
2
3
4
Rooney C and Devis T (1996). ‘Mortality trends by cause of death
in England and Wales 1980–94; the impact of introducing
automated cause coding and related changes in 1993’. Population
Trends 86, pp. 29–35. HMSO, London.
Ashley J, Smith T, and Dunnell K (1991). ‘Deaths in Great Britain
associated with the influenza epidemic of 1989/90’. Population
Trends 65, pp. 16–20. HMSO, London.
Christophersen O (1997). ‘Mortality during the 1996/7 winter’.
Population Trends 90, pp. 11–17. TSO, London.
Devis T and Rooney C (1997). ‘The time taken to register a death’.
Population Trends 88, pp. 48–55. TSO, London.
Since April 1996 ONS has published weekly deaths data in a Monthly
News Release. Users of these data should remember that they are
43
Office for National Statistics
Health Statistics Quar terly 01
Spring 1999
Annual Update:
1997 Mortality Statistics:
Cause (England and Wales)
The ONS publication Mortality statistics: cause, England and Wales
1997, Series DH2 no. 24, was published in December 1998. It contains
statistics of deaths in England and Wales analysed by cause of death,
sex and age group. Deaths are classified according to the 9th revision of
the International Classification of Diseases (ICD9).
The number of deaths that occurred in England and Wales in 1997 was
555,281, just under 1 per cent fewer than in 1996 (Table 1). The crude
death rate (deaths per 1,000 population) was 10.6 in 1997.
The age-standardised death rate (see Notes to tables) continues to fall
for both sexes. Between 1971 and 1997, the rate fell by a third for
males and just over a quarter for females.
Age-standardised rates by underlying cause of death show that over
twice as many men as women died from lung cancer in 1997. However,
since 1984 the female rate has remained fairly constant between 280
and 300 deaths per million, while the male rate has fallen by over a
third from 998 to 649 deaths per million. Male and female death rates
from cerebrovascular disease have continued to fall and are both now
50 per cent lower than they were in 1971. Over the same period, the
mortality rate for both sexes from chronic liver disease and cirrhosis
has more than doubled.
Table 1
There are many diseases associated with older adults although there are
few diseases that can be said to be totally age-affected. Figure 1 shows
the percentage change in some of these diseases using two-year average
age-specific mortality rates over the period 1994/95 and 1996/97, for
men and women aged 65 and over. Mortality from pneumonia (ICD9
codes 480-486), accidental falls (E880-E888) and osteoporosis (733.0)
has increased for both sexes. The age-specific rate for accidental falls in
men in 1996/97 (29 per 100,000) is over a quarter lower than that for
women (40 per 100,000). Based on single-year data death rates for
senile dementia (290.0) increased between 1994 and 1995 but have
since fallen by between 6 and 16 per cent. Of the seven diseases
analysed rheumatoid arthritis (714.0) shows the only opposing trend for
men and women, with the rate for women at its lowest level for 20
years at 11 per 100,000.
Looking at these data broken down into 10-year age groups (65–74,
75–84 and 85 and over) gives a more in-depth analysis. Mortality from
osteoporosis in women has increased from 42 per 100,000 in 1994/95 to
44 per 100,000 in 1996/97 in the 85 and over age group, whilst the rates
have decreased for both the 65–74 and 75–84 year age groups. The
opposite is true for men dying from rheumatoid arthritis, where the rate
has decreased in the 85 and over age group, but risen for the younger
ages. Mortality from Parkinson’s disease (332) for men aged 65 and
Deaths and death rates in England and Wales
Males
Year
Total
deaths
Crude death
rate*
Age-standardised
rate†
1971
1981
1991
1992
567,262
577,890
570,044
558,313
11.6
11.6
11.2
10.9
1993
1994
1995
1996
1997
578,799
553,194
569,683
560,135
555,281
11.3
10.7
11.0
10.8
10.6
Females
Deaths
Crude death
rate*
Age-standardised
rate†
Deaths
Crude death
rate*
Age-standardised
rate†
10,326
9,374
7,987
7,724
288,359
289,022
277,582
271,732
12.2
12.0
11.1
10.8
13,464
12,200
10,234
9,870
278,903
288,868
292,462
286,581
11.1
11.3
11.2
10.9
8,186
7,433
6,399
6,197
7,878
7,486
7,574
7,376
7,239
279,561
267,555
274,449
268,682
264,865
11.1
10.6
10.8
10.5
10.3
10,010
9,502
9,582
9,271
9,019
299,238
285,639
295,234
291,453
290,416
11.4
10.9
11.2
11.0
10.9
6,347
6,039
6,128
5,995
5,926
* Deaths per 1,000 population.
† Deaths per million. (See Notes to tables.)
Note: Figures up to 1992 represent those deaths that were registered during the year, figures from 1993 onwards those that occurred during the year.
Office for National Statistics
44
Health Statistics Quar terly 01
over has not changed between the years covered and for women in the
same age group the rise was minimal. However, the rate for men aged
75–84 has decreased by nearly 9 per cent and that for women aged 65–
74 has decreased by over 15 per cent. The death rate for accidental falls
in men aged 65 and over has increased by nearly 4 per cent, although a
rise (of 11 per cent) is seen in the 65–74 year age group, with rates for
the older ages falling by 2 and 3 per cent respectively.
Table 2 shows that age-specific death rates have fallen sharply for all
age groups since 1971. The largest decrease has been for the youngest
age group, 0–14, where the rate has fallen by almost two-thirds for each
sex. Since 1971, rates for males have fallen proportionately more than
those for females.
However looking just at the trend during the 1990s, shows that the rates
among women aged between 15 and 44 have shown the smallest
decreases, of between 2 and 5 per cent. The rate for men aged 25–34
has increased during the 1990s by 2 per cent, having peaked in 1994.
The rates for most age groups for each sex have decreased from 1996.
The exceptions are for men aged 15–24 and for women aged 45–54 and
for women aged 85 and over which have shown small increases.
Table 2
Spring 1999
For the first time since 1986 the DH2 includes information on the
number of causes mentioned at death certification. The underlying
cause will usually be selected from one of the mentioned causes
although occasionally it may be inferred from two or more mentioned
causes.
The tables in the DH2 present the mean number of mentioned causes
per death by underlying cause, the age distribution of mentioned causes
and the ratio of mentioned causes to underlying cause broken down by
age and sex.
There is little variation overall in the ratio of mentions to underlying
causes when analysed by sex. The ratio is 2.31 for males and 2.21 for
females.
However, there are marked variations between different conditions. For
male deaths in 1997, septicaemia was given as a mentioned cause over
11 times as often as it was assigned as an underlying cause, but for
stomach cancer the ratio was just 1.1.
Further analysis of mentioned causes will appear in future ONS publications and additional information is available to researchers on request.
Age-specific death rates per million population, England and Wales, 1971–97
Age group
Year
0–14
15–24
25–34
35–44
45–54
55–64
65–74
75–84
85+
Males
1971
1981
1991
1992
1993
1994
1995
1996
1997
1,806
1,118
844
736
693
665
644
631
610
915
825
800
723
722
705
730
709
735
974
891
907
907
911
962
951
953
928
2,311
1,830
1,741
1,714
1,674
1,660
1,669
1,622
1,543
7,069
6,111
4,562
4,285
4,244
3,993
4,081
4,016
3,940
20,083
17,740
13,866
13,417
13,258
12,415
12,264
11,970
11,453
50,510
45,641
38,092
37,262
37,893
36,171
36,085
34,499
33,221
112,955
105,224
92,560
90,146
93,338
89,549
89,445
85,061
82,489
231,842
226,232
202,602
193,881
202,285
188,636
195,997
192,139
190,327
Females
1971
1981
1991
1992
1993
1994
1995
1996
1997
1,359
835
657
571
548
515
496
488
480
405
334
314
308
294
277
279
300
297
605
518
439
430
447
435
456
454
425
1,592
1,259
1,053
1,077
1,056
1,064
1,047
1,032
1,031
4,318
3,798
2,871
2,727
2,726
2,683
2,722
2,619
2,630
10,047
9,492
8,153
7,915
7,858
7,325
7,285
7,087
6,942
26,127
24,145
21,846
21,469
21,973
21,266
21,393
20,653
20,189
73,645
66,224
58,169
56,922
59,396
56,871
57,075
55,782
54,572
185,684
178,183
155,527
148,803
156,467
146,650
153,080
150,777
151,749
Persons
1971
1981
1991
1992
1993
1994
1995
1996
1997
1,589
980
753
655
622
592
572
561
547
664
584
563
521
514
497
510
510
522
792
706
676
673
683
704
709
709
683
1,955
1,547
1,398
1,396
1,366
1,363
1,360
1,330
1,289
5,671
4,954
3,717
3,507
3,485
3,338
3,401
3,317
3,285
14,817
13,448
10,955
10,617
10,514
9,832
9,742
9,499
9,171
36,380
33,545
29,155
28,601
29,184
28,033
28,095
27,004
26,200
86,638
79,690
70,836
69,221
72,004
69,059
69,291
66,935
65,307
197,537
189,380
166,816
159,753
167,786
157,126
163,947
161,394
161,807
Figures up to 1992 represent those deaths that were registered during the year, figures from 1993 onwards those that occurred during the year.
45
Office for National Statistics
Health Statistics Quar terly 01
Figure 1
Spring 1999
Percentage change in death rates by cause and sex, ages 65 and over, 1994/95–1996/97, England and Wales
Percentage change
15
Men
Women
10
5
0
-5
-10
-15
Pneumonia
Accidental falls
Diabetes
Senile dementia
Cause of death
Office for National Statistics
46
Parkinson's disease
Rheumatoid arthritis
Osteoporosis
Tables
Page
Table
Population
1.1
1.2
International
National
1.3
Subnational
1.4
Subnational
1.5
Age and sex
1.6
Age, sex and legal marital status
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
Selected countries
Constituent countries of
the United Kingdom
Health Regional Office areas
of England
Government Office Regions
of England
Constituent countries of
the United Kingdom
England and Wales
48
50
51
52
53
56
Vital statistics
2.1
Summary
2.2
Key demographic and health indicators
○
○
○
○
○
○
○
○
○
○
○
○
○
○
Constituent countries of
the United Kingdom
Constituent countries of
the United Kingdom
58
60
Live births
Age of mother
Outside marriage: age of mother and type of registration
3.1
3.2
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
England and Wales
England and Wales
61
62
England and Wales (residents)
England and Wales
63
64
Constituent countries of
the United Kingdom
65
○
Conceptions and abortions
Age of women at conception
Abortions: age and gestation.
4.1
4.2
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
Expectation of life
(In years) at birth and selected age
5.1
○
○
○
○
○
○
○
Deaths
6.1
6.2
Age and sex
Subnational
6.3
Selected causes and sex
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
England and Wales
Health Regional Office areas
of England
England and Wales
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
○
Notes to tables
66
67
68
70
Symbols
..
:
blank
not available
not applicable
nil or less than half the final digit shown
not yet available
47
Office for National Statistics
Health Statistics Quar terly 01
Table 1.1
Year
Spring 1999
Population and vital rates: international
United
Kingdom (1)
Austria
(2)
Belgium
(2)
Denmark
(2)
Finland
(2)
7,501
7,566
7,569
7,588
7,818
9,673
9,811
9,859
9,862
10,005
4,963
5,073
5,122
5,121
5,154
7,989
8,028
8,047
8,059
10,085
10,116
10,137
10,157
Population changes (per 1,000 per annum)
1971–76
1.0
1.7
2.9
1976–81
0.5
0.1
1.0
1981–86
1.8
0.5
0.1
1986–91
1.7
Population (thousands)
1971
55,928
1976
56,216
1981
56,352
1986
56,852
1991
57,808
1992**
58,006
1993
1994
1995
1996
1997
58,191
58,395
58,606
58,801
59,009
1991–92
1992–93
1993–94
1994–95
1995–96
1996–97
3.4
3.2
3.5
3.6
3.3
3.5
12.3
9.3
4.9
2.4
1.6
Live birth rate (per 1,000 per annum)
1971–75
14.1
13.3
1976–80
12.5
11.5
1981–85
12.9
12.0
1986–90**
1991
13.7
12.1
1992
13.5
12.1
1993
1994
1995
1996
1997
13.1
12.9
12.5
12.5
12.3
11.9
11.5
11.0
11.0
10.4
Death rate (per 1,000 per annum)
1971–75
11.8
12.6
1976–80
11.9
12.3
1981–85
11.7
12.0
1986–90**
1991
11.3
10.7
1992
11.0
10.5
1993
1994
1995
1996
1997
*
†
**
‡
≠
11.3
10.7
10.9
10.9
10.8
10.3
10.0
10.1
10.0
9.8
Excluding former GDR throughout.
Including former GDR throughout.
Not currently available.
Provisional.
Estimates prepared by the Population
Division of the United Nations.
+ Rates are for 1990–95.
Selected countries
France
Germany Germany
(2) (Fed. Rep (2))* (2)†
Greece
(2)
Irish
Republic (2)
Italy
(2)
Luxembourg (2)
Netherlands (2)
Portugal
(2)
4,612
4,726
4,800
4,918
5,014
51,251
52,909
54,182
55,547
57,055
61,302
61,531
61,682
61,066
64,074
78,352
78,321
78,419
77,694
80,014
8,831
9,167
9,729
9,967
10,247
2,978
3,228
3,443
3,541
3,526
54,074
55,718
56,510
56,596
56,751
342
361
365
368
387
13,195
13,774
14,247
14,572
15,070
8,644
9,355
9,851
10,011
9,871
5,189
5,205
5,228
5,262
5,066
5,089
5,108
5,125
57,654
57,899
58,137 ‡
58,374 ‡
65,534
65,858
66,715
81,156
81,438
81,678
10,380
10,426
10,454
10,475
3,574
3,587 ‡
3,605 ‡
3,626 ‡
57,049
57,204
57,301
57,397
398
404
410
416
15,290
15,383
15,459
15,531
9,881
9,902
9,917
9,927
4.4
1.9
0.0
4.9
3.1
4.9
6.5
4.8
5.0
0.7
0.5
–2.0
-0.1
0.3
-1.8
7.6
12.3
4.9
16.8
13.3
5.7
6.1
2.8
0.3
10.7
2.5
1.8
8.8
6.9
4.6
16.5
10.6
3.2
4.1
3.9
3.1
2.1
1.9
3.2
3.7
3.0
4.4
6.4
5.6
4.8
4.4
3.7
3.3
5.6
4.9
4.2
4.1 ‡
4.1 ‡
12.3
10.3
4.9
13.0
7.6
6.6
3.5
2.9
7.3
5.6
4.5
2.7
2.0
8.8
4.8
3.9 ‡
5.0 ‡
5.8 ‡
1.9
3.4
2.7
1.7
1.7
13.9
14.3
14.3
14.6
14.4
7.6
7.0
6.1
4.9
4.6
–0.4
1.4
2.2
1.4
1.1
13.4
12.5
12.0
14.6
12.0
10.2
13.1
13.6
13.4
16.0
14.1
14.2
10.8
9.7
9.8
10.5
10.5
10.7
15.8
15.6
13.3
22.2
21.3
19.2
16.0
12.6
10.6
11.6
11.2
11.6
14.9
12.6
12.2
20.3
17.9
14.5
12.6
12.4
12.5
13.1
13.0
13.3
13.3
13.0
11.3
11.1
10.4
10.1
10.1
10.1
15.0
14.4
9.9
9.7
12.9
13.1
13.2
13.0
11.8
11.6
13.0
13.4
13.4
12.9 ‡
12.8 ‡
12.8
12.8
12.3
11.8
11.5 ‡
12.3
12.3
12.5 ‡
12.6 ‡
12.4 ‡
11.0
10.5
10.2
10.6 ‡
9.8
10.0
9.7
9.6 ‡
9.7 ‡
13.8
13.4 ‡
13.5 ‡
13.9 ‡
14.3 ‡
9.6
9.3
9.2 ‡
9.2 ‡
9.2 ‡
13.4
13.5
13.2
13.7
13.1
12.8
12.7
12.3
12.2
12.2 ‡
11.5
11.0
10.8
11.1
11.4
12.0
11.5
11.4
11.4
11.4
‡
‡
‡
‡
‡
9.9
9.5
9.4
9.7 ‡
12.1
11.6
11.4
10.1
10.5
11.1
9.5
9.3
9.3
10.7
10.2
10.1
11.9
11.7
11.6
12.3
12.2
12.0
8.6
8.8
9.0
11.0
10.2
9.4
9.8
9.7
9.5
12.2
11.5
11.2
8.3
8.1
8.3
11.0
10.1
9.6
10.5
10.3
11.6
11.8
9.8
9.9
9.2
9.1
11.1
10.7
11.4
11.0
9.3
9.5
8.9
8.7
9.7
9.6
9.7
10.2
8.6
8.6
10.5
10.2
9.2
9.0
9.1 ‡
9.2 ‡
9.1 ‡
10.9
10.7
10.6
10.7
11.1
10.9
10.8
10.8
9.4
9.4
9.6
9.6
9.6
8.7
8.6 ‡
9.0
8.8 ‡
8.6 ‡
9.7
9.7
9.5
9.5
9.6
9.0
8.7
8.8
8.9 ‡
8.7 ‡
10.7
10.0
10.4
10.8
10.5
10.7
10.4
10.5
10.4
10.2
‡
‡
‡
‡
‡
12.1
11.7
12.1
11.6 ‡
11.3 ‡
(1)
(2)
(3)
(4)
(5)
(6)
(7)
Population estimated at 30 June each year.
Average of estimated populations at start and end of year as given in Council of Europe report Recent demographic developments in Europe 1997.
EU as constituted 1 January 1986 and including countries subsequently admitted.
Population estimated at 1 June each year.
Population estimated at 31 December each year.
Population estimated at 1 July except for 1991 (1 March).
Population estimated at 1 October. (Rates for Japan are based on population of Japanese nationality only.)
Note: Figures may not add exactly due to rounding.
Office for National Statistics
48
10.1
9.4
9.6
9.6
9.6
‡
‡
‡
‡
9.8
9.4
9.3
9.4
9.4
Health Statistics Quar terly 01
Population and vital rates: international
Table 1.1
continued
Spain
(2)
Spring 1999
Sweden
(2)
European
Union (3)
Selected countries
Russian
Federation (2)
Australia
(1)
Canada
(4)
New
Zealand (5)
139,422
144,475
148,624
13,067
14,033
14,923
16,018
17,284
22,026
23,517
24,900
26,204
28,120
2,899
3,163
3,195
3,317
3,450
852,290
943,033
1,011,219
1,086,733
1,170,052
148,520
148,336
148,141
147,739
17,667
17,855
18,072
18,311
28,947
29,256
29,615
29,964 ‡
3,556
3,604
3,658
3,716
1,190,360 ≠
1,208,841 ≠
1,221,462 ≠
7.2
14.8
12.7
14.7
13.5
11.8
10.5
18.2
2.0
7.6
19.9
15.2
15.5
Population changes (per 1,000 per annum)
23.9
15.1
10.0
1971–76
18.8
8.5
10.9
1976–81
27.3
6.4
9.3
1981–86
1986–91
0.4
-1.1
-1.2
-1.3
-2.7
12.2
9.9
10.6
12.2
13.2
15.0
14.2
10.7
12.3
11.8‡
19.0
11.5
13.5
15.0
15.8
11.6
5.7
15.5
10.4
19.0
18.5
39.2
18.7
18.8
15.7
15.6
15.9
15.5
15.1
20.4
16.8
15.8
27.2
18.6
19.2
35.6
33.4
..
12.1
10.7
14.9
15.1
14.3
14.0
17.4
17.2
9.3
9.5
9.2
8.8
8.6
14.7
14.5
14.2
13.9
13.4
13.2
12.8
17.1
16.4
16.3
18.3+
28.7
28.7
28.3
8.2
7.6
7.3
7.4
7.2
7.0
8.4
8.2
8.1
7.3
6.6
6.7
15.5
13.8
..
34,190
35,937
37,742
38,537
38,920
8,098
8,222
8,321
8,370
8,617
342,631
350,384
356,511
359,543
366,256
39,086
39,150
39,210
39,270
8,719
8,781
8,827
8,841
369,706 ‡
371,005 ‡
372,122 ‡
10.2
10.0
4.2
3.1
2.4
1.2
4.5
3.5
1.7
2.3
2.0
1.6
1.5
1.5
5.9
5.8
7.1
5.3
1.6
4.9
4.5
3.5
3.0
19.2
17.1
12.8
13.5
11.6
11.3
14.7
13.1
12.2
10.2
10.2
14.3
14.2
11.7
11.5
9.9
9.5
9.3 ‡
9.1 ‡
9.2 ‡
13.5
12.8
11.7
10.8
10.2
11.2 ‡
10.9 ‡
10.7 ‡
8.5
8.0
7.7
10.5
10.9
11.0
10.8
10.6
10.4
8.6
8.5
11.0
10.9
10.2
10.0
11.4
12.2
6.9
7.1
7.0
6.9
7.8
7.9
8.7
8.6
8.8 ‡
8.9 ‡
8.9 ‡
11.1
10.5
10.6
10.6
10.5
10.2‡
9.9‡
10.0‡
14.3
15.5
14.9
14.1
13.7
6.8
7.1
6.9
7.0
7.1
7.1
7.1
7.2
7.7
7.5
7.6
7.6
‡
‡
‡
‡
China
(5)
≠
≠
≠
≠
India
(6)
Japan
(7)
551,311
617,248
676,218
767,199
851,661
105,145
113,094
117,902
121,672
123,102
833,910
918,570 ≠
935,744 ≠
123,788
124,069
124,299
124,709
29.5
29.0
9.8
10.1
7.2+
9.3
9.3
9.0
3.0
2.5
2.3
1.9
3.3
USA
(1)
Year
Population (thousands)
207,661
1971
218,035
1976
230,138
1981
240,680
1986
252,177
1991
1992
257,783
260,341
262,755
265,284
11.5
10.6
9.9
9.3
9.6
1993
1994
1995
1996
1997
1991–92
1992–93
1993–94
1994–95
1995–96
1996–97
Live birth rate (per 1,000 per annum)
18.6
15.3
1971–75
14.9
15.2
1976–80
12.6
15.7
1981–85
1986–90
9.9
16.3
1991
9.7
16.0
1992
9.5
9.9
9.5
9.6 ‡
15.5
15.2
14.8
14.8 ‡
Death rate (per 1,000 per annum)
6.4
9.1
1971–75
6.1
8.7
1976–80
6.1
8.6
1981–85
1986–90
6.7
8.6
1991
6.9
8.5
1992
7.0
7.0
7.4
7.1
8.8
8.7
8.8
8.8 ‡
See notes opposite.
49
1993
1994
1995
1996
1997
Office f or National Statistics
1993
1994
1995
1996
1997
Health Statistics Quar terly 01
Table 1.2
Spring 1999
Population: national
Numbers (thousands) and percentage age distribution
Mid-year
Constituent countries of the United Kingdom
United
Kingdom
Great
Britain
England
and Wales
England
Wales
Scotland
Northern
Ireland
Estimates
1971
1976
1981
1986
1991
1992
55,928
56,216
56,352
56,852
57,808
58,006
54,388
54,693
54,815
55,285
56,207
56,388
49,152
49,459
49,634
50,162
51,100
51,277
46,412
46,660
46,821
47,342
48,208
48,378
2,740
2,799
2,813
2,820
2,891
2,899
5,236
5,233
5,180
5,123
5,107
5,111
1,540
1,524
1,538
1,567
1,601
1,618
1993
1994
1995
1996
1997
58,191
58,395
58,606
58,801
59,009
56,559
56,753
56,957
57,138
57,334
51,439
51,621
51,820
52,010
52,211
48,533
48,707
48,903
49,089
49,284
2,906
2,913
2,917
2,921
2,927
5,120
5,132
5,137
5,128
5,123
1,632
1,642
1,649
1,663
1,675
6.3
14.2
40.9
20.4
10.9
7.2
6.3
14.1
40.9
20.5
10.9
7.3
6.3
14.2
40.8
20.5
10.9
7.3
6.3
14.1
41.0
20.5
10.8
7.3
6.0
14.5
38.4
21.2
12.0
7.9
6.0
13.9
41.8
20.4
11.3
6.6
7.4
17.2
41.9
18.5
9.6
5.5
59,618
60,287
60,929
61,605
62,244
57,924
58,576
59,209
59,880
60,519
52,818
53,492
54,151
54,849
55,526
49,871
50,526
51,161
51,832
52,484
2,947
2,966
2,989
3,017
3,043
5,106
5,084
5,059
5,031
4,993
1,694
1,711
1,720
1,725
1,724
5.6
12.2
35.7
27.3
10.6
8.6
5.6
12.1
35.7
27.3
10.6
8.6
5.6
12.2
35.8
27.2
10.6
8.7
5.6
12.2
35.8
27.3
10.5
8.6
5.6
12.4
35.2
26.2
11.3
9.4
5.3
11.8
34.6
28.7
11.1
8.5
5.8
13.1
36.7
27.0
9.6
7.7
of which (percentages)
0–4
5–15
16–44
45–64M/59F
65M/60F–74
75 and over
Projections≠
2001
2006
2011
2016
2021
of which (percentages)
0–4
5–15
16–44
45–64†
65–74†
75 and over
≠ These projections are based on the mid-1996 population estimates.
† Between 2010 and 2020, state retirement age will change from 65 years for men and 60 years for women, to 65 years for both sexes.
Note: Figures may not add exactly due to rounding.
Office for National Statistics
50
Health Statistics Quar terly 01
Table 1.3
Population: subnational
Numbers (thousands) and percentage age distribution
Mid-year
Spring 1999
Health Regional Office areas of England*
Northern and
Yorkshire +
Trent +
Anglia and
Oxford
North
Thames
South
Thames
South and
West
West
Midlands
North
West
Estimates
1971
1976
1981
1986
1991
1992
6,482
6,512
6,238
6,207
6,285
6.309
4,483
4,557
4,921
4,945
5,035
5,060
4,272
4,531
4,745
4,979
5,174
5,206
6,914
6,695
6,598
6,652
6,744
6,766
6,642
6,567
6,489
6,567
6,679
6,696
5,569
5,789
5,988
6,224
6,426
6,459
5,146
5,178
5,187
5,197
5,265
5,278
6,903
6,832
6,657
6,570
6,600
6,603
1993
1994
1995
1996
1997
6,323
6,332
6,337
6,338
6,336
5,081
5,096
5,109
5,121
5,128
5,226
5,261
5,315
5,361
5,410
6,795
6,830
6,872
6,934
6,988
6,715
6,749
6,782
6,819
6,865
6,486
6,529
6,569
6,594
6,639
5,290
5,295
5,306
5,317
5,321
6,617
6,616
6,614
6,605
6,598
6.1
14.4
40.4
20.6
11.3
7.1
6.1
14.1
40.3
20.9
11.3
7.3
6.4
14.3
41.5
20.8
10.2
6.8
6.9
13.8
44.1
19.2
9.6
6.5
6.4
13.5
41.6
20.1
10.6
7.9
5.8
13.7
39.1
21.1
11.8
8.6
6.4
14.6
40.0
20.9
11.1
7.0
6.3
14.9
40.2
20.6
11.0
7.1
6,365
6,382
6,405
6,435
6,464
5,184
5,232
5,277
5,324
5,371
5,568
5,747
5,906
6,057
6,198
7,088
7,220
7,352
7,487
7,614
6,955
7,077
7,198
7,326
7,455
6,786
6,958
7,122
7,291
7,456
5,343
5,358
5,372
5,391
5,411
6,582
6,553
6,530
6,521
6,515
4.5
12.8
37.6
26.6
12.8
8.6
4.4
12.7
37.4
23.7
12.9
8.9
4.6
13.1
37.8
23.6
12.5
8.5
5.1
13.4
40.8
22.8
10.8
7.1
4.7
12.9
38.3
23.5
12.0
8.6
4.2
12.2
35.8
23.7
13.9
10.3
4.6
13.3
37.1
23.5
12.8
8.7
4.6
13.2
37.8
23.5
12.5
8.4
of which (percentages)
0–4
5–15
16–44
45–64M/59F
65M/60F–74
75 and over
Projections≠
2001
2006
2011
2016
2021
of which (percentages)
0–4
5–15
16–44
45–64†
65–74†
75 and over
* Areas as constituted in 1996. Population figures for years before 1981 may relate to different areas where boundaries have changed.
≠ These projections are based on the mid-1996 population estimates.
+ From 1 April 1996 boundary changes due to local government reorganisation have led to changes in the constitution of the Northern and Yorkshire and Trent Regional Office areas. South
Humber Health Authority with 311.3 thousand people – mid-1996 is now included in the Trent Regional Office area rather than in the Northern and Yorkshire area.
† Between 2010 and 2020, state retirement age will change from 65 years for men and 60 years for women, to 65 years for both sexes.
Note: Figures may not add exactly because of rounding.
51
Office f or National Statistics
Health Statistics Quar terly 01
Table 1.4
Spring 1999
Population: subnational
Numbers (thousands) and percentage age distribution
Mid-year
Government Office Regions of England*
North
East
North
West and
Merseyside
North
West
Merseyside
Yorkshire
and the
Humber
East
Midlands
West
Midlands
Eastern
London
South
East
South
West
Estimates
1971
1976
1981
1986
1991
1992
2,679
2,671
2,636
2,601
2,603
2,609
7,108
7,043
6,940
6,852
6,885
6,890
5,446
5,457
5,418
5,381
5,436
5,444
1,662
1,586
1,522
1,471
1,450
1.446
4,902
4,924
4,918
4,906
4,983
5,002
3,652
3,774
3,853
3,919
4,035
4,062
5,146
5,178
5,187
5,197
5,265
5,278
4,454
4,672
4,854
5,012
5,150
5,175
7,529
7,089
6,806
6,803
6,890
6,905
6,830
7,029
7,245
7,492
7,679
7,712
4,112
4,280
4,381
4,560
4,718
4,746
1993
1994
1995
1996
1997
2,612
2,610
2,605
2,600
2,594
6,903
6,902
6,900
6,891
6,885
5,462
5,468
5,473
5,471
5,471
1,441
1,434
1,427
1,420
1,413
5,014
5,025
5,029
5,036
5,037
4,083
4,102
4,124
4,141
4,156
5,290
5,295
5,306
5,317
5,321
5,193
5,223
5,257
5,293
5,334
6,933
6,968
7,007
7,074
7,122
7,737
7,784
7,847
7,895
7,959
4,768
4,798
4,827
4,842
4,876
6.0
14.5
40.3
20.6
11.7
6.9
6.2
14.8
40.1
20.6
11.0
7.1
6.3
14.8
40.1
20.8
10.9
7.1
6.1
15.0
40.3
19.9
11.5
7.2
6.3
14.4
40.6
20.4
11.0
7.3
6.1
14.2
40.3
21.1
11.1
7.2
6.4
14.6
40.0
20.9
11.1
7.0
6.3
13.9
40.3
21.1
11.0
7.4
7.1
13.6
46.1
18.2
8.9
6.2
6.2
13.9
40.4
21.0
10.8
7.8
5.8
13.5
38.5
21.1
12.1
9.0
2,579
2,555
2,536
2,521
2,509
6,871
6,843
6,820
6,813
6,808
5,485
5,490
5,947
5,514
5,530
1,386
1,353
1,323
1,299
1,278
5,071
5,098
5,130
5,165
5,200
4,234
4,312
4,384
4,455
4,523
5,343
5,358
5,372
5,391
5,411
5,448
5,582
5,702
5,823
5,941
7,215
7,337
7,470
7,609
7,736
8,134
8,344
8,534
8,722
8,905
4,977
5,098
5,213
5,333
5,452
4.4
12.7
37.3
23.9
13.1
8.6
4.6
13.2
37.6
23.6
12.6
8.5
4.6
13.2
37.6
23.6
12.6
8.4
4.5
13.1
37.8
23.2
12.7
8.7
4.5
12.9
38.0
23.5
12.5
8.5
4.4
12.8
37.3
23.7
12.9
8.8
4.6
13.3
37.1
23.5
12.8
8.7
4.5
13.0
36.7
23.5
13.1
9.3
5.3
13.3
43.1
22.6
9.7
6.1
4.5
12.9
37.0
23.7
12.7
9.2
4.1
12.1
35.1
23.9
14.3
10.7
of which (percentages)
0–4
5–15
16–44
45–64M/59F†
65M/60F–74†
75 and over
Projections≠
2001
2006
2011
2016
2021
of which (percentages)
0–4
5–15
16–44
45–64†
65–74†
75 and over
* See map on page 71.
≠ These projections are based on the 1996 population estimates.
† Between 2010 and 2020, state retirement age will change from 65 years for men and 60 years for women, to 65 years for both sexes.
Note: Figures may not add exactly because of rounding.
Office for National Statistics
52
Health Statistics Quar terly 01
Table 1.5
Population: age and sex
Numbers (thousands)
Spring 1999
Constituent countries of the United Kingdom
Age group
Mid-year
All ages
United Kingdom
Persons
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
55,928
56,216
56,352
56,852
57,808
58,006
58,191
58,395
58,606
58,801
59,009
Males
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
Females
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
Under 1
1–4
5–14
15–24
25–34
35–44
45–59
60–64
65–74
75–84
85–89
90 and
over
Under
16
16–
64/59
65/60
and over
899
677
730
749
794
787
759
759
734
719
736
3,654
3,043
2,725
2,893
3,092
3,124
3,129
3,117
3,102
3,044
2,977
8,916
9.176
8,147
7,163
7,175
7,289
7,417
7,483
7,526
7,595
7,665
8,144
8.126
9,019
9.283
8,247
7,969
7,723
7,554
7,450
7,325
7,219
6,971
7,868
8,010
8,048
9,057
9,199
9,295
9,375
9,409
9,420
9,362
6,512
6,361
6,774
7,717
7,955
7,818
7,787
7,837
7,931
8,093
8,296
10,202
9,836
9,540
9,210
9,500
9,814
10,070
10,277
10,445
10,582
10,702
3,222
3,131
2,935
3,067
2,888
2,868
2,839
2,808
2,784
2,772
2,783
4,764
5,112
5,195
5,017
5,067
5,104
5,169
5,223
5,127
5,058
5,006
2,159
2,348
2,675
2,986
3,136
3,100
3,020
2,952
3,054
3,125
3,175
358
390
428
516
639
662
688
703
719
728
732
127
147
174
203
257
274
294
308
325
339
356
14,257
13,797
12,541
11,679
11,741
11,850
11,965
12,075
12,106
12,098
12,104
32,548
32,757
33,780
34,846
35,469
35,533
35,589
35,689
35,848
36,035
36,214
9,123
9,663
10,031
10,328
10,597
10,623
10,637
10,630
10,652
10,668
10,691
27,167
27,360
27,409
27,698
28,246
28,362
28,474
28,592
28,727
28,856
28,990
461
348
374
384
407
403
389
389
376
369
377
1,874
1,564
1,399
1,484
1,588
1,602
1,603
1,596
1,589
1,560
1,526
4,576
4,711
4,184
3,682
3,688
3,744
3,808
3,840
3,861
3,897
3,932
4,137
4,145
4,596
4,747
4,227
4,087
3,965
3,879
3,825
3,760
3,705
3,530
3,981
4,035
4,063
4,591
4,670
4,723
4,767
4,793
4,805
4,780
3,271
3,214
3,409
3,871
3,986
3,917
3,904
3,929
3,984
4,072
4,182
4,970
4,820
4,711
4,572
4,732
4,889
5,017
5,118
5,201
5,270
5,329
1,507
1,466
1,376
1,462
1,390
1,384
1,374
1,363
1,358
1,355
1,361
1,999
2,204
2,264
2,205
2,272
2,297
2,333
2,363
2,330
2,310
2,299
716
775
921
1,063
1,151
1,143
1,117
1,096
1,147
1,185
1,215
97
101
105
128
167
176
186
193
201
206
211
29
31
35
38
47
51
55
58
62
67
71
7,318
7,083
6,438
5,999
6,033
6,084
6,140
6,194
6,208
6,205
6,298
17,008
17.167
17,646
18,266
18,576
18,611
18,642
18,687
18,779
18,882
18,984
2,841
3,111
3,325
3,433
3,637
3,666
3,692
3,710
3,740
3,768
3,796
28,761
28,856
28,943
29,153
29,562
29,645
29,718
29,803
29,878
29,946
30,019
437
330
356
365
387
384
370
370
358
350
359
1,779
1,479
1,326
1,409
1,505
1,522
1,526
1,521
1,513
1,484
1,450
4,340
4,465
3,963
3,480
3,487
3,545
3,609
3,643
3,665
3,698
3,733
4,008
3,7980
4,423
4,536
4,021
3,882
3,758
3,674
3,625
3,565
3,514
3,441
3,887
3,975
3,986
4,466
4,530
4,572
4,608
4,616
4,615
4,581
3,241
3,147
3,365
3,846
3,968
3,900
3,883
3,908
3,947
4,020
4,114
5,231
5,015
4,829
4,638
4,769
4,925
5,054
5,159
5,244
5,312
5,374
1,715
1,665
1,559
1,605
1,498
1,484
1,466
1,444
1,427
1,418
1,422
2,765
2,908
2,931
2,813
2,795
2,807
2,836
2,861
2,797
2,748
2,707
1,443
1,573
1,755
1,923
1,986
1,957
1,903
1,856
1,907
1,940
1,960
261
289
322
388
472
486
502
510
518
522
521
97
116
139
164
210
223
239
249
262
273
285
6,938
6,714
6,103
5,679
5,708
5,766
5,826
5,881
5,898
5,893
5,895
15,540
15,590
16,134
16,580
16,893
16,922
16,946
17,002
17,069
17,153
17,229
6,282
6,552
6,706
6,894
6,961
6,957
6,946
6,920
6,911
6,900
6,894
England and Wales
Persons
1971
49,152
1976
49,459
1981
49,634
1986
50,162
1991
51,100
1992
51,277
1993
51,439
1994
51,621
1995
51,820
1996
52,010
1997
52,211
782
585
634
655
702
694
670
671
649
636
651
3,170
2,642
2,372
2,528
2,728
2,760
2,764
2,752
2,739
2,688
2,632
7,705
7,967
7,085
6,243
6,281
6,385
6,504
6,568
6,613
6,683
6,751
7,117
7,077
7,873
8,134
7,237
6,985
6,768
6,612
6,521
6,411
6,332
6,164
6,979
7,086
7,088
8,008
8,137
8,219
8,293
8,329
8,342
8,290
5,736
5,608
5,996
6,863
7,056
6,924
6,887
6,925
7,003
7,146
7,325
9,034
8,707
8,433
8,136
8,407
8,695
8,929
9,118
9,272
9,397
9,503
2,853
2,777
2,607
2,725
2,553
2,534
2,507
2,478
2,458
2,447
2,456
4,228
4,540
4,619
4,470
4,506
4,538
4,596
4,644
4,554
4,490
4,440
1,926
2,093
2,388
2,673
2,810
2,777
2,704
2,642
2,734
2,800
2,844
323
351
383
465
576
598
623
636
651
658
661
115
135
157
184
233
249
268
281
297
311
327
12,334
11,973
10,910
10,190
10,303
10,407
10,515
10,618
10,653
10,655
10,672
28,710
28,894
29,796
30,759
31,351
31,402
31,445
31,530
31,676
31,851
32,018
8,108
8,593
8,928
9,213
9,446
9,467
9,480
9,473
9,491
9,505
9,522
Males
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
23,897
24,089
24,160
24,456
24,995
25,099
25,198
25,304
25,433
25,557
25,684
402
300
324
336
360
356
343
344
333
327
334
1,626
1,358
1,218
1,297
1,401
1,416
1,416
1,410
1,403
1,378
1,350
3,957
4,091
3,639
3,211
3,231
3,282
3,341
3,371
3,394
3,430
3,463
3,615
3,610
4,011
4,156
3,710
3,583
3,476
3,396
3,348
3,291
3,249
3,129
3,532
3,569
3,579
4,065
4,137
4,184
4,225
4,252
4,265
4,243
2,891
2,843
3,024
3,445
3,539
3,472
3,456
3,475
3,523
3,602
3,700
4,414
4,280
4,178
4,053
4,199
4,343
4,458
4,551
4,626
4,689
4,740
1,337
1,304
1,227
1,302
1,234
1,228
1,218
1,209
1,204
1,201
1,206
1,778
1,963
2,020
1,972
2,027
2,049
2,082
2,109
2,078
2,059
2,048
637
690
825
954
1,035
1,028
1,004
985
1,032
1,066
1,094
86
91
94
115
151
160
170
175
183
188
192
26
29
32
35
43
46
51
53
57
61
66
6,334
6,148
5,601
5,236
5,296
5,346
5,397
5,448
5,465
5,466
5,475
15,036
15,169
15,589
16,143
16,442
16,470
16,495
16,533
16,619
16,716
16,810
2,527
2,773
2,970
3,076
3,257
3,283
3,306
3,323
3,349
3,375
3,399
Females
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
25,255
25,370
25,474
25,706
26,104
26,178
26,241
26,317
26,387
26,453
26,527
380
285
310
319
342
339
326
327
316
310
317
1,544
1,284
1,154
1,231
1,328
1,345
1,348
1,342
1,335
1,310
1,282
3,749
3,876
3,446
3,032
3,050
3,104
3,163
3,197
3,219
3,253
3,287
3,502
3,467
3,863
3,978
3,527
3,403
3,293
3,216
3,172
3,120
3,083
3,036
3,447
3,517
3,509
3,943
4,001
4,035
4,069
4,076
4,077
4,046
2,845
2,765
2,972
3,418
3,517
3,452
3,431
3,449
3,480
3,544
3,625
4,620
4,428
4,255
4,083
4,208
4,352
4,471
4,567
4,646
4,709
4,763
1,516
1,473
1,380
1,422
1,319
1,306
1,289
1,270
1,254
1,246
1,250
2,450
2,577
2,599
2,498
2,479
2,488
2,514
2,536
2,477
2,430
2,392
1,289
1,403
1,564
1,718
1,775
1,749
1,700
1,656
1,702
1,733
1,750
236
261
289
349
425
439
453
461
468
471
470
89
106
126
149
191
203
218
228
240
250
262
6,000
5,826
5,309
4,953
5,007
5,062
5,117
5,170
5,188
5,188
5,196
13,673
13,725
14,207
14,616
14,908
14,932
14,950
14,997
15,058
15,134
15,208
5,581
5,820
5,958
6,137
6,189
6,184
6,173
6,150
6,141
6,130
6,123
53
Office f or National Statistics
Health Statistics Quar terly 01
Table 1.5
continued
Spring 1999
Population: age and sex
Numbers (thousands)
Constituent countries of the United Kingdom
Age group
Mid-year
All ages
Under 1
1–4
5–14
15–24
25–34
35–44
45–59
60–64
65–74
75–84
85–89
90 and
over
Under
16
16–
64/59
65/60
and over
England
Persons
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
46,412
46,660
46,821
47,342
48,208
48,378
48,533
48,707
48,903
49,089
49,284
739
551
598
618
663
657
633
634
615
603
616
2,996
2,491
2,235
2,385
2,574
2,606
2,611
2,601
2,589
2,543
2,490
7,272
7,513
6,678
5,885
5,916
6,014
6,125
6,186
6,231
6,298
6,364
6,731
6,688
7,440
7,692
6,840
6,601
6,394
6,246
6,158
6,054
5,980
5,840
6,599
6,703
6,717
7,599
7,724
7,803
7,873
7,909
7,922
7,873
5,421
5,298
5,663
6,484
6,665
6,541
6,508
6,545
6.622
6,761
6,933
8,515
8,199
7,948
7,672
7,920
8,193
8,415
8,593
8,738
8,856
8,956
2,690
2,616
2,449
2,559
2,399
2,381
2,356
2,329
2,310
2,299
2,308
3,976
4,274
4,347
4,199
4,222
4,252
4,308
4,355
4,270
4,210
4,164
1,816
1,972
2,249
2,518
2,645
2,612
2,541
2,481
2,568
2,629
2,670
306
332
362
438
543
564
587
600
613
620
623
109
127
149
174
220
235
253
265
280
293
308
11,648
11,293
10,285
9,608
9,711
9,811
9,913
10,012
10,048
10,053
10,071
27,128
27,275
28,133
29,070
29,627
29,678
29,720
29,803
29,946
30,114
30,275
7,636
8,092
8,403
8,665
8,870
8,889
8,899
8,893
8,909
8,922
8,939
Males
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
22,569
22,728
22,795
23,086
23,588
23,688
23,782
23,882
24,008
24,129
24,251
380
283
306
317
340
337
325
326
315
309
316
1,537
1,280
1,147
1.224
1,322
1,336
1,338
1,332
1,327
1,304
1,278
3,734
3,858
3,430
3,026
3,043
3,091
3,146
3,175
3,198
3,233
3,265
3,421
3,413
3,790
3,931
3,507
3,385
3,282
3,207
3,160
3,106
3,067
2,965
3,339
3,377
3,392
3,859
3,928
3,974
4,012
4,039
4,051
4,030
2,733
2,686
2,856
3,255
3,344
3,281
3,267
3,286
3,333
3,410
3,504
4,161
4,031
3,938
3,822
3,957
4,093
4,202
4,289
4,360
4,420
4,468
1,261
1,228
1,154
1,224
1,159
1,154
1,145
1,136
1,132
1,129
1,134
1,671
1,849
1,902
1,853
1,900
1,920
1,951
1,977
1,948
1,931
1,921
599
649
777
900
975
968
945
926
969
1,002
1,027
107
85
89
109
143
151
160
166
173
177
181
25
27
30
33
41
44
48
50
54
58
62
5,982
5,798
5,280
4,937
4,991
5,039
5,089
5,137
5,155
5,158
5,168
14,209
14,320
14,717
15,254
15,539
15,566
15,590
15,626
15,709
15,803
15,893
2,377
2,610
2,798
2,895
3,058
3,082
3,103
3,119
3,144
3,167
3,191
Females
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
23,843
23,932
24,026
24,257
24,620
24,691
24,751
24,825
24,896
24,960
25,033
359
269
292
301
324
320
309
309
300
293
300
1,459
1,211
1,088
1,161
1,253
1,270
1,273
1,268
1,262
1,239
1,213
3,538
3,656
3,248
2,859
2,873
2,923
2,979
3,010
3,033
3,065
3,099
3,310
3,275
3,650
3,761
3,333
3,215
3,111
3,039
2,998
2,948
2,913
2,875
3,260
3,327
3,325
3,739
3,795
3,829
3,862
3,871
3,872
3,843
2,688
2,612
2,807
3,229
3,322
3,260
3,241
3,259
3,289
3,351
3,429
4,354
4,168
4,009
3,850
3,964
4,100
4,212
4,304
4,378
4,437
4,488
1,429
1,387
1,295
1,335
1,239
1,227
1,211
1,193
1,178
1,170
1,174
2,305
2,425
2,445
2,346
2,323
2,332
2,357
2,378
2,322
2,279
2,244
1,217
1,323
1,472
1,618
1,670
1,644
1,597
1,555
1,598
1,627
1,643
309
246
273
330
400
413
427
434
441
443
442
85
100
119
141
179
191
205
214
226
235
246
5,666
5,495
5,004
4,671
4,720
4,772
4,824
4,874
4,893
4,894
4,903
12,918
14,968
13,416
13,816
14,088
14,112
14,131
14,177
14,237
14,311
14,382
5,259
5,481
5,605
5,770
5,812
5,807
5,796
5,774
5,765
5,755
5,748
Wales
Persons
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
2,740
2,799
2,813
2,820
2,891
2,899
2,906
2,913
2,917
2,921
2,927
43
33
36
37
39
38
36
36
35
34
35
173
151
136
143
154
154
153
151
149
145
141
433
453
407
358
365
372
379
382
383
385
387
386
388
434
441
397
385
375
367
363
357
352
325
379
383
371
409
413
416
420
420
420
417
315
309
333
378
391
383
379
379
380
385
392
519
509
485
464
486
501
514
525
534
541
547
164
161
158
166
154
152
151
149
148
148
148
252
267
272
271
284
286
288
289
284
280
276
110
121
139
155
165
165
163
161
166
171
174
16
19
21
26
33
34
36
36
37
38
39
6
7
8
10
13
14
15
16
17
18
19
686
680
626
582
592
596
602
606
605
602
601
1,582
1,618
1,663
1,690
1,724
1,724
1,725
1,727
1,730
1,737
1,743
472
501
525
548
576
578
580
580
581
582
583
Males
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
1,329
1,361
1,365
1,370
1,407
1,411
1,417
1,422
1,425
1,428
1,433
22
17
18
19
20
19
19
19
18
17
18
89
78
70
73
79
79
78
77
76
74
72
222
233
209
185
188
191
195
196
196
197
198
194
197
221
225
203
198
193
190
188
185
182
164
193
193
187
206
208
210
213
214
214
214
158
157
168
190
195
191
189
189
190
192
196
253
249
240
231
242
250
256
262
266
269
272
76
75
73
79
74
74
73
72
72
72
72
107
114
118
119
128
129
131
131
130
128
127
38
41
48
54
60
60
60
60
62
65
67
6
5
5
7
8
9
9
10
10
10
11
1
2
2
2
2
3
3
3
4
4
4
352
350
321
300
305
306
309
311
310
308
308
827
849
871
889
904
904
905
907
910
913
917
150
162
173
181
199
201
203
204
206
207
208
Females
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
1,412
1,438
1,448
1,450
1,484
1,488
1,490
1,491
1,491
1,493
1,494
21
16
18
18
19
19
18
18
17
16
17
85
73
66
70
75
75
75
74
73
71
69
211
220
199
173
177
181
185
186
187
188
189
191
191
213
217
194
187
181
177
175
172
170
161
187
190
184
203
205
206
207
206
206
204
157
153
165
188
195
192
190
190
190
193
196
265
260
246
233
244
252
258
263
268
272
275
88
86
85
87
80
79
78
77
76
76
76
146
152
154
152
156
157
157
158
154
151
148
73
80
91
101
105
105
103
101
104
106
107
16
14
16
20
25
25
26
27
27
28
28
4
6
6
8
11
12
13
13
14
15
15
335
330
305
282
288
290
293
295
295
294
293
755
770
791
800
820
820
819
820
820
824
826
322
339
352
367
377
377
377
376
376
375
375
Office for National Statistics
54
Health Statistics Quar terly 01
Table 1.5
continued
Population: age and sex
Numbers (thousands)
Spring 1999
Constituent countries of the United Kingdom
Age group
Mid-year
All ages
Under 1
1–4
5–14
15–24
25–34
35–44
45–59
60–64
65–74
75–84
85–89
90 and
over
Under
16
16–
64/59
65/60
and over
Scotland
Persons
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
5,236
5,233
5,180
5,123
5,107
5,111
5,120
5,132
5,137
5,128
5,123
86
67
69
66
66
67
64
63
61
59
60
358
291
249
257
259
260
260
261
261
255
247
912
904
780
657
634
641
648
651
649
647
649
781
806
875
870
754
727
705
690
677
663
651
617
692
724
742
809
817
825
829
827
821
809
612
591
603
665
699
692
694
703
715
728
744
926
897
880
849
853
873
888
902
911
919
924
294
282
260
273
265
264
262
260
258
256
255
430
460
460
435
441
445
451
456
450
446
443
183
202
232
251
259
256
249
243
250
255
259
29
31
35
41
50
51
52
53
55
56
56
9
11
14
15
19
20
21
21
22
23
24
1,440
1,352
1,188
1,063
1,023
1,025
1,032
1,038
1,036
1,028
1,021
2,986
3,023
3,110
3,171
3,174
3,174
3,176
3,183
3,187
3,185
3,185
810
858
882
889
910
912
912
911
914
915
917
Males
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
2,516
2,517
2,495
2,474
2,470
2,473
2,479
2,486
2,489
2,486
2,484
44
34
35
34
34
34
33
32
31
30
31
184
149
128
131
133
133
133
133
133
130
126
467
463
400
337
325
328
332
333
332
331
332
394
408
445
445
385
371
360
353
346
339
333
306
347
364
375
407
412
415
418
416
413
407
299
290
298
332
348
344
345
350
356
362
371
440
429
424
410
415
426
434
441
446
450
453
134
128
118
127
124
124
123
122
121
121
121
176
193
194
184
192
194
197
200
198
197
196
60
65
77
86
91
90
88
86
90
92
95
8
8
8
10
12
13
13
14
14
15
15
2
2
3
3
3
4
4
4
4
4
5
738
693
610
545
524
525
528
531
530
526
522
1,530
1,556
1,603
1,647
1,646
1,646
1,648
1,651
1,653
1,651
1,651
247
269
282
283
299
301
302
304
307
309
311
Females
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
2,720
2,716
2,685
2,649
2,637
2,638
2,642
2,646
2,647
2,642
2,638
42
32
33
32
32
33
32
31
30
29
29
174
142
121
126
126
126
127
128
128
125
121
445
440
380
320
309
313
316
318
317
316
317
387
398
430
425
369
356
345
337
331
324
318
311
345
359
368
402
406
409
412
411
408
403
313
301
305
334
351
348
349
353
359
366
374
485
468
456
439
437
447
454
461
465
469
471
160
154
142
146
141
141
139
138
136
135
135
254
267
265
250
249
251
254
256
252
249
247
122
137
155
165
168
165
161
157
160
163
164
20
23
27
32
37
38
39
40
40
41
41
7
8
11
12
16
16
17
17
18
19
19
701
659
579
518
499
500
504
507
506
502
498
1,455
1,468
1,506
1,525
1,528
1,527
1,528
1,532
1,534
1,534
1,534
563
589
600
606
611
611
609
607
607
606
605
Northern Ireland
Persons
1971
1,540
1976
1,524
1981
1,538
1986
1,567
1991
1,601
1992
1,618
1993
1,632
1994
1,642
1995
1,649
1996
1,663
1997
1,675
31
26
27
28
26
26
25
25
24
24
25
126
111
104
108
104
104
105
104
102
101
99
299
306
282
262
260
263
265
265
264
264
265
247
243
271
280
256
256
250
251
252
251
236
189
198
200
218
240
245
251
253
253
257
262
165
163
175
189
200
202
205
209
213
218
227
243
231
227
225
241
247
253
257
261
266
275
74
73
68
69
70
70
70
70
69
69
72
106
111
116
113
119
121
123
123
123
123
123
51
53
55
62
67
67
67
68
69
71
72
7
8
9
10
12
13
13
13
14
14
14
2
2
4
4
5
5
5
5
5
5
5
483
471
442
426
415
417
419
419
417
416
412
853
840
874
915
945
957
968
976
985
999
1,011
205
212
221
226
241
244
246
246
247
249
252
Males
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
755
754
754
768
781
791
797
802
805
812
821
16
13
14
14
13
13
13
12
12
12
13
64
58
53
55
54
53
54
53
52
51
51
152
157
145
134
133
134
135
136
135
136
136
127
127
140
146
132
132
129
130
130
129
123
95
102
102
109
119
121
124
125
125
127
130
81
81
87
94
100
101
102
104
105
108
112
116
111
109
109
118
121
124
126
128
131
136
36
34
32
32
32
32
33
33
32
33
34
45
47
50
48
52
53
54
54
54
54
55
19
19
20
22
24
24
25
25
26
26
27
2
3
3
3
3
3
3
4
4
4
4
1
0
1
1
1
1
1
1
1
1
1
246
242
227
218
213
214
214
214
214
213
211
441
442
454
476
487
495
500
504
508
515
523
67
70
73
74
81
82
83
83
84
85
86
Females
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
786
769
783
798
820
828
835
840
844
851
854
15
13
13
13
13
13
12
12
12
12
12
62
53
51
52
51
51
51
51
50
49
48
147
149
137
128
127
128
129
129
129
129
129
119
116
130
133
125
124
121
121
122
121
113
95
96
98
108
121
124
127
128
128
130
132
84
81
88
95
100
101
103
105
107
111
115
126
120
118
116
123
126
129
131
133
135
139
39
38
37
37
38
38
38
37
36
36
37
61
64
66
64
67
68
69
69
69
69
68
32
33
36
40
43
43
43
43
44
45
46
5
6
7
7
9
9
10
10
10
10
11
2
2
3
3
4
4
4
4
4
4
4
237
229
215
208
203
204
204
205
203
203
200
411
398
420
439
457
463
468
472
477
484
487
138
143
148
152
160
162
163
163
163
164
166
55
Office f or National Statistics
Health Statistics Quar terly 01
Table 1.6
Spring 1999
Population: age, sex and legal marital status
Numbers (thousands)
England and Wales
Males
Mid-year
Single
Married
Divorced
16 and over
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
4,173
4,369
5,013
5,673
6,024
6,089
6,147
6,221
6,345
6,482
6,622
12,522
12,511
12,238
11,886
11,745
11,663
11,580
11,492
11,415
11,339
11,256
187
376
611
919
1,200
1,269
1,342
1,413
1,480
1,543
1,604
16–19
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
1,327
1,454
1,675
1,601
1,372
1,301
1,242
1,212
1,220
1,251
1,291
34
28
20
10
8
5
4
3
3
2
2
20–24
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
1,211
1,167
1,420
1,794
1,764
1,760
1,742
1,699
1,658
1,597
1,536
25–29
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
Females
Widowed
Total
Single
Married
682
686
698
695
731
732
732
730
729
728
726
17,563
17,941
18,559
19,173
19,699
19,753
19,801
19,855
19,968
20,091
20,209
3,583
3,597
4,114
4,613
4,822
4,871
4,906
4,958
5,058
5,171
5,292
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1,362
1,482
1,694
1,611
1,380
1,306
1,246
1,215
1,222
1,253
1,293
689
557
466
322
249
214
182
152
127
105
87
3
4
10
14
12
10
8
7
6
5
4
0
0
1
0
0
0
0
0
0
0
0
431
533
588
841
1,183
1,232
1,263
1,293
1,326
1,368
1,401
1,206
1,326
1,057
956
894
852
807
754
696
639
577
16
39
54
79
85
83
80
76
70
64
58
30–34
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
206
236
318
356
535
596
662
732
799
855
903
1,244
1,338
1,451
1,200
1,206
1,202
1,194
1,187
1,177
1,155
1,125
35–44
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
317
286
316
397
482
497
522
556
601
657
725
2,513
2,442
2,519
2,743
2,658
2,561
2,500
2,463
2,446
2,449
2,458
Total
population
Divorced
Widowed
Total
12,566
12,538
12,284
11,994
11,838
11,749
11,661
11,583
11,488
11,406
11,319
296
533
828
1,164
1,459
1,533
1,610
1,684
1,754
1,819
1,882
2,810
2,877
2,939
2,943
2,978
2,963
2,946
2,922
2,898
2,870
2,838
19,255
19,545
20,165
20,714
21,097
21,116
21,124
21,147
21,199
21,265
21,331
36,818
37,486
38,724
39,887
40,796
40,869
40,925
41,003
41,167
41,356
41,540
1,163
1,289
1,523
1,483
1,267
1,209
1,157
1,131
1,139
1,171
1,212
142
129
93
49
32
24
18
14
13
12
11
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1,305
1,419
1,616
1,533
1,300
1,233
1,175
1,145
1,152
1,183
1,224
2,666
2,901
3,310
3,144
2,680
2,539
2,421
2,360
2,374
2,436
2,517
1,904
1,728
1,896
2,130
2,025
1,984
1,933
1,858
1,791
1,707
1,628
745
725
1,007
1,382
1,421
1,434
1,432
1,416
1,404
1,369
1,333
1,113
925
811
658
490
434
381
330
282
238
204
9
16
27
32
29
26
23
20
17
15
12
2
2
2
1
1
1
1
1
0
0
0
1,869
1,667
1,847
2,072
1,941
1,895
1,838
1,767
1,703
1,622
1,549
3,773
3,395
3,744
4,203
3,966
3,879
3,770
3,625
3,495
3,329
3,177
1
2
1
1
1
1
1
1
1
1
1
1,654
1,900
1,700
1,877
2,163
2,169
2,152
2,124
2,092
2,071
2,037
215
267
331
527
800
848
880
908
936
977
1,014
1,367
1,522
1,247
1,204
1,158
1,112
1,062
1,011
947
887
818
29
65
89
113
123
124
124
122
116
109
101
4
5
4
4
2
2
2
2
2
2
2
1,614
1,859
1,671
1,847
2,083
2,087
2,069
2,044
2,002
1,975
1,935
3,267
3,758
3,372
3,724
4,246
4,256
4,220
4,168
4,094
4,045
3,972
23
55
97
125
160
167
174
179
182
181
177
3
3
3
2
2
2
2
2
2
2
3
1,475
1,632
1,869
1,683
1,903
1,968
2,032
2,100
2,160
2,194
2,207
111
118
165
206
335
375
418
467
518
560
598
1,269
1,388
1,544
1,292
1,330
1,336
1,338
1,340
1,333
1,316
1,287
34
75
129
154
189
198
205
213
218
221
222
8
8
9
6
5
5
5
5
5
5
5
1,422
1,588
1,846
1,658
1,859
1,914
1,967
2,025
2,075
2,103
2,111
2,897
3,220
3,715
3,341
3,762
3,882
3,999
4,126
4,235
4,296
4,318
48
104
178
293
388
403
423
444
464
483
503
13
12
12
12
12
11
12
12
12
13
13
2,891
2,843
3,024
3,444
3,539
3,472
3,456
3,475
3,523
3,602
3,700
201
167
170
213
280
295
316
343
374
414
459
2,529
2,427
2,540
2,816
2,760
2,669
2,612
2,587
2,568
2,575
2,593
66
129
222
350
444
456
473
491
509
527
545
48
42
41
39
34
32
31
29
29
28
28
2,845
2,765
2,972
3,419
3,517
3,452
3,431
3,449
3,480
3,544
3,625
5,736
5,608
5,996
6,863
7,056
6,924
6,887
6,925
7,003
7,146
7,325
Aged
Note: Population estimates by marital status for 1971 and 1976 are based on the 1971 Census and those for 1981 and 1986 are based on the 1981 Census and have not been rebased using
the 1991 Census.
Office for National Statistics
56
Health Statistics Quar terly 01
Table 1.6
continued
Population: age, sex and legal marital status
Numbers (thousands)
England
England and
andWales
Wales
Males
Mid-year
Spring 1999
Females
Widowed
Total
Single
Married
Divorced
Total
population
Single
Married
Divorced
Widowed
Total
45–64
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
502
496
480
461
456
468
479
489
500
512
524
4,995
4,787
4,560
4,423
4,394
4,479
4,532
4,564
4,581
4,587
4,590
81
141
218
332
456
499
544
587
630
673
715
173
160
147
141
127
125
122
120
119
118
117
5,751
5,583
5,405
5,356
5,433
5,571
5,677
5,759
5,830
5,890
5,946
569
462
386
326
292
295
297
300
305
310
318
4,709
4,568
4,358
4,221
4,211
4,308
4,376
4,422
4,452
4,473
4,494
125
188
271
388
521
568
615
659
703
746
789
733
683
620
569
503
487
471
456
440
425
412
6,136
5,901
5,635
5,504
5,527
5,658
5,759
5,837
5,900
5,954
6,013
11,887
11,484
11,040
10,860
10,960
11,228
11,436
11,596
11,730
11,844
11,959
65 and over
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
179
197
216
223
231
235
237
239
241
242
242
1,840
2,033
2,167
2,233
2,337
2,349
2,360
2,368
2,385
2,401
2,417
17
33
54
76
99
106
113
121
128
137
147
492
510
534
539
589
593
596
595
595
594
593
2,527
2,773
2,971
3,070
3,257
3,283
3,306
3,323
3,349
3,375
3,399
580
569
533
475
427
416
405
393
382
370
358
1,437
1,579
1,692
1,754
1,858
1,866
1,873
1,879
1,893
1,904
1,912
32
60
90
127
153
161
170
179
190
201
213
2,016
2,138
2,263
2,325
2,433
2,436
2,436
2,429
2,422
2,410
2,390
4,065
4,347
4,578
4,681
4,870
4,879
4,885
4,880
4,887
4,884
4,873
6,592
7,119
7,548
7,752
8,127
8,162
8,191
8,203
8,237
8,259
8,272
See note opposite.
57
Office f or National Statistics
Health Statistics Quar terly 01
Table 2.1
Year and
quarter
Spring 1999
Vital statistics summary
Numbers (thousands) and rates
All live
births
Number
Rate*
Live births
outside marriage
Constituent countries of the United Kingdom
Marriages
Number
Rate†
Number
Rate**
Divorces
Deaths
Number
Rate††
Number
Rate*
Infant
mortality***
Number
Neonatal
mortality†††
Rate†
Number
Rate†
Perinatal
mortality****
Number
Rate††††
United Kingdom
1971
1976
1981
1986
1991
1992
901.6
675.5
730.8
755.0
792.5
781.0
16.1
12.0
13.0
13.3
13.7
13.5
73.9
61.1
91.3
158.5
236.1
240.8
82
90
125
210
298
308
459.4
406.0
397.8
393.9
349.7
356.0
..
..
49.4
43.5
36.0
..
79.6
135.4
156.4
168.2
173.5
175.1
..
..
11.3
12.5
13.0
..
645.1
680.8
658.0
660.7
646.2
634.2
11.5
12.1
11.7
11.6
11.3
10.9
16.2
9.79
8.16
7.18
5.82
5.14
17.9
14.5
11.2
9.5
7.4
6.6
10.8
6.68
4.93
4.00
3.46
3.37
12.0
9.9
6.7
5.3
4.4
4.3
20.7
12.3
8.79
7.31
6.45
6.01
22.6
18.0
12.0
9.6
8.1
7.7
1993
1994
1995
1996
1997
761.7
750.7
732.0
733.4
725.8
13.1
12.9
12.5
12.5
12.3
241.8
240.1
245.7
260.4
266.7
318
320
336
355
367
341.6
331.2
322.3
317.5‡
308.7‡
..
..
..
..
..
180.0
173.6
170.0
171.7‡
161.1‡
..
..
..
..
..
658.5
627.6
645.5
636.0
629.7‡
11.3
10.7
11.0
10.8
10.7‡
4.83
4.63
4.52
4.50
4.28
6.3
6.2
6.2
6.1
5.9
3.18
3.09
3.05
3.00
2.83
4.2
4.1
4.2
4.1
3.9
6.73
6.74
6.52
6.41
6.08
8.8
9.0
8.9
8.7
8.3
1996 Sept
1996 Dec
191.4
185.6
13.0
12.6
68.8
68.6
359
369
129.4‡
55.8‡
..
..
143.2
166.4
9.7
11.3
1.13
1.12
5.9
6.0
0.80
0.72
4.2
3.9
1.66
1.55
8.6
8.3
1997 March
1997 June
1997 Sept
1997 Dec
178.4
184.7
185.8
176.9
12.3
12.6
12.5
11.9
65.5
66.2
68.7
66.3
367
358
370
375
39.0‡
86.8‡
128.6‡
54.3‡
..
..
..
..
182.7‡
146.6‡
141.2‡
159.3‡
12.6‡
10.0‡
9.5‡
10.7‡
1.10
1.10
1.01
1.08
6.1
6.0
5.4
6.1
0.70
0.73
0.69
0.71
3.9
4.0
3.7
4.0
1.53
1.53
1.49
1.53
8.5
8.2
8.0
8.6
1998 March
1998 June
1998 Sept
175.6‡
178.6‡
12.1‡
12.1‡
166.7‡
152.3‡
142.8‡
11.5‡
10.3‡
9.6‡
1.02‡
0.97‡
5.8‡
5.4‡
0.67‡
0.65‡
3.8‡
3.7‡
1.51‡
1.45‡
8.5‡
8.1‡
England and Wales
1971
783.2
1976
584.3
1981
634.5
1986
661.0
1991
699.2
1992
689.7
15.9
11.8
12.8
13.2
13.7
13.4
65.7
53.8
81.0
141.3
211.3
215.2
84
92
128
214
302
312
404.7
358.6
352.0
347.9
306.8
311.6
69.0
57.7
49.6
43.5
35.6
35.8
74.4
126.7
145.7
153.9
158.7
160.4
5.9
10.1
11.9
12.9
13.5
13.7
567.3
598.5
577.9
581.2
570.0
558.3
11.5
12.1
11.6
11.6
11.2
10.9
13.7
8.34
7.02
6.31
5.16
4.54
17.5
14.3
11.1
9.6
7.4
6.6
9.11
5.66
4.23
3.49
3.05
2.96
11.6
9.7
6.7
5.3
4.4
4.3
17.6
10.5
7.56
6.37
5.65
5.24
22.3
17.7
11.8
9.6
8.0
7.6
1993
1994
1995
1996
1997
673.5
664.7
648.1
649.5
642.1
13.1
12.9
12.5
12.5
12.3
216.5
215.5
219.9
232.7
237.9
322
324
339
358
370
299.2
291.1
283.0
279.0‡
271.1‡
33.9
32.6
31.0
30.0‡
28.6‡
165.0
158.2
155.5
157.1‡
146.7‡
14.2
13.7
13.6
13.8‡
13.0‡
578.8
553.2
569.7
560.1
555.3
11.3
10.7
11.0
10.8
10.6‡
4.24
4.10
3.98
3.99
3.83
6.3
6.2
6.1
6.1
6.0
2.80
2.74
2.70
2.68
2.54
4.2
4.1
4.2
4.1
3.9
6.03
5.95
5.70
5.62
5.40
8.9
8.9
8.8
8.6
8.4
1996 Sept
1996 Dec
169.9
164.2
13.0
12.6
61.6
61.2
363
373
114.1‡
47.9‡
48.8‡
20.5‡
40.3‡
34.8‡
14.1‡
12.2‡
125.9
147.1
9.6
11.2
1.01
1.00
5.9
6.1
0.72
0.65
4.2
3.9
1.48
1.37
8.7
8.3
1997 March
1997 June
1997 Sept
1997 Dec
158.1
162.9
164.4
156.7
12.3
12.5
12.5
11.9
58.5
58.8
61.3
59.3
370
361
373
378
34.7‡
76.4‡
112.9‡
47.1‡
14.8‡
32.3‡
47.2‡
19.7‡
34.9‡
39.6‡
37.2‡
35.0‡
12.5‡
14.1‡
13.1‡
12.3‡
161.9
128.6
124.3
140.4
12.6
9.9
9.4
10.7
0.98
0.99
0.91
0.96
6.2
6.0
5.5
6.1
0.63
0.65
0.62
0.64
4.0
4.0
3.8
4.1
1.36
1.36
1.32
1.36
8.6
8.3
8.0
8.6
1998 March
1998 June
1998 Sept
155.4‡
158.3‡
165.8‡
12.1‡
12.2‡
12.6‡
58.4‡
58.3‡
63.1‡
375‡
368‡
381‡
146.9‡
134.0‡
125.6‡
11.4‡
10.3‡
9.6‡
0.93‡
0.83‡
6.0‡
5.3‡
0.60‡
0.57‡
3.9‡
3.6‡
1.34‡
1.28‡
8.6‡
8.1‡
England
1971
1976
1981
1986
1991
1992
740.1
550.4
598.2
623.6
660.8
651.8
15.9
11.8
12.8
13.2
13.7
13.5
62.6
50.8
76.9
133.5
198.9
202.4
85
92
129
214
301
311
382.3
339.0
332.2
328.4
290.1
295.0
..
..
..
..
..
..
..
..
..
146.0
150.1
151.5
..
..
..
..
..
..
532.4
560.3
541.0
544.5
534.0
522.7
11.5
12.0
11.6
11.5
11.2
10.8
12.9
7.83
6.50
5.92
4.86
4.26
17.5
14.2
10.9
9.5
7.3
6.5
8.58
5.32
3.93
3.27
2.87
2.79
11.6
9.7
6.6
5.2
4.3
4.3
16.6
9.81
7.04
5.98
5.33
4.95
22.1
17.6
11.7
9.5
8.0
7.6
1993
1994
1995
1996
1997
636.5
629.0
613.2
614.2
607.2
13.1
13.0
12.5
12.5
12.3
203.6
202.7
206.8
218.2
223.0
320
322
337
355
367
283.3
275.5
268.3
264.2‡
256.5‡
..
..
..
..
..
156.1
149.6
147.5
148.7‡
138.7‡
..
..
..
..
..
541.1
517.6
532.6
524.0
519.1
11.1
10.6
10.9
10.7
10.5
4.00
3.83
3.74
3.74
3.60
6.3
6.1
6.1
6.1
5.9
2.65
2.57
2.55
2.53
2.39
4.2
4.1
4.2
4.1
3.9
5.70
5.58
5.41
5.36
5.11
8.9
8.8
8.8
8.7
8.4
1996 Sept
1996 Dec
160.7
155.1
13.0
12.6
57.8
57.4
360
370
108.0‡
45.3‡
..
..
38.0‡
32.8‡
..
..
117.6
135.3
9.5
11.0
0.94
0.94
5.8
6.0
0.68
0.62
4.2
4.0
1.42
1.31
8.8
8.4
1997 March
1997 June
1997 Sept
1997 Dec
149.3
154.2
155.5
148.2
12.3
12.6
12.5
11.9
54.8
55.1
57.6
55.5
367
358
370
374
32.8‡
72.4‡
106.7‡
44.6‡
..
..
..
..
33.0‡
37.5‡
35.1‡
33.1‡
..
..
..
..
151.5
120.1
116.1
131.4
12.5
9.8
9.3
10.6
0.91
0.91
0.86
0.91
6.1
5.9
5.5
6.1
0.58
0.61
0.60
0.61
3.9
3.9
3.8
4.1
1.29
1.28
1.26
1.28
8.6
8.2
8.0
8.6
1998 March
1998 June
1998 Sept
147.2‡
150.0‡
157.1‡
12.1‡
12.2‡
12.6‡
54.8‡
54.8‡
59.2‡
372‡
365‡
377‡
137.6‡
125.4‡
117.5‡
11.3‡
9.6‡
8.9‡
0.87‡
0.77‡
5.9‡
5.2‡
0.57‡
0.54‡
3.9‡
3.6‡
1.26‡
1.21‡
8.5‡
8.0‡
* Per 1,000 population of all ages. † Per 1,000 live births. ** Persons marrying per 1,000 unmarried population 16 and over. †† Per 1,000 married population. *** Deaths under 1 year.
††† Deaths under 4 weeks. **** Stillbirths and deaths under 1 week. In October 1992 the legal definition of a stillbirth was changed, from baby born dead after 28 completed weeks of
gestation or more, to one born dead after 24 completed weeks of gestation or more. †††† Per 1,000 live and stillbirths. ‡ Provisional.
Office for National Statistics
58
Health Statistics Quar terly 01
Table 2.1
continued
Year and
quarter
Vital statistics summary
Numbers (thousands) and rates
All live
births
Number
Wales
1 971
Rate*
Live births
outside marriage
Spring 1999
Constituent countries of the United Kingdom
Marriages
Number
Rate†
Number
Rate**
Divorces
Number
Deaths
Rate††
Number
Rate*
Infant
mortality***
Number
Neonatal
mortality†††
Rate†
Number
Rate†
Perinatal
mortality****
Number
Rate††††
1976
1981
1986
1991
1992
43.1
33.4
35.8
37.0
38.1
37.5
15.7
11.9
12.7
13.1
13.2
12.9
3.1
2.9
4.0
7.8
12.3
12.8
71
86
112
211
323
340
22.4
19.5
19.8
19.5
16.6
16.6
..
..
..
..
..
..
..
..
..
7.9
8.6
8.9
..
..
..
..
..
..
34.8
36.3
35.0
34.7
34.1
33.8
12.7
13.0
12.4
12.3
11.8
11.7
0.79
0.46
0.45
0.35
0.25
0.23
18.4
13.7
12.6
9.5
6.6
6.0
0.53
0.32
0.29
0.21
0.16
0.14
12.3
9.6
8.1
5.6
4.1
3.8
1.07
0.64
0.51
0.38
0.30
0.26
24.4
19.0
14.1
10.3
7.9
7.0
1993
1994
1995
1996
1997
36.6
35.4
34.5
34.9
34.5
12.6
12.2
11.8
11.9
11.8
12.9
12.7
13.1
14.4
14.8
352
360
381
412
428
15.9
15.5
14.7
14.8‡
14.5‡
..
..
..
8.9
8.6
8.0
8.4‡
8.0‡
..
..
..
..
..
35.9
33.9
35.6
34.6
34.6
12.4
11.6
12.2
11.8
11.8
0.20
0.22
0.20
0.20
0.20
5.5
6.1
5.9
5.6
5.9
0.12
0.14
0.13
0.13
0.14
3.3
4.1
3.9
3.6
3.9
0.30
0.33
0.27
0.26
0.28
8.2
9.3
7.9
7.5
7.9
1996 Sept
1996 Dec
9.1
9.0
12.4
12.3
3.8
3.8
413
425
6.1‡
2.5‡
..
..
2.1‡
2.0‡
..
..
7.9
8.9
10.1
12.1
0.06
0.04
6.4
4.7
0.04
0.02
3.8
2.5
0.06
0.06
7.0
6.7
1997 March
1997 June
1997 Sept
1997 Dec
8.7
8.6
8.8
8.4
12.0
11.8
12.0
11.4
3.7
3.6
3.7
3.7
427
422
420
446
1.8‡
4.0‡
6.2‡
2.5‡
..
..
..
..
1.9‡
2.1‡
2.0‡
1.9‡
..
..
..
..
10.1
8.1
7.7
8.7
14.0
11.1
10.5
11.8
0.06
0.06
0.04
0.04
6.8
7.4
4.9
4.5
0.04
0.05
0.02
0.03
4.6
5.2
2.5
3.3
0.06
0.08
0.06
0.07
7.4
9.4
7.2
7.7
1998 March
1998 June
1998 Sept
8.3‡
8.4‡
8.8‡
11.4‡
11.4‡
11.9‡
3.6‡
3.6‡
3.9‡
435‡
428‡
444‡
12.5‡
11.3‡
10.5‡
0.04‡
0.04‡
5.3‡
5.3‡
0.03‡
0.03‡
3.4‡
3.7‡
0.07‡
0.07‡
7.9‡
8.5‡
9.0‡
8.3‡
7.8
Scotland
1971
1976
1981
1986
1991
1992
86.7
64.9
69.1
65.8
67.0
65.8
16.6
12.5
13.4
12.9
13.1
12.9
7.0
6.0
8.5
13.6
19.5
20.0
81
93
122
206
291
303
42.5
37.5
36.2
35.8
33.8
35.1
64.1
53.8
47.5
42.8
38.7
39.9
4.8
8.1
9.9
12.8
12.4
12.5
3.9
6.5
8.0
10.7
10.6
10.7
61.6
65.3
63.8
63.5
61.0
60.9
11.8
12.5
12.3
12.4
12.0
11.9
1.72
0.96
0.78
0.58
0.47
0.45
19.9
14.8
11.3
8.8
7.1
6.8
1.17
0.67
0.47
0.34
0.29
0.30
13.5
10.3
6.9
5.2
4.4
4.6
2.15
1.20
0.81
0.67
0.58
0.60
24.5
18.3
11.6
10.2
8.6
9.0
1993
1994
1995
1996
1997
63.3
61.7
60.1
59.3
59.4
12.4
12.0
11.7
11.6
11.6
19.9
19.2
20.3
21.4
22.4
313
312
337
360
377
33.4
31.5
30.7
30.2
29.6
37.6
35.1
33.7
32.8
31.7
12.8
13.1
12.2
12.3
12.2
11.0
11.4
10.7
10.9
11.0
64.0
59.3
60.5
60.7
59.5
12.5
11.6
11.8
11.8
11.6
0.41
0.38
0.38
0.37
0.32
6.5
6.2
6.2
6.2
5.3
0.25
0.25
0.24
0.23
0.19
4.0
4.0
4.0
3.9
3.2
0.61
0.56
0.58
0.55
0.47
9.6
9.0
9.6
9.2
7.8
1996 Sept
1996 Dec
15.1
15.4
11.7
12.0
5.5
5.7
366
368
11.7
6.5
50.6
28.2
3.1
3.0
10.8
10.6
13.8
15.5
10.7
12.1
0.09
0.09
5.8
5.6
0.06
0.06
3.9
3.8
0.13
0.14
8.2
9.1
1997 March
1997 June
1997 Sept
1997 Dec
14.3
15.4
15.2
14.6
11.3
12.1
11.7
11.3
5.4
5.6
5.7
5.6
376
370
379
382
3.4
8.3
12.1
5.8
14.7
35.6
51.4
24.6
2.8
3.3
3.1
3.0
10.2
11.7
11.1
10.8
16.6
14.4
13.4
15.1
13.1
11.3
10.4
11.7
0.10
0.09
0.06
0.07
6.7
5.7
4.0
4.9
0.06
0.05
0.04
0.04
3.9
3.4
2.8
2.6
0.12
0.11
0.11
0.12
8.4
7.3
7.4
8.1
1998 March
1998 June
1998 Sept
14.2‡
14.3‡
11.2‡
11.2‡
3.5‡
8.4‡
11.9‡
15.0‡
35.9‡
50.6‡
15.7‡
14.4‡
13.6‡
12.5‡
11.3‡
10.5‡
0.06‡
0.09‡
4.2‡
6.4‡
0.04‡
0.06‡
2.8‡
4.0‡
0.12‡
0.12‡
8.1‡
8.5‡
Northern Ireland
1971
1976
1981
1986
1991
1992
31.8
26.4
27.3
28.2
26.3
25.6
20.7
17.3
17.8
18.0
16.5
15.9
1.2
1.3
1.9
3.6
5.3
5.6
38
50
70
127
202
219
12.2
9.9
9.6
10.2
9.2
9.4
..
..
45.4
..
37.7
..
0.3
0.6
1.4
1.5
2.3
2.3
..
..
4.2
..
6.8
..
17.6
17.0
16.3
16.1
15.1
15.0
12.8
11.2
10.6
10.3
9.4
9.3
0.72
0.48
0.36
0.36
0.19
0.15
22.7
18.3
13.2
13.2
7.4
6.0
0.51
0.35
0.23
0.23
0.12
0.10
15.9
13.3
8.3
8.3
4.6
4.1
0.88
0.59
0.42
0.42
0.22
0.21
27.2
22.3
15.3
15.3
8.4
8.2
1993
1994
1995
1996
1997
24.9
24.3
23.9
24.6
24.3
15.3
14.9
14.5
14.8
14.5
5.5
5.4
5.5
6.4
6.4
219
220
231
259
266
9.0
8.7
8.6
8.3
8.1
..
..
..
..
..
2.2
2.3
2.3
2.3
2.2
..
..
..
..
..
15.6
15.1
15.3
15.2
15.0
9.6
9.2
9.3
9.1
9.0
0.18
0.15
0.17
0.14
0.14
7.1
6.1
7.1
5.8
5.6
0.12
0.10
0.13
0.09
0.10
4.9
4.2
5.5
3.7
4.2
0.22
0.24
0.25
0.23
0.21
8.8
9.7
10.4
9.4
8.6
1996 Sept
1996 Dec
6.4
6.0
14.4
14.8
1.6
1.6
251
271
3.6
1.4
..
..
3.5
3.8
8.3
9.0
0.04
0.03
5.5
5.6
0.02
0.02
3.8
3.3
0.05
0.05
8.5
7.7
1997 March
1997 June
1997 Sept
1997 Dec
6.1
6.3
6.3
5.6
14.7
15.2
14.8
13.3
1.6
1.7
1.7
1.5
269
261
269
264
1.0
2.1
3.6
1.4
..
..
..
..
4.2‡
3.6‡
3.4‡
3.7‡
10.2
8.7
8.2
8.9
0.02
0.05
0.04
0.05
3.6
4.9
5.7
8.6
0.02
0.02
0.03
0.03
3.1
3.6
4.2
6.1
0.05
0.06
0.05
0.06
7.6
8.8
8.1
10.3
1998 March
1998 June
1998 Sept
5.9‡
6.1‡
6.2‡
14.3‡
14.8‡
14.7
1.7‡
1.7‡
1.8‡
286‡
272‡
288‡
4.1‡
3.8‡
3.4‡
9.9
9.2
8.1
0.04‡
0.04‡
5.8‡
7.1‡
0.03‡
0.03‡
4.3‡
4.4‡
0.05‡
0.05‡
8.8‡
7.7‡
Notes: 1. Rates for the most recent quarters will be particularly subject to revision, even when standard detail is given, as they are based on provisional numbers or on estimates derived from
events registered in the period. 2. Figures for England and Wales represent the numbers of deaths registered in each year up to 1992, and the number of deaths occurring in each
year from 1993. Provisional figures are registrations. 3.The marriage and divorce rates for 1991 onwards differ in part from those previously published because of a revision of the
denominators. 4. From 1972 births for England and Wales are excluded if the mother was usually resident outside England and Wales, but included in the total for the United Kingdom.
5. From 1972 deaths for England and for Wales separately exclude deaths to persons usually resident outside England and Wales, but these deaths are included in the totals for England
and Wales combined, and the United Kingdom.
59
Office f or National Statistics
Health Statistics Quar terly 01
Table 2.2
Spring 1999
Key demographic and health indicators
Numbers (thousands), rates, percentages, mean age
Constituent countries of the United Kingdom
Dependency ratio
Population
Live births
Deaths
Children*
Elderly†
Live births
TPFR**
Expectation of
life (in years)
at birth
Outside
marriage as
percentage
of total
live births
Mean age
of mother
at birth
(years)
Agestandardised
mortality
rate††
Males
Females
Infant
mortality
rate***
68.8
69.6
70.8
71.9
73.2
73.4
73.7
73.9
74.1
74.2‡
75.0
75.2
76.8
77.7
78.8
78.9
79.1
79.2
79.4
79.5‡
17.9
14.5
11.2
9.5
7.4
6.6
8.3
6.2
6.2
6.1
5.9‡
71.1
72.0
73.4
73.7
74.0
74.1
74.4
74.6
77.0
77.9
79.0
79.1
79.3
79.4
79.6
79.7
United Kingdom
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
55,928.0
56,216.1
56,352.2
56,851.9
57,807.9
58,006.5
58,191.2
58,394.6
58,605.8
58,801.5
59,008.6
901.6
675.5
730.8
755.0
792.5
781.0
761.7
750.7
732.0
733.4
725.8
645.1
680.8
658.0
660.7
646.2
634.2
658.5
627.6
645.5
636.0
629.7
43.8
42.1
37.1
33.5
33.1
33.3
33.6
33.8
33.8
33.6
33.4
28.0
29.5
29.7
29.6
29.9
29.9
29.9
29.8
29.7
29.6
29.5
2.41
1.74
1.82
1.78
1.82
1.79
1.76
1.74
1.71
1.72
1.72
8.2
9.0
12.5
21.0
29.8
30.8
31.8
32.0
33.6
35.5
36.7
26.2
26.4
26.8
27.0
27.6
27.9
28.1
28.4
28.5
28.6
28.8
10,448
10,486
9,506
8,897
8,107
7,860
8,037
7,622
7,706
7,522
7,370
England
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
46,411.7
46,659.9
46,820.8
47,342.4
48,208.1
48,378.3
48,532.7
48,707.5
48,903.4
49,089.1
49,284.2
740.1
550.4
598.2
623.6
660.8
651.8
636.5
629.0
613.2
614.2
607.2
532.4
560.3
541.0
544.5
534.0
522.7
541.1
517.6
532.6
524.0
519.1
42.9
41.4
36.4
33.1
32.8
33.1
33.4
33.6
33.6
33.4
33.3
28.1
29.7
29.9
29.8
29.9
30.0
29.9
29.8
29.8
29.6
29.5
2.37
1.70
1.79
1.87
1.81
1.79
1.76
1.74
1.71
1.73
1.72
8.5
9.2
12.9
21.4
30.1
31.1
32.0
32.2
33.7
35.5
36.7
26.4
26.8
27.0
27.7
27.9
28.1
28.4
28.6
28.7
28.2
10,278
10,271
9,298
8,694
7,941
7,678
7,825
7,440
7,526
7,333
7,190
Wales
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
2,740.3
2,799.3
2,813.5
2,819.6
2,891.5
2,898.5
2,906.5
2,913.0
2,916.8
2,921.1
2,926.9
43.1
33.4
35.8
37.0
38.1
37.5
36.6
35.4
34.5
34.9
34.5
34.8
36.3
35.0
34.7
34.1
33.8
35.9
33.9
35.6
34.6
34.6
43.4
42.0
37.6
34.4
34.4
34.6
34.9
35.1
35.0
34.7
34.5
29.8
30.9
31.6
32.5
33.4
33.6
33.6
33.6
33.6
33.5
33.5
2.44
1.79
1.87
1.86
1.88
1.87
1.84
1.79
1.78
1.82
1.82
7.2
8.7
11.2
21.1
32.3
34.0
35.2
36.0
38.1
41.2
42.8
26.0
26.6
26.5
27.0
27.3
27.4
27.7
27.8
27.8
28.0
11,175
10,858
9,846
9,012
8,074
7,886
8,227
7,753
7,953
7,664
7,578
Scotland
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
5,235.6
5,233.4
5,180.2
5,123.0
5,107.0
5,111.2
5,120.2
5,132.4
5,136.6
5,128.0
5,122.5
86.7
64.9
69.1
65.8
67.0
65.8
63.3
61.7
60.1
59.3
59.4
61.6
65.3
63.8
63.5
61.0
60.9
64.0
59.3
60.5
60.7
59.5
48.2
44.7
38.2
33.5
32.2
32.3
32.5
32.6
32.5
32.3
32.0
27.1
28.4
28.4
28.0
28.7
28.7
28.7
28.6
28.7
28.7
28.8
2.53
1.80
1.84
1.67
1.70
1.67
1.62
1.58
1.55
1.55
1.57
8.1
9.3
12.2
20.6
29.1
30.3
31.3
31.2
33.7
36.0
37.7
26.0
26.3
26.6
27.4
27.7
27.9
28.2
28.4
28.5
28.6
Northern Ireland
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
1,540.4
1,523.5
1,537.7
1,566.8
1,601.4
1,618.4
1,631.8
1,641.7
1,649.0
1,663.3
1,675.0
31.8
26.4
27.3
28.2
26.3
25.6
24.9
24.3
23.9
24.6
24.3
17.6
17.0
16.3
16.1
15.1
15.0
15.6
15.1
15.3
15.2
15.0
56.6
56.1
50.6
46.5
44.0
43.6
43.3
42.9
42.3
41.6
40.8
24.0
25.3
25.3
24.7
25.6
25.4
25.4
25.2
25.1
24.9
25.0
3.13
2.70
2.60
2.46
2.18
2.09
2.01
1.95
1.91
1.95
1.93
3.8
5.0
7.0
12.7
20.2
21.9
21.9
22.0
23.1
25.9
26.6
27.4
27.6
27.6
28.0
28.1
28.4
28.6
28.8
28.8
29.0
60
18.4
13.7
12.6
9.5
6.6
6.0
5.5
6.1
5.8
5.6
5.9‡
70.4
71.6
73.2
73.3
73.5
73.5
73.8
73.9
76.4
77.6
78.9
78.9
79.0
79.0
79.2
79.2
11,444
11,675
10,849
10,135
9,254
9,146
9,529
8,840
8,887
8,868
8,623
67.3
68.2
69.1
70.2
71.4
71.5
71.7
71.9
72.1
72.2‡
73.7
74.4
75.3
76.2
77.1
77.1
77.3
77.4
77.6
77.8‡
19.9
14.8
11.3
8.8
7.1
6.8
6.5
6.2
6.2
6.2
5.3‡
11,607
11,746
10,567
10,071
8,564
8,347
8,600
8,256
8,255
8,057
7,810
67.6
67.5
69.1
70.6
72.3
72.5
72.8
72.9
73.3
73.6‡
73.7
73.8
75.4
76.7
78.1
78.3
78.4
78.4
78.7
78.9‡
22.7
18.3
13.2
10.2
7.4
6.0
7.1
6.1
7.1
5.8
5.6‡
‡ Provisional.
* Percentage of children under 16 to working population (males 16–64 and females 16–59).
† Percentage of males 65 and over and females 60 and over to working population (males 16–64 and females 16–59).
** TPFR (the total period fertility rate) is the number of children that would be born to a woman if current patterns of fertility persisted throughout her childbearing life.
†† Per million population. The age-standardised mortality rate makes allowances for changes in the age structure of the population. See Notes to tables.
***Deaths under one year per 1,000 live births.
Notes: 1. Some of these indicators are also in other tables. They are brought together to make comparison easier.
2. Figures for England and Wales represent the number of deaths registered in each year up to 1992, and the number of deaths occurring in each year from 1993.
Office for National Statistics
17.5
14.2
10.9
9.5
7.3
6.5
6.3
6.1
6.1
6.1
5.9‡
Health Statistics Quar terly 01
Table 3.1
Live births: age of mother
Numbers (thousands), rates, mean age and TPFRs
England and Wales
Age of mother at birth
Year and
quarter
Spring 1999
All
ages
Under
20
20–24
25–29
30–34
Age of mother at birth
35–39
40 and
over
All
ages
Under
20
Total live births (numbers)
20–24
25–29
30–34
35–39
40 and
over
Mean
age
(years)
TPFR†
Age-specific fertility rates*
1961
1964(max)†
1966
1971
1976
1977(min)†
1981
1986
1991
811.3
876.0
849.8
783.2
584.3
569.3
634.5
661.0
699.2
59.8
76.7
86.7
82.6
57.9
54.5
56.6
57.4
52.4
249.8
276.1
285.8
285.7
182.2
174.5
194.5
192.1
173.4
248.5
270.7
253.7
247.2
220.7
207.9
215.8
229.0
248.7
152.3
153.5
136.4
109.6
90.8
100.8
126.6
129.5
161.3
77.5
75.4
67.0
45.2
26.1
25.5
34.2
45.5
53.6
23.3
23.6
20.1
12.7
6.5
6.0
6.9
7.6
9.8
89.2
92.9
90.5
83.5
60.4
58.1
61.3
60.6
63.6
37.3
42.5
47.7
50.6
32.2
29.4
28.1
30.1
33.0
172.6
181.6
176.0
152.9
109.3
103.7
105.3
92.7
89.3
176.9
187.3
174.0
153.2
118.7
117.5
129.1
124.0
119.4
103.1
107.7
97.3
77.1
57.2
58.6
68.6
78.1
86.7
48.1
49.8
45.3
32.8
18.6
18.2
21.7
24.6
32.1
15.0
13.7
12.5
8.7
4.8
4.4
4.9
4.8
5.3
27.6
27.2
26.8
26.2
26.4
26.5
26.8
27.0
27.7
2.77
2.93
2.75
2.37
1.71
1.66
1.80
1.77
1.82
1992
1993
1994
1995
1996
1997
1998
689.7
673.5
664.7
648.1
649.5
642.1
47.9
45.1
42.0
41.9
44.7
46.3
163.3
152.0
140.2
130.7
125.7
118.4
244.8
236.0
229.1
217.4
211.1
202.5
166.8
171.1
179.6
181.2
186.4
187.2
56.7
58.8
63.1
65.5
69.5
74.8
10.2
10.5
10.7
11.3
12.1
12.9
63.5
62.6
61.9
60.4
60.5
59.7
31.7
31.0
29.0
28.5
29.8
30.2
86.2
82.7
79.4
76.8
77.5
76.4
117.3
114.1
112.1
108.6
106.9
104.6
87.2
87.0
88.7
87.3
88.6
88.7
33.4
34.1
35.8
36.2
37.2
38.8
5.8
6.2
6.4
6.8
7.2
7.6
27.9
28.1
28.4
28.5
28.6
28.8
1.80
1.76
1.75
1.72
1.73
1.73
1996
Sept
Dec
169.9
164.2
11.6
12.0
33.1
32.1
55.4
52.6
48.8
46.6
17.9
17.7
3.1
3.2
61.5
61.9
30
32
80
82
109
109
90
90
37
38
7
8
28.6
28.6
1.77
1.79
1997
March
June
Sept
Dec
158.1
162.9
164.4
156.7
11.5
11.3
11.8
11.8
29.8
29.5
30.3
29.0
50.4
51.5
51.9
48.7
45.6
48.3
48.0
45.3
17.7
19.2
19.2
18.7
3.1
3.3
3.3
3.2
60.6
60.4
59.1
58.7
31
29
30
31
78
75
76
76
107
106
104
102
89
91
88
87
38
40
38
39
8
8
7
8
28.7
28.9
28.8
28.8
1.75
1.74
1.71
1.71
1998
March‡ 155.4
June‡ 158.3
11.7
11.4
27.7
27.4
47.8
48.5
46.1
48.0
18.8
19.6
3.3
3.3
59.3
58.5
31
29
75
72
103
102
90
91
39
40
8
8
28.9
29.0
1.73
1.70
* Births per 1,000 women in the age-group; all quarterly rates and total period fertility rates (TPFRs) are seasonally adjusted.
† TPFR (the total period fertility rate) is the number of children that would be born to a woman if current patterns of fertility persisted throughout her childbearing life. During the post
Second World War period the TPFR reached a maximum in 1964 and a minimum in 1997.
‡ Provisional.
Note: The rates for women of all ages, under 20, and 40 and over are based upon the populations of women aged 15–44, 15–19, and 40–44 respectively.
61
Office f or National Statistics
Health Statistics Quar terly 01
Table 3.2
Spring 1999
Live births outside marriage: age of mother and type of registration
Numbers (thousands), mean age and percentages
Age of mother at birth
Year and
quarter
All
ages
Under
20
20–24
25–29
30–34
England and Wales
Age of mother at birth
35–39
40 and
over
Mean
age
(years)
All
ages
Under
20
20–24
25–29
30–34
Registration*
35–39
40 and
over
Joint
Sole
Same Different
address† address†
Percentage of total births
As a percentage of all
births outside marriage
{
Live births outside marriage (numbers)
65.7
53.8
81.0
21.6
19.8
26.4
22.0
16.6
28.8
11.5
9.7
14.3
6.2
4.7
7.9
3.2
2.3
1.3
1.1
0.7
0.9
23.7
23.3
23.4
8.4
9.2
12.8
26.1
34.2
46.7
7.7
9.1
14.8
4.7
4.4
6.6
5.7
5.2
6.2
7.0
8.6
3.9
9.0
10.1
12.5
45.5
51.0
58.2
54.5
49.0
41.8
1986
1991
1992
141.3
211.3
215.2
39.6
43.4
40.1
54.1
77.8
77.1
27.7
52.4
55.9
13.1
25.7
28.9
5.7
9.8
10.9
1.1
2.1
2.3
23.8
24.8
25.2
21.4
30.2
31.2
69.0
82.9
83.7
28.2
44.9
47.2
12.1
21.1
22.8
10.1
16.0
17.3
12.6
18.3
19.3
14.7
21.3
22.9
46.6
54.6
55.4
19.6
19.8
20.7
33.8
25.6
23.9
1993
1994
1995
1996
1997
1998
216.5
215.5
219.9
232.7
237.9
38.2
35.9
36.3
39.3
41.1
75.0
71.0
69.7
71.1
69.4
57.5
58.5
59.6
62.3
63.3
31.4
34.0
37.0
40.5
42.2
11.9
13.4
14.4
16.2
18.2
2.5
2.7
3.0
3.2
3.7
25.4
25.8
25.9
26.0
26.2
32.2
32.4
33.9
35.8
37.0
84.8
85.5
86.6
88.0
88.7
49.4
50.6
53.3
56.5
58.6
24.4
25.5
27.4
29.5
31.3
18.4
18.9
20.4
21.7
22.5
20.2
21.2
22.0
23.4
24.3
23.5
25.2
26.2
26.7
28.6
54.8
57.5
58.1
58.1
59.6
22.0
19.8
20.1
19.9
19.3
23.2
22.7
21.8
21.9
21.3
1996 June
1996 Sept
1996 Dec
54.8
61.6
61.3
9.2
10.3
10.6
16.6
18.8
18.7
14.6
16.6
16.4
9.7
10.8
10.5
3.9
4.3
4.2
0.8
1.0
0.8
26.1
26.1
26.0
34.6
36.3
37.3
88.2
88.1
87.9
55.6
57.0
58.2
28.2
29.9
31.2
20.9
22.1
22.6
23.2
23.8
23.9
27.7
31.2
26.7
58.1
58.2
58.2
19.9
20.1
19.9
22.0
21.7
21.8
1997 March
1997 June
1997 Sept
1997 Dec
58.5
58.8
61.3
59.3
10.2
10.0
10.5
10.4
17.4
17.1
17.8
17.1
15.7
15.5
16.5
15.7
10.2
10.6
10.9
10.4
4.2
4.7
4.7
4.6
0.9
0.9
0.9
0.9
26.1
26.2
26.2
26.2
37.0
36.1
37.3
37.8
88.7
89.1
88.8
88.3
58.4
58.0
58.9
59.2
31.0
30.1
31.8
32.2
22.4
22.0
22.7
23.0
23.9
24.3
24.4
24.8
28.8
28.3
27.9
29.4
58.4
59.6
59.9
60.0
19.5
19.5
18.9
19.2
22.0
21.0
21.2
20.7
1998 March‡
1998 June‡
58.4
58.3
10.4
10.2
16.5
16.2
15.3
15.4
10.6
10.8
4.6
4.7
1.0
0.9
26.2
26.4
37.5
36.8
89.0
89.6
59.4
59.1
31.9
31.8
23.1
22.5
24.4
24.0
29.6
28.3
60.4
61.0
18.4
18.2
21.1
20.8
{
1971
1976
1981
* Births outside marriage can be registered by both the mother and father (joint) or by the mother alone (sole).
† Usual address of parents.
‡ Provisional.
Office for National Statistics
62
Health Statistics Quar terly 01
Table 4.1
Conceptions: age of woman at conception
Numbers (thousands) and rates; and percentage terminated by abortion
Spring 1999
England and Wales (residents)
Age of woman at conception
Year and quarter
All ages
Under 16
Under 20
20–24
25–29
30–34
35–39
40 and over
(a) numbers (thousands)
1990
1991
1992
1993
1994
1995
1996
871.5
853.6
828.0
819.0
801.6
790.3
816.0
8.6
7.8
7.3
7.2
7.8
8.0
8.8
115.1
103.3
93.0
86.7
85.0
86.2
94.4
245.2
234.1
215.0
202.9
189.6
180.4
179.1
283.8
281.1
274.8
271.4
261.5
249.9
251.9
160.2
166.3
172.9
181.9
185.9
191.2
200.5
55.4
56.9
60.1
63.5
66.7
69.2
75.9
11.8
11.9
12.2
12.6
12.9
13.3
14.2
1995 March
1995 June
1995 Sept
1995 Dec
193.2
194.1
195.2
207.8
1.9
2.0
2.1
2.0
20.9
21.3
21.0
23.0
45.2
44.7
43.3
47.2
61.4
61.1
62.0
65.4
45.8
46.4
48.4
50.6
16.6
17.2
17.2
18.2
3.3
3.4
3.3
3.4
1996 March
1996 June
1996 Sept
1996 Dec
206.3
200.7
202.3
206.6
2.3
2.3
2.1
2.1
24.1
23.7
22.5
24.1
47.2
44.4
42.9
44.7
63.8
61.9
63.0
63.2
49.4
48.9
51.0
51.2
18.4
18.4
19.3
19.8
3.4
3.6
3.6
3.6
1997 March‡
1997 June‡
1997 Sept‡
194.0
198.1
199.2
2.0
2.2
2.0
23.1
23.8
23.3
41.4
41.5
40.3
59.4
59.6
60.6
47.8
49.9
51.3
18.8
19.6
20.1
3.6
3.8
3.6
(b) rates (conceptions per thousand women in age group)
1990
79.2
10.1
1991
77.7
9.3
1992
76.3
8.5
1993
76.1
8.1
1994
74.7
8.3
1995
73.7
8.5
1996
76.0
9.4
69.1
65.1
61.7
59.6
58.6
58.7
63.0
124.4
120.6
113.5
110.4
107.3
105.9
110.4
137.8
135.0
131.7
131.2
128.0
124.8
127.5
89.1
89.4
90.4
92.5
91.8
92.1
95.3
33.2
34.0
35.4
36.8
37.8
38.2
40.6
6.4
6.4
7.0
7.4
7.7
8.0
8.5
1995 March
1995 June
1995 Sept
1995 Dec
73.0
72.6
72.2
76.8
8.2
8.7
8.7
8.6
58.0
58.2
56.5
61.7
106.0
104.9
101.5
112.0
123.5
122.1
123.1
130.2
90.3
89.9
92.5
96.2
37.6
38.2
37.5
39.5
8.0
8.1
7.8
8.0
1996 March
1996 June
1996 Sept
1996 Dec
77.3
75.2
74.9
76.5
9.9
9.8
9.1
9.0
65.2
63.7
59.6
63.3
114.9
109.3
105.8
111.4
129.4
125.8
127.3
128.6
94.9
93.7
96.4
96.9
40.1
39.7
41.0
41.7
8.1
8.6
8.6
8.5
1997 March‡
1997 June‡
1997 Sept‡
73.2
73.9
73.5
8.7
9.4
8.7
61.6
62.3
60.5
106.6
106.8
104.4
123.5
123.2
125.6
91.9
94.8
97.4
39.9
41.0
41.6
8.6
8.9
8.5
(c) percentage terminated by abortion
1990
19.9
1991
19.3
1992
19.3
1993
19.2
1994
19.5
1995
19.7
1996
20.8
50.6
51.0
51.0
52.1
52.8
49.8
51.5
35.6
34.4
34.5
34.9
35.3
35.2
36.8
22.2
22.1
22.4
22.9
23.5
24.3
25.9
13.5
13.4
13.9
13.8
14.3
14.8
15.6
13.8
13.7
13.7
13.4
13.4
13.4
14.0
23.3
22.0
21.8
21.2
20.8
20.3
20.9
43.3
41.8
40.6
39.4
40.0
37.2
36.7
1995 March
1995 June
1995 Sept
1995 Dec
19.8
20.2
19.3
19.6
48.7
48.8
52.1
49.3
35.3
35.4
35.1
34.8
23.8
24.7
24.1
24.5
14.9
15.4
14.4
14.5
13.8
13.8
12.9
13.4
20.6
20.8
19.6
20.3
38.0
37.4
38.1
35.3
1996 March
1996 June
1996 Sept
1996 Dec
21.0
21.3
19.8
21.1
49.4
51.6
52.9
52.5
36.7
37.2
35.8
37.7
25.6
26.6
24.8
26.4
15.8
16.0
14.9
15.9
14.3
14.3
13.2
14.2
21.5
21.6
20.0
20.7
36.1
37.2
36.4
37.1
1997 March‡
1997 June‡
1997 Sept‡
21.4
21.7
20.6
50.8
51.9
50.6
36.6
37.2
37.4
26.7
27.2
26.1
16.6
16.7
15.8
14.4
14.5
13.5
20.6
21.6
20.4
37.7
38.1
35.5
‡ Provisional
Notes: 1. Conceptions are estimates derived from birth registrations and abortion notifications.
2. Rates for women of all ages, under 20 and 40 and over are based on the population of women aged 15–44, 15–19 and 40–44 respectively. Some figures for September 1996
onwards have been amended.
These rates use mid-1997 population estimates.
63
Office f or National Statistics
Health Statistics Quar terly 01
Table 4.2
Spring 1999
Abortions: residents and non-residents; age and gestation (residents only)
Numbers (thousands) and rates; and percentages for gestation weeks
England
England and
andWales
Wales
All women (residents)
All ages
Year and quarter
All
women
Residents
Age group
Nonresidents
Under
16
16–19
20–34
Gestation weeks
35–44
45 and
over
Not
stated
Numbers (thousands)
Under
10
10–12
13–19
20 and
over
Not
stated
Percentages
1971
1976
1981
1986
1991
126.8
129.7
162.5
172.3
179.5
94.6
101.9
128.6
147.6
167.4
32.2
27.8
33.9
24.7
12.1
2.3
3.4
3.5
3.9
3.2
18.2
24.0
31.4
33.8
31.1
56.0
57.5
74.9
92.0
114.7
15.9
14.7
17.6
17.5
17.9
0.5
0.5
0.6
0.4
0.4
1.8
1.8
0.6
0.0
0.0
30.5
41.6
45.4
47.7
50.7
44.0
38.9
39.0
39.4
37.4
21.8
15.0
13.5
11.5
10.6
1.0
1.1
1.3
1.4
1.2
2.8
3.4
0.8
0.0
0.0
1992
1993
1994
1995
1996
1997
172.1
168.7
166.9
163.6
177.5
179.7
160.5
157.8
156.5
154.3
167.9
170.1
11.6
10.9
10.3
9.3
9.6
9.6
3.0
3.1
3.2
3.2
3.6
3.4
27.6
25.8
25.1
24.9
28.8
29.9
113.5
109.7
108.1
106.4
113.9
114.0
18.1
18.8
19.1
19.2
21.1
22.3
0.5
0.5
0.4
0.5
0.4
0.5
0.0
0.0
0.0
0.0
0.0
0.0
51.3
52.9
54.2
55.6
54.1
56.2
37.3
36.0
34.7
33.6
34.5
32.9
10.3
9.9
9.9
9.6
10.1
9.6
1.2
1.2
1.2
1.2
1.3
1.2
0.0
0.0
0.0
0.0
0.0
0.0
1995
March
June
Sept
Dec
42.9
40.0
41.3
39.5
40.4
37.7
38.9
37.3
2.4
2.3
2.4
2.2
0.9
0.8
0.8
0.8
6.5
6.0
6.4
6.1
28.0
26.1
26.7
25.6
5.0
4.7
4.9
4.6
0.1
0.1
0.1
0.1
0.0
0.0
0.0
0.0
52.9
55.6
55.8
58.6
35.7
33.8
33.2
31.4
10.4
9.4
9.7
8.7
1.1
1.2
1.3
1.2
0.0
0.0
0.0
0.0
1996
March
June
Sept
Dec
45.7
45.5
44.0
42.4
43.2
42.9
41.6
40.1
2.4
2.5
2.4
2.2
0.9
0.9
0.9
0.9
7.4
7.3
7.1
7.0
29.7
29.2
28.0
26.9
5.2
5.4
5.4
5.2
0.1
0.1
0.1
0.1
0.0
0.0
0.0
0.0
52.0
52.8
54.3
57.5
36.5
35.3
34.1
32.0
10.2
10.5
10.3
9.3
1.2
1.4
1.4
1.3
0.0
0.0
0.0
0.0
1997
March
June
Sept
Dec
46.2
45.2
45.1
43.3
43.6
42.8
42.7
41.0
2.5
2.4
2.4
2.3
0.9
0.8
0.9
0.8
7.7
7.4
7.5
7.4
29.4
28.9
28.4
27.3
5.5
5.6
5.8
5.4
0.1
0.1
0.1
0.1
0.0
0.0
0.0
0.0
51.7
56.4
57.2
59.8
36.0
32.9
32.0
30.7
11.1
9.4
9.6
8.3
1.3
1.2
1.2
1.2
0.0
0.0
0.0
0.0
1998
March‡
June‡
Sept‡
47.9
46.1
46.6
45.4
43.7
44.2
2.5
2.4
2.4
1.0
0.9
1.0
8.6
8.1
8.2
30.0
28.9
28.8
5.8
5.8
6.1
0.1
0.1
0.1
0.0
0.0
0.0
52.7
55.7
57.0
35.2
33.4
32.3
10.8
9.5
9.5
1.3
1.4
1.2
0.0
0.0
0.0
Rates (per thousand women 14–49)
1971
:
1976
:
1981
:
1986
:
1991
:
8.4
8.9
10.6
11.7
13.1
:
:
:
:
:
3.5
4.4
4.5
5.4
5.6
13.9
16.9
19.4
22.0
24.0
11.4
11.2
14.0
16.5
19.6
5.6
5.3
5.9
5.1
5.1
0.3
0.3
0.4
0.3
0.3
1992
1993
1994
1995
1996
1997
:
:
:
:
:
:
12.5
12.3
12.2
12.0
13.0
13.3
:
:
:
:
:
:
5.4
5.3
5.2
5.2
5.8
5.5
22.4
22.0
22.0
21.7
24.3
24.5
18.9
18.7
18.6
18.4
20.0
20.4
5.2
5.5
5.6
5.5
6.0
6.1
0.3
0.3
0.2
0.2
0.2
0.3
1995
March
June
Sept
Dec
:
:
:
:
12.8
11.8
12.0
11.5
:
:
:
:
5.6
4.9
5.1
5.2
22.9
21.0
21.9
21.0
19.6
18.1
18.3
17.6
5.8
5.4
5.6
5.2
0.3
0.2
0.3
0.2
1996
March
June
Sept
Dec
:
:
:
:
13.5
13.4
12.8
12.4
:
:
:
:
5.7
5.9
5.9
5.8
25.0
24.9
24.0
23.5
21.0
20.6
19.6
18.8
5.9
6.1
6.1
5.8
0.2
0.2
0.2
0.3
1997
March
June
Sept
Dec
:
:
:
:
13.8
13.4
13.2
12.7
:
:
:
:
5.7
5.4
5.7
5.3
25.5
24.1
24.3
24.0
21.3
20.7
20.2
19.4
6.2
6.2
6.3
5.9
0.2
0.3
0.3
0.3
1998
March‡
June‡
Sept‡
:
:
:
14.4
13.7
13.7
:
:
:
6.3
5.7
6.1
28.5
26.4
26.7
21.7
20.7
20.4
6.5
6.4
6.7
0.3
0.3
0.3
‡ Provisional.
Office for National Statistics
64
Health Statistics Quar terly 01
Table 5.1
Expectation of life (in years) at birth and selected age
Constituent countries of the United Kingdom
Males
Year
At
birth
Spring 1999
Females
At age
Year
5
20
30
50
60
70
80
At
birth
At age
5
20
30
50
60
70
80
United Kingdom
1971
1976
68.8
69.6
65.3
66.0
50.9
51.4
41.3
41.9
23.0
23.4
15.3
15.7
9.5
9.6
5.5
5.6
1971
1976
75.0
75.2
71.4
72.0
56.7
57.3
47.0
47.5
28.3
28.7
19.8
20.3
12.5
12.9
6.9
7.2
1981
1986
70.8
71.9
66.9
67.8
52.3
53.1
42.7
43.5
24.1
24.9
16.3
16.8
10.1
10.5
5.7
6.0
1981
1986
76.8
77.7
72.8
73.4
57.9
58.6
48.2
48.8
29.2
29.8
20.8
21.2
13.3
13.7
7.4
7.8
1991
1992
1993
1994
1995
1996‡
73.2
73.4
73.7
73.9
74.1
74.2
68.9
69.1
69.3
69.5
69.7
69.8
54.2
54.4
54.6
54.8
55.0
55.1
44.7
44.8
45.1
45.2
45.5
45.5
26.0
26.1
26.4
26.5
26.8
26.8
17.7
17.8
18.0
18.1
18.4
18.4
11.1
11.1
11.3
11.3
11.5
11.5
6.4
6.4
6.5
6.5
6.6
6.6
1991
1992
1993
1994
1995
1996‡
78.8
78.9
79.1
79.2
79.4
79.5
74.4
74.4
74.6
74.7
74.9
75.0
59.6
59.6
59.8
59.9
60.1
60.1
49.7
49.8
50.0
50.0
50.2
50.3
30.7
30.7
30.9
31.0
31.2
31.2
21.9
21.9
22.1
22.2
22.4
22.4
14.4
14.4
14.5
14.5
14.6
14.6
8.3
8.3
8.4
8.4
8.5
8.5
England and Wales
1971
69.0
1976
69.9
65.6
66.2
51.1
51.6
41.5
42.1
23.1
23.5
15.4
15.8
9.5
9.7
5.5
5.7
1971
1976
75.2
76.0
71.6
72.2
56.9
57.4
47.1
47.7
28.4
28.8
20.0
20.4
12.6
13.0
7.0
7.2
1981
1986
71.0
72.1
67.1
68.0
52.5
53.4
42.9
43.8
24.3
25.0
16.4
16.9
10.1
10.5
5.8
6.0
1981
1986
77.0
77.8
72.9
73.6
58.1
58.8
48.3
49.0
29.4
29.9
20.9
21.3
13.4
13.8
7.5
7.8
1991
1992
1993
1994
1995
1996‡
73.4
73.6
74.0
74.1
74.4
74.5
69.1
69.3
69.6
69.7
70.0
70.1
54.5
54.6
54.9
55.0
55.2
55.4
44.9
45.0
45.3
45.4
45.7
45.9
26.2
26.3
26.5
26.7
26.9
27.1
17.9
17.9
18.2
18.3
18.5
18.7
11.2
11.2
11.4
11.4
11.6
11.7
6.4
6.4
6.5
6.5
6.6
6.7
1991
1992
1993
1994
1995
1996‡
79.0
79.1
79.3
79.4
79.6
79.6
74.6
74.6
74.8
74.9
75.1
75.1
59.8
59.8
60.0
60.1
60.3
60.3
49.9
50.0
50.2
50.3
50.4
50.5
30.8
30.9
31.1
31.2
31.3
31.4
22.1
22.1
22.3
22.3
22.5
22.5
14.5
14.5
14.6
14.6
14.7
14.7
8.4
8.4
8.5
8.5
8.6
8.5
England
1981
1986
71.1
72.2
67.1
68.1
52.5
53.4
42.9
43.8
24.3
25.1
16.4
17.0
10.1
10.6
5.8
6.1
1981
1986
77.0
77.9
72.9
73.7
58.2
58.9
48.4
49.1
29.4
30.0
20.9
21.4
13.4
13.9
7.5
7.9
1991
1992
1993
1994
1995
1996‡
73.4
73.7
74.0
74.1
74.4
74.6
69.1
69.3
69.6
69.7
70.0
70.2
54.5
54.6
54.9
55.0
55.3
55.5
44.9
45.0
45.3
45.5
45.7
45.9
26.2
26.3
26.6
26.7
27.0
27.1
17.9
18.0
18.2
18.3
18.5
18.7
11.2
11.2
11.4
11.4
11.6
11.7
6.4
6.4
6.5
6.6
6.6
6.7
1991
1992
1993
1994
1995
1996‡
79.0
79.1
79.3
79.4
79.6
79.7
74.6
74.6
74.9
74.9
75.1
75.2
59.8
59.8
60.0
60.1
60.3
60.3
49.9
50.0
50.2
50.3
50.5
50.5
30.9
30.9
31.1
31.2
31.4
31.4
22.1
22.1
22.3
22.4
22.5
22.6
14.5
14.5
14.6
14.6
14.7
14.7
8.4
8.4
8.5
8.5
8.6
8.6
Wales
1981
1986
70.4
71.6
66.5
67.5
51.9
52.9
42.2
43.3
23.6
24.6
15.8
16.6
9.7
10.4
5.5
6.0
1981
1986
76.4
77.6
72.3
73.3
57.5
58.5
47.7
48.7
28.9
29.7
20.4
21.1
13.1
13.8
7.4
7.8
1991
1992
1993
1994
1995
1996‡
73.2
73.3
73.5
73.5
73.8
73.9
68.9
68.9
69.1
69.1
69.4
69.5
54.2
54.2
54.4
54.4
54.7
54.8
44.6
44.7
44.9
44.9
45.2
45.3
25.9
25.9
26.1
26.2
26.5
26.6
17.6
17.7
17.8
17.9
18.1
18.3
11.0
11.0
11.2
11.1
11.3
11.4
6.4
6.4
6.6
6.5
6.6
6.5
1991
1992
1993
1994
1995
1996‡
78.9
78.9
79.0
79.0
79.2
79.2
74.4
74.4
74.5
74.5
74.7
74.7
59.6
59.6
59.7
59.7
59.8
59.8
49.8
49.8
49.9
49.8
50.0
50.0
30.7
30.7
30.8
30.8
30.9
31.0
21.9
21.9
22.0
22.0
22.2
22.2
14.4
14.3
14.4
14.4
14.5
14.5
8.4
8.3
8.4
8.4
8.5
8.5
Scotland
1971
1976
67.3
68.2
64.0
64.4
49.5
49.9
40.1
40.4
22.0
22.3
14.6
14.9
9.1
9.2
5.4
5.3
1971
1976
73.7
74.4
70.1
70.6
55.4
55.9
45.6
46.1
27.2
27.6
19.0
19.4
11.9
12.4
6.7
6.9
1981
1986
69.1
70.2
65.0
66.0
50.0
51.4
41.0
41.9
22.8
23.5
15.3
15.8
9.5
9.9
5.4
5.7
1981
1986
75.3
76.2
71.1
71.9
56.3
57.1
46.5
47.3
27.8
28.4
19.6
20.0
12.6
13.0
7.2
7.4
1991
1992
1993
1994
1995
1996‡
71.4
71.5
71.7
71.9
72.1
72.2
67.1
67.2
67.3
67.5
67.7
67.8
52.5
52.5
52.7
52.8
53.1
53.1
43.0
43.1
43.2
43.4
43.6
43.7
24.6
24.6
24.8
24.9
25.2
25.3
16.6
16.6
16.8
16.9
17.2
17.3
10.4
10.4
10.5
10.6
10.8
10.9
6.1
6.0
6.0
6.1
6.2
6.2
1991
1992
1993
1994
1995
1996‡
77.1
77.1
77.3
77.4
77.6
77.8
72.6
72.6
72.8
72.9
73.2
73.2
57.8
57.8
58.0
58.1
58.3
58.4
48.1
48.1
48.2
48.3
48.6
48.7
29.1
29.1
29.3
29.4
29.6
29.7
20.6
20.6
20.7
20.8
21.0
21.1
13.4
13.4
13.4
13.5
13.7
13.7
7.8
7.7
7.8
7.8
7.9
7.9
Northern Ireland
1971
1976
67.6
67.5
64.6
64.1
50.1
49.7
40.7
40.5
22.6
22.5
15.0
14.9
9.4
9.2
5.3
4.8
1971
1976
73.7
73.8
70.4
70.4
55.6
55.7
45.9
46.0
27.3
27.3
18.9
19.0
11.7
11.8
6.5
6.1
1981
1986
69.1
70.6
65.5
66.4
50.9
51.8
41.6
42.4
23.2
23.8
15.6
16.0
9.6
10.0
5.4
5.7
1981
1986
75.4
76.7
71.5
72.5
56.8
57.7
47.1
47.9
28.2
28.9
19.9
20.4
12.6
13.0
7.0
7.3
1991
1992
1993
1994
1995
1996‡
72.3
72.5
72.8
72.9
73.3
73.6
67.9
68.2
68.4
68.6
68.9
69.2
53.3
53.6
53.8
54.0
54.3
54.5
43.8
44.1
44.4
44.5
44.8
45.0
25.2
25.5
25.6
25.8
26.1
26.3
17.0
17.2
17.4
17.5
17.7
17.9
10.6
10.7
10.8
10.9
11.0
11.0
6.0
6.0
6.1
6.1
6.0
6.0
1991
1992
1993
1994
1995
1996‡
78.1
78.3
78.4
78.4
78.7
78.9
73.7
73.9
74.0
74.0
74.2
79.5
58.9
59.1
59.2
59.2
59.4
59.6
49.1
49.3
49.4
49.4
49.6
49.8
30.0
30.2
30.3
30.3
30.5
30.7
21.3
21.5
21.6
21.6
21.7
21.9
13.8
13.9
14.0
14.0
14.0
14.1
7.8
8.0
7.9
7.9
7.9
8.0
Note: Figures from 1981 are calculated from the population estimates revised in the light of the 1991 Census. All figures are based on a three-year period; see Notes to tables for further
information.
‡
Provisional.
65
Office f or National Statistics
Health Statistics Quar terly 01
Table 6.1
Spring 1999
Deaths: age and sex
Numbers (thousands) and rates
England and Wales
Age group
Year and quarter
All ages
Under 1*
1–4
5–9
10–14
15–19
20–24
25–34
35–44
45–54
55–64
65–74
75–84
85 and over
Numbers (thousands)
Males
1971
1976
1981
1986
1991
1992
288.4
300.1
289.0
287.9
277.6
271.7
7.97
4.88
4.12
3.72
2.97
2.61
1.23
0.88
0.65
0.57
0.55
0.49
0.92
0.68
0.45
0.32
0.34
0.30
0.69
0.64
0.57
0.38
0.35
0.32
1.54
1.66
1.73
1.43
1.21
0.97
1.77
1.66
1.58
1.75
1.76
1.62
3.05
3.24
3.18
3.10
3.69
3.75
6.68
5.93
5.54
5.77
6.16
5.95
21.0
20.4
16.9
14.4
13.3
13.1
55.7
52.0
46.9
43.6
34.9
33.7
89.8
98.7
92.2
84.4
77.2
76.4
71.9
80.3
86.8
96.2
95.8
92.7
26.1
29.0
28.5
32.2
39.3
39.9
1993
1994
1995
1996
1997
279.6
267.6
274.4
268.7
264.9
2.41
2.37
2.31
2.27
2.14
0.51
0.43
0.39
0.44
0.41
0.28
0.28
0.27
0.24
0.27
0.34
0.33
0.34
0.29
0.33
0.91
0.84
0.91
0.93
0.95
1.60
1.55
1.53
1.41
1.44
3.81
4.07
4.04
4.06
3.94
5.78
5.77
5.88
5.84
5.71
13.4
12.9
13.5
13.6
13.5
33.3
31.3
31.0
30.1
28.9
78.9
76.3
75.0
71.0
68.0
93.8
88.2
92.3
90.7
90.2
44.5
43.2
47.1
47.8
49.1
Females
1971
1976
1981
1986
1991
1992
278.9
298.5
288.9
293.3
292.5
286.6
5.75
3.46
2.90
2.59
2.19
1.93
0.98
0.59
0.53
0.49
0.44
0.39
0.57
0.45
0.30
0.25
0.25
0.21
0.42
0.42
0.37
0.27
0.22
0.20
0.63
0.62
0.65
0.56
0.46
0.43
0.79
0.67
0.64
0.67
0.64
0.62
1.84
1.94
1.82
1.65
1.73
1.72
4.53
4.04
3.74
3.83
3.70
3.72
13.3
12.8
10.5
8.8
8.4
8.3
30.8
29.6
27.2
25.8
21.3
20.6
64.0
67.1
62.8
58.4
54.2
53.4
95.0
104.7
103.6
106.5
103.3
99.5
60.4
72.1
73.9
83.6
95.7
95.5
1993
1994
1995
1996
1997
299.2
285.6
295.2
291.5
290.4
1.84
1.75
1.68
1.69
1.66
0.37
0.36
0.33
0.32
0.30
0.19
0.19
0.20
0.18
0.18
0.25
0.20
0.21
0.20
0.21
0.39
0.36
0.38
0.43
0.43
0.58
0.54
0.50
0.51
0.49
1.80
1.77
1.86
1.85
1.72
3.63
3.67
3.64
3.66
3.74
8.6
8.7
9.0
8.9
9.0
20.4
19.0
18.9
18.2
18.0
55.2
53.9
53.0
50.2
48.3
100.9
94.2
97.2
96.7
95.5
105.0
101.0
108.4
108.7
110.9
Rates (deaths per 1,000 population in each age group)
Males
1971
1976
1981
1986
1991
1992
12.1
12.5
12.0
11.8
11.2
10.8
19.8
16.2
12.6
11.0
8.3
7.3
0.76
0.65
0.53
0.44
0.40
0.34
0.44
0.34
0.27
0.21
0.21
0.18
0.37
0.31
0.29
0.23
0.23
0.20
0.90
0.88
0.82
0.71
0.69
0.61
0.93
0.96
0.83
0.82
0.86
0.82
0.97
0.92
0.89
0.87
0.94
0.91
2.31
2.09
1.83
1.67
1.76
1.71
7.07
6.97
6.11
5.27
4.62
4.29
20.1
19.6
17.7
16.6
13.8
13.4
50.5
50.3
45.6
42.9
38.5
37.3
113.0
116.4
105.2
101.1
93.6
90.1
231.8
243.2
226.5
214.8
197.1
193.9
1993
1994
1995
1996
1997
11.1
10.6
10.8
10.5
10.3
7.0
6.9
6.9
7.0
6.5
0.36
0.31
0.28
0.32
0.31
0.16
0.16
0.15
0.13
0.15
0.21
0.20
0.21
0.18
0.19
0.59
0.55
0.58
0.58
0.58
0.83
0.83
0.86
0.83
0.89
0.91
0.96
0.95
095
0.93
1.67
1.66
1.67
1.62
1.54
4.24
3.99
4.08
4.02
3.94
13.3
12.4
12.3
12.0
11.5
37.9
36.2
36.1
34.5
33.2
93.3
89.5
89.4
85.1
82.5
202.3
188.6
196.0
192.1
190.3
1996 Dec
10.9
6.8
0.37
0.15
0.14
0.61
0.82
0.92
1.65
4.21
12.2
35.6
89.5
202.6
1997 March
1997 June
1997 Sept
1997 Dec
11.9
9.7
9.3
10.4
7.0
6.6
5.9
6.6
0.32
0.32
0.24
0.34
0.15
0.17
0.16
0.12
0.21
0.16
0.20
0.19
0.60
0.55
0.57
0.62
0.95
0.85
0.89
0.86
0.96
0.89
0.93
0.94
1.55
1.52
1.48
1.61
4.11
3.85
3.76
4.04
12.3
11.0
10.9
11.7
37.1
32.2
30.5
33.2
98.4
76.3
72.4
83.1
235.2
170.6
161.8
194.4
1998 March‡
1998 June‡
1998 Sept‡
11.0
10.1
9.3
6.5
6.0
5.7
0.38
0.29
0.27
0.17
0.14
0.13
0.19
0.21
0.15
0.60
0.65
0.50
0.93
0.87
0.84
1.02
0.93
0.88
1.64
1.57
1.51
4.13
3.90
3.87
12.1
11.3
10.7
34.7
31.6
29.4
88.0
80.8
74.7
210.9
186.5
169.6
Females
1971
1976
1981
1986
1991
1992
11.0
11.8
11.3
11.4
11.3
10.9
15.1
12.2
9.4
8.0
6.4
5.7
0.63
0.46
0.46
0.40
0.33
0.29
0.29
0.24
0.19
0.17
0.16
0.14
0.24
0.21
0.19
0.17
0.15
0.13
0.39
0.35
0.32
0.29
0.28
0.29
0.42
0.40
0.35
0.33
0.33
0.32
0.60
0.56
0.52
0.47
0.45
0.43
1.59
1.46
1.26
1.12
1.06
1.08
4.32
4.30
3.80
3.23
2.91
2.73
10.0
10.1
9.5
9.2
8.1
7.9
26.1
26.0
24.1
23.4
22.0
21.5
73.6
74.6
66.2
62.5
58.6
56.9
185.7
196.6
178.2
171.0
163.8
148.8
1993
1994
1995
1996
1997
11.4
10.9
11.2
11.0
10.9
5.6
5.4
5.3
5.4
5.3
0.28
0.27
0.25
0.24
0.23
0.12
0.11
0.12
0.10
0.10
0.16
0.13
0.13
0.12
0.13
0.27
0.25
0.26
0.29
0.28
0.31
0.30
0.29
0.31
0.32
0.45
0.44
0.46
0.45
0.42
1.06
1.06
1.05
1.03
1.03
2.73
2.68
2.72
2.62
2.63
7.9
7.3
7.3
7.1
6.9
22.0
21.3
21.4
20.7
20.2
59.4
56.9
57.1
55.8
54.6
156.5
146.6
153.1
150.8
151.8
1996 Dec
11.6
5.1
0.24
0.11
0.15
0.28
0.34
0.43
1.02
2.74
7.4
21.3
58.7
159.6
1997 March
1997 June
1997 Sept
1997 Dec
13.3
10.0
9.6
10.9
5.6
5.3
4.8
5.6
0.29
0.24
0.18
0.22
0.12
0.10
0.08
0.12
0.10
0.13
0.15
0.15
0.28
0.27
0.26
0.31
0.28
0.35
0.33
0.30
0.43
0.40
0.43
0.44
1.12
1.00
1.01
1.00
2.76
2.57
2.57
2.63
7.8
6.5
6.6
6.9
23.1
19.1
18.3
20.3
65.8
50.6
48.0
54.1
195.0
134.3
127.2
151.3
1998 March‡
1998 June‡
1998 Sept‡
11.8
10.5
9.7
5.4
4.4
4.6
0.25
0.21
0.18
0.12
0.11
0.08
0.18
0.09
0.10
0.30
0.29
0.28
0.31
0.29
0.29
0.41
0.45
0.45
1.03
1.00
0.98
2.64
2.68
2.51
7.1
6.8
6.4
21.3
18.9
17.8
58.5
52.4
48.0
169.3
146.6
134.8
* Rates per 1,000 live births.
‡ Provisional registrations.
Note: Figures represent the numbers of deaths registered in each year up to 1992 and the numbers of deaths occurring in each year from 1993.
Office for National Statistics
66
Health Statistics Quar terly 01
Table 6.2
Year and
quarter
Deaths: subnational
Rates
Northern and
Yorkshire
Spring 1999
Health Regional Office areas of England*
Trent
Anglia and
Oxford
North
Thames
South
Thames
South and
West
West
Midlands
North
West
Total deaths (deaths per 1,000 population of all ages)
1991
11.8
11.2
1992
11.4
11.0
9.7
9.5
10.0
9.6
11.3
11.1
11.5
11.3
10.8
10.6
12.0
11.7
1993
1994
1995
1996
1997
11.8
11.2
11.3
11.2
11.0
11.4
10.8
11.0
10.9
10.8
9.8
9.4
9.6
9.6
9.4
9.9
9.5
9.7
9.4
9.2
11.5
10.9
11.1
10.9
10.7
11.6
11.1
11.5
11.2
11.2
11.0
10.5
10.9
10.6
10.5
12.1
11.5
11.6
11.5
11.4
1997 March
1997 June
1997 Sept
1997 Dec
12.9
10.4
9.8
11.1
12.7
10.1
9.6
11.1
11.2
8.7
8.3
9.4
10.9
8.5
8.3
9.1
12.9
9.7
9.4
10.6
13.6
10.4
9.8
11.2
12.3
9.8
9.4
10.6
13.1
10.6
10.2
11.6
1998 March‡
1998 June‡
1998 Sept‡
12.4
10.9
10.0
11.7
10.7
9.5
10.0
9.2
8.7
9.8
8.6
8.2
11.3
10.1
9.4
11.9
11.0
10.0
11.2
10.1
9.5
12.1
11.1
10.3
Infant mortality (deaths under 1 year per 1,000 live births)
1991
8.5
8.0
1992
6.9
6.8
6.8
5.4
6.5
6.4
6.5
6.0
6.4
5.6
8.7
8.2
7.5
7.3
1993
1994
1995
1996
1997
6.9
6.8
6.6
6.4
6.3
7.0
7.2
6.4
6.3
6.0
5.3
5.6
5.2
5.8
5.0
6.2
6.1
5.7
5.6
5.4
6.4
5.2
5.8
6.1
5.3
5.6
5.0
5.6
5.5
5.8
7.0
7.2
7.1
6.8
7.1
6.5
6.2
6.6
6.4
6.8
1997 March
1997 June
1997 Sept
1997 Dec
6.3
6.2
6.5
6.0
6.6
6.4
5.8
5.2
4.5
5.3
4.4
6.0
5.3
5.1
5.0
5.9
5.8
5.1
5.8
4.5
6.5
5.8
5.1
6.0
7.5
7.2
5.8
8.0
6.7
6.8
5.9
7.8
1998 March‡
1998 June‡
6.5
6.4
6.0
5.6
5.1
4.4
5.2
5.4
5.8
4.4
5.1
3.8
7.2
6.3
6.8
5.2
Neonatal mortality (deaths under 4 weeks per 1,000 live births)
1991
4.9
4.7
3.8
1992
4.5
4.5
3.5
4.2
4.2
3.8
3.9
3.6
3.4
5.9
5.9
4.0
4.4
1993
1994
1995
1996
1997
4.3
4.4
4.5
4.1
4.1
4.7
5.1
4.5
4.2
4.0
3.6
3.8
3.4
3.7
3.3
4.4
3.9
3.9
3.9
3.5
4.2
3.7
3.9
4.1
3.6
3.6
3.1
3.9
3.9
3.9
4.8
5.4
5.3
5.0
5.0
4.0
3.9
4.2
4.1
4.3
1997 March
1997 June
1997 Sept
1997 Dec
4.3
3.9
4.5
3.8
4.2
4.3
4.2
3.5
3.1
3.8
2.6
3.8
3.7
3.3
3.5
3.4
3.7
3.5
4.1
3.2
3.9
3.7
3.6
4.3
4.5
5.6
4.3
5.6
3.8
4.0
3.9
5.5
1998 March‡
1998 June‡
3.8
4.0
4.2
4.2
2.9
3.6
3.2
3.6
3.9
3.1
3.3
2.5
5.2
4.9
4.8
3.2
Perinatal mortality (stillbirths and deaths under 1 week per 1,000 total births)†
1991
8.7
8.6
7.2
1992
7.3
8.5
6.2
8.0
7.5
7.4
7.2
7.0
6.7
9.9
9.2
7.8
8.1
1993
1994
1995
1996
1997
9.4
9.1
9.4
8.6
8.2
8.6
9.1
9.5
8.7
7.9
8.5
7.9
7.2
7.7
7.5
9.2
9.1
9.0
9.0
8.5
8.9
8.1
8.6
8.6
7.9
7.8
7.8
7.7
7.5
8.4
9.9
10.6
10.2
10.2
9.7
8.9
9.2
8.6
8.7
8.8
1997 March
1997 June
1997 Sept
1997 Dec
8.0
8.0
8.6
8.4
8.0
7.8
8.4
7.5
8.6
7.4
6.5
7.6
9.0
8.1
8.0
8.8
8.7
7.5
7.9
7.3
8.8
8.4
7.3
9.4
9.5
9.7
8.5
11.1
8.0
9.1
9.1
9.1
1998 March‡
1998 June‡
9.5
9.0
8.9
8.8
6.5
7.4
8.5
8.2
7.8
7.5
6.5
5.9
10.6
9.5
10.3
8.3
* As constituted on 1 April 1996.
† In October 1992 the legal definition of a stillbirth was changed, from baby born dead after 28 completed weeks of gestation or more, to one born dead after 24 completed weeks of
gestation or more.
‡ Provisional registrations.
Note: Figures represent the numbers of deaths registered in each year up to 1992 and the number of deaths occurring in each year from 1993.
67
Office f or National Statistics
Health Statistics Quar terly 01
Table 6.3
Spring 1999
Deaths: selected causes (International Classification)* and sex
Number (thousands) and rate for all deaths and age-standardised rates† per million population for selected causes
England and Wales
Malignant neoplasms
Year and
quarter
All deaths
Number
(thousands)
Males
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
All causes Oesophagus
Stomach
Colon,
rectum,
rectosigmoid
junction and
anus
Trachea,
bronchus
and lung
Melanoma
of skin
Other
neoplasm
of skin
Breast
Cervix
uteri
Ovary and
other
uterine
Prostate
(150)
(151)
(153,154)
(162)
(172)
(173)
(174)
(180)
(183)
(185)
Rate**
288.4
300.1
289.0
287.9
277.6
271.7
279.6
267.6
274.4
268.7
264.9
1,207
1,246
1,196
1,177
1,121
1,083
1,109
1,057
1,079
1,051
1,031
13,464
13,613
12,200
11,349
10,234
9,870
10,010
9,502
9,582
9,271
9,019
76
84
90
101
117
120
123
128
126
126
125
317
292
251
224
185
179
162
162
148
145
136
331
339
316
313
310
316
294
283
281
272
267
1,066
1,091
1,028
949
841
810
766
743
712
681
649
10
14
17
18
23
22
25
24
26
25
25
12
12
9
9
10
10
8
9
9
8
7
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
198
211
214
263
302
303
296
295
296
287
277
1996 Dec
69.3
1,078
9,634
132
146
282
684
25
8
:
:
:
306
1997 Mar
1997 Jun
1997 Sep
1997 Dec
75.2
62.1
60.1
67.5
1,187
970
928
1,042
10,330
8,501
8,142
9,126
119
125
132
125
139
128
138
141
274
262
267
267
668
634
630
663
22
27
27
23
7
5
8
7
:
:
:
:
:
:
:
:
:
:
:
:
275
269
277
285
1998 Mar‡
1998 Jun‡
1998 Sep‡
69.7
64.4
60.5
1,100
1,006
934
9,625
8,712
8,195
130
126
134
133
127
135
276
257
262
667
621
639
26
26
25
8
8
7
:
:
:
:
:
:
:
:
:
277
275
274
Females
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
278.9
298.5
288.9
293.3
292.5
286.6
299.2
285.6
295.2
291.5
290.4
1,104
1,176
1,134
1,141
1,127
1,095
1,140
1,085
1,119
1,102
1,095
8,186
8,303
7,433
6,947
6,399
6,197
6,347
6,039
6,128
5,995
5,925
40
43
42
47
49
49
51
50
52
51
51
149
136
111
89
74
73
66
66
61
55
57
255
262
231
220
207
206
190
187
179
174
169
183
219
252
285
300
297
294
298
294
292
285
14
16
16
19
18
17
22
22
20
20
20
6
6
5
4
4
5
3
4
4
3
3
379
393
405
420
401
395
376
370
359
343
336
83
78
69
69
54
52
47
42
42
41
37
127
125
122
121
118
118
116
114
116
122
115
:
:
:
:
:
:
:
:
:
:
:
1996 Dec
75.5
1,136
6,259
55
56
172
300
20
4
350
38
130
:
1997 Mar
1997 Jun
1997 Sep
1997 Dec
86.7
66.5
64.3
73.0
1,326
1,005
961
1,091
6,992
5,499
5,309
5,921
50
49
54
51
56
55
58
58
168
170
169
170
292
268
293
286
21
19
20
20
4
3
3
3
336
327
340
338
33
41
38
37
116
112
116
118
:
:
:
:
1998 Mar‡
1998 Jun‡
1998 Sep‡
77.3
69.6
65.1
1,181
1,053
974
6,321
5,650
5,323
45
48
49
50
57
58
158
160
165
295
273
287
21
18
23
4
3
3
331
312
329
37
35
34
120
113
120
:
:
:
* The Ninth Revision of the International Classification of Diseases, 1975, came into operation in England and Wales on 1 January 1979. ONS has produced a publication containing details of
the effect of this Revision (Mortality statistics: comparison of the 8th and 9th revision of the International Classification of Diseases, 1978 (sample), (Series DH1 no.10).
‡ Provisional registrations.
† Directly age-standardised to the European Standard population. See Notes to Tables.
** Per 100,000 population.
Notes 1. Between 1 January 1984 and 31 December 1992, ONS applied the International Classification of Diseases Selection Rule 3 in the coding of deaths where terminal events and other
‘modes of dying’ such as cardiac arrest, cardiac failure, certain thromboembolic disorders, and unspecified pneumonia and bronchopneumonia, were stated by the certifier to be the
underlying cause of death and other major pathology appeared on the certificate. In these cases Rule 3 allows the terminal event to be considered a direct sequel to the major
pathology and that primary condition was selected as the underlying cause of death. Prior to 1984 and from 1993 onwards, such certificates are coded to the terminal event. ONS
also introduced automated coding of cause of death in 1993, which may also affect comparisons of deaths by cause from 1993. Further details may be found in the annual volumes
Mortality statistics: Cause 1984, Series DH2 no.11, and Mortality statistics: Cause 1993 (revised) and 1994, Series DH2 no.21.
2. On 1 January 1986 a new certificate for deaths within the first 28 days of life was introduced. It is not possible to assign one underlying cause of death from this certificate. The
‘cause’ figures for 1986 onwards therefore exclude deaths at ages under 28 days.
3. Figures represent the numbers of deaths registered in each year up to 1992, and the number of deaths occurring in each year from 1993. Provisional figures are registrations.
Office for National Statistics
68
Health Statistics Quar terly 01
Table 6.3
continued
Spring 1999
Deaths: selected causes (International Classification)* and sex
Number (thousands) and rate for all deaths and age-standardised rates† per million population for selected causes
England and Wales
Malignant neoplasms
Bladder
Leukaemia
Diabetes
mellitus
Ischaemic
heart
disease
Cerebrovascular
disease
Pneumonia
Bronchitis,
emphysema
and allied
conditions
Asthma
Gastric,
duodenal
and peptic
ulcers
Chronic
liver
disease
and
cirrhosis
Chronic
renal
failure
Motor
vehicle
traffic
accidents
Suicides and
undetermined
deaths
Year and
quarter
(188)
(204-208)
(250)
(410-414)
(430-438)
(480-486)
(490-492,
496)
(493)
(531-533)
(571)
(585)
124
128
121
120
121
123
114
109
111
104
100
74
76
74
75
76
72
69
68
70
65
66
82
91
82
134
130
127
100
97
100
96
94
3,801
3,930
3,664
3,463
2,981
2,854
2,829
2,595
2,535
2,410
2,261
1,541
1,357
1,141
1,071
939
886
794
755
754
743
714
920
1,237
1,054
460
390
361
759
679
753
725
741
944
852
683
725
605
573
566
494
524
480
475
21
17
28
33
31
27
24
23
20
19
19
107
108
90
85
73
69
67
67
63
63
61
35
45
49
56
70
70
67
67
75
88
95
48
61
44
38
24
14
21
20
21
19
17
198
170
113
130
117
109
90
86
83
87
86
124
135
151
154
158
157
149
148
146
137
140
Males
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
104
66
99
2,513
759
800
502
21
69
97
18
90
131
1996 Dec
102
98
101
100
66
64
68
65
110
85
87
93
2,588
2,189
1,976
2,296
825
670
637
726
1,119
603
531
717
699
398
347
458
22
16
20
19
68
57
53
65
93
89
89
108
19
15
17
19
79
91
90
84
138
134
145
145
1997 Mar
1997 Jun
1997 Sep
1997 Dec
94
101
100
64
65
67
99
87
85
2,419
2,167
1,977
758
685
636
843
689
552
560
424
362
17
18
17
64
57
54
107
105
110
18
18
16
92
86
76
158
153
142
1998 Mar‡
1998 Jun‡
1998 Sep‡
32
35
35
36
34
35
34
34
32
31
31
47
48
46
46
43
42
43
42
41
40
43
89
81
66
100
95
94
73
69
72
67
65
1,668
1,774
1,601
1,554
1,404
1,347
1,330
1,222
1,179
1,126
1,060
1,352
1,212
1,012
930
809
773
711
677
677
667
639
623
824
741
349
324
284
569
499
553
534
559
193
183
155
194
211
216
223
202
227
220
225
25
22
30
35
30
29
27
24
24
21
23
44
49
57
52
46
46
45
43
42
43
41
26
29
36
38
45
43
43
46
49
52
55
30
35
28
21
13
8
12
12
11
10
9
80
65
39
49
44
40
34
33
29
29
28
84
83
81
67
51
51
48
44
46
44
45
Females
1971
1976
1981
1986
1991
1992
1993
1994
1995
1996
1997
31
42
68
1,181
686
587
243
22
45
49
11
31
43
1996 Dec
31
30
30
32
43
38
43
47
73
59
64
66
1,237
1,013
934
1,061
752
608
559
640
908
424
378
530
339
178
160
227
29
19
19
23
50
40
36
38
57
53
55
55
10
9
10
9
29
27
28
30
44
48
43
46
1997 Mar
1997 Jun
1997 Sep
1997 Dec
32
32
32
41
35
39
66
61
60
1,140
1,020
931
691
614
575
651
493
380
288
198
157
24
19
19
45
41
35
57
55
55
13
11
9
29
34
27
46
43
45
1998 Mar‡
1998 Jun‡
1998 Sep‡
(E810-E819) (E950-E959,
E980-E989
exc. E9888)
69
Office f or National Statistics
Health Statistics Quar terly 01
Spring 1999
Notes to tables
Changes to tables
With the introduction of Health Statistics
Quarterly, the previous Population Trends
tables have been reviewed and some small
changes introduced, in particular, a new table,
Table 2.2, showing key demographic and
health indicators for the constituent countries
of the United Kingdom.
For most tables, years start at 1971 and then
continue at five-year intervals until 1991.
Individual years are shown thereafter. If a year
is not present the data are not available.
Population
The estimated and projected populations of an
area include all those usually resident in the
area, whatever their nationality. Members of
HM forces stationed outside the United
Kingdom are excluded. Students are taken to
be resident at their term-time addresses.
Figures for the United Kingdom do not include
the population of the Channel Islands or the
Isle of Man.
The population estimated for mid-1991
onwards are final figures based on the 1991
Census of Population with allowance for
subsequent births, deaths and migration.
Live births
For England and Wales, figures relate to
numbers occurring in a period; for Scotland
and Northern Ireland, figures relate to those
registered in a period. See also Note on page
63 of Population Trends 67.
Perinatal mortality
In October 1992 the legal definition of a
stillbirth was changed, from baby born dead
after 28 completed weeks of gestation or more,
to one born dead after 24 completed weeks of
gestation or more.
Expectation of life
The life tables on which these expectations are
Office for National Statistics
70
based use current death rates to describe
mortality levels for each year. Each
individual year shown is based on a threeyear period, so that for instance 1986
represents 1985–87. More details may be
found in Population Trends 60, page 23.
Deaths
Figures for England and Wales represent the
numbers of deaths registered in each year up
to 1992, and the number of deaths occurring
in each year from 1993. Provisional figures
are registrations.
Figures for both Scotland and Northern
Ireland represent the number of deaths
registered in each year.
Age-standardised mortality
Directly age-standardised rates make
allowances for changes in the age structure
of the population. The age-standardised rate
for a particular condition is that which
would have occurred if the observed agespecific rates for the condition had applied
in a given standard population. Tables 2.2
and 6.3 use the European Standard
Population. This is a hypothetical population
standard which is the same for both males
and females allowing standardised rates to
be compared for each sex, and between
males and females.
Abortions
Figures relate to numbers occurring in a
period.
Marriages and divorces
Marriages are tabulated according to date of
solemnisation. Divorces are tabulated
according to date of decree absolute, and the
term ‘divorces’ includes decrees of nullity.
Government Office Regions
Figures refer to Government Office Regions
(GORs) of England which were adopted as
the primary classification for the
presentation of regional statistics from April
1997. A map showing the GORs is included
on page 71.
Health Regional Office areas
Figures refer to new health regions of
England which are as constituted on 1 April
1996.
Sources
Figures for Scotland and Northern Ireland
shown in these tables (or included in totals
for the United Kingdom or Great Britain)
have been provided by their respective
General Register Offices, except for the
projections in Table1.2 which are provided
by the Government Actuary.
Rounding
All figures are rounded independently;
constituent parts may not add to totals.
Generally numbers and rates per 1,000
population are rounded to one decimal place
(e.g. 123.4); where appropriate, for small
figures (below 10.0), two decimal places are
given (e.g. 7.62). Figures which are
provisional or estimated are given in less
detail (e.g. 123 or 7.6 respectively) if their
reliability does not justify giving the
standard amount of detail. Where, for some
other reason, figures need to be treated with
particular caution, an explanation is given as
a footnote.
Latest figures
Figures for the latest quarters and years may
be provisional (see note above on rounding)
and will be updated in future issues when
later information becomes available. Where
figures are not yet available, cells are left
blank. Population estimates and rates based
on them may be revised in the light of
results from future censuses of populations.
Health Statistics Quar terly 01
Spring 1999
Recent ONS publications
All Change? The Health Education Monitoring Survey one year on
(The Stationery Office, September, £30, ISBN 0 11 621065 6).
Travel Trends 1997 (The Stationery Office, November, £39.50, ISBN 0
11 621090 7).
Mortality Statistics 1996: injury and poisoning (The Stationery
Office, September, £30, ISBN 0 11 621070 2).
Family spending 1997–98 (The Stationery Office, November, £39.50,
ISBN 0 11 621047 8).
Focus on the South West (The Stationery Office, October, £30, ISBN
0 11 621064 8).
Mortality Statistics 1997 – Cause (The Stationery Office, December,
£35, ISBN 0 11 621095 8).
Social Focus on Women and Men (The Stationery Office, October,
£30, ISBN, 0 11 621069 9).
Mortality Statistics 1996 – General (The Stationery Office,
December, £25, ISBN 0 11 621094 X).
Smoking-related behaviour and attitudes, 1997 (ONS, October,
£8, ISBN 1 85774 289 3).
Abortion Statistics 1997 (The Stationery Office, December, £25,
ISBN 0 11 621093 1).
Psychiatric morbidity among prisoners in England and Wales
(The Stationery Office, October, £45, ISBN 0 11 621045 1).
Social Trends 1999 (The Stationery Office, January, £39.50, ISBN 0 11
621067 2).
Cancer statistics: registrations 1992 (The Stationery Office,
November, £30, ISBN 0 11 621091 5).
Congenital Anomaly Statistics (MB3 no. 26) (The Stationery
Office, January, £25, ISBN 0 11 621156 3).
G OVERNMENT O FFICE REGIONS
ENGLAND
AND COUNTIES IN
Government Office Regions
Northumberland
NORTH Tyne & Wear
EAST
Cumbria
Durham
NORTH
WEST
Cleveland
North Yorkshire
YORKSHIRE
AND THE
HUMBER Humberside
Lancashire
W Yorks
Greater
Manchester
MERSEYSIDE
Merseyside
S Yorks
Derbyshire
Cheshire
Staffordshire
Notts
EAST
MIDLANDS
Lincolnshire
Norfolk
Leicestershire
Shropshire
West
Midlands
WEST
MIDLANDS
Hereford &
Worcester
Northants
Cambridgeshire
Warwickshire
EASTERN
Beds
Suffolk
Bucks
Herts
Gloucestershire
Essex
Oxfordshire
LONDON
Avon
Berkshire
Wiltshire
Surrey
Hampshire
Somerset
West Sussex
SOUTH WEST
Devon
Kent
SOUTH EAST
East Sussex
Dorset
Isle of Wight
Cornwall
71
Office for National Statistics