The health transition in the West

1
The health transition in the West

• MStud Public Health, Oct 2002
• John Powles

2
Public health

• 'the science and art of preventing disease,
  prolonging life and promoting health through the
  organised efforts of society
• Acheson ED (chairman). Public health in
  England The report of the Committee of Inquiry
  into the future development of the Public Health
  Function. London HMSO 1988. (Cm 289), p 1

3
Public health practice

• Rests on assumptions about what determines health
  levels in human populations
• These assumptions need to be tested against
  historical experience

4
Exploring the range of human health experience

• Three main topics
• The health transition in the West
• The health transition in the Third World
• Trends, and determinants of, adult mortality
• (incl. the failed transition in CCEE)

5
How can health levels in populations be compared?

• As a first approximation
• By their mortality levels
• These will tend to understate differences in
  health
• Because levels of health among the living tend to
  be less favourable in populations with high
  mortality
• (unless injury plays a large role in elevating
  mortality eg hunter-gatherers)

6
Four survival patterns and transitions between
them
7
Mortality patterns are closely linked to
fertility patterns

• In the long run death rate birth rate
• Except in stationary populations
• The distribution of deaths by age at death, and
• The age structure of the population
• Are much more sensitive to the birth rate than to
  the death rate
• Population aging has been much more
  powerfully influenced by the decline in fertility
  than the decline in mortality

8
How do we know about mortality and fertility
levels in the past?

• Official systems for vital registration
• In the UK from 1837
• Family reconstitution
• In England from the 16th century

9
What is a vital registration system?

• A system for the official and compulsory -
  registration of vital events including
• Births
• Deaths
• Marriages
• Divorces
• plus associated systems for collation and
  analysis of such data
• In the UK the General Register Office began
  operations in 1837

10
What is family reconstitution?

• In eg early modern England, vital events were
  recorded by the established Church (baptisms,
  burials, marriages)
• Starting eg with a record of a baptism
• Back to marriage and baptism of parents
• Forward to death of subject, and so
• Reconstitute families and then estimate vital
  rates

11
Comparing survival patterns

• Summary measure
• Life expectancy at birth
• Mean of all life durations
• Or Mean age at death (in a life table
  population)
• But
• Mean does not convey distribution (ie survival
  patterns) well when many lives are very short

12
(No Transcript)
13
(No Transcript)
14
So when considering transitions across a wide
range of survival (e0) levels

• It is more informative to consider separately,
• changes in
• Survival chances in childhood
• And
• Survival chances in adulthood
• Additionally because
• these have varied independently through time

15
Summary measures of child survival

• Infant mortality rates
• Conceptually
• probability of death by age 1
• Operationally measured as
• Drawback
• Direct estimation requires vital statistical
  system ie not very practical in high mortality
  populations

16
Under 5 mortality rates

• Conceptually
• probability of death before 5
• Operational estimation
• Indirect from survey Qs to women about
• Number of children born
• Number who have died?
• (Brass technique)
• Main measure now used for low and middle income
  countries
• (IMR estimates also derived this way, but less
  robustly)

17
For an intuitive summary of population experience

• May use
• chance of surviving to (or dying before)
  adulthood (taken as 15)
• Even at e0 35
• corresponds closely to U5MR
• 93 of those surviving to 5 survive to 15
• At e0 75
• 99.8 of those surviving to 5 survive to 15

18
Summary measures of adult survival

• Adult mortality rate
• probability of dying before 60
• Given survival to 15
• Most widely used measure eg by World Bank
• But 60 is rather low for low mortality countries
  (especially when you are my age!)
• I use probability of surviving/dying between 15
  and 65
• (difference will be small)

19
Life expectancy in England since the C17
20
Sweden life expectancy since the mid C18
21
Changing survival chances England C17 to late
C18 - childhood
22
Changing survival chances England C17 to late
C18 - adulthood
23
Changing survival chances England since the late
C18
24
Changing survival chances England since the C17
NB In early modern times, 70 of those alive at
15 died before 65. The idea that that high
mortality was concentrated in childhood is wrong!
25
Why have survival chances improved since early
modern times?
Increased knowledge and changed outlook
Changed responses to disease (medicine)
Social and econ-omic development
Better health
26
The intervention hypothesis
Increased knowledge and changed outlook
Better health
27
Major advances in effectiveness of clinical
interventions
28
Subtitle Medical planning based on evaluation
of medical achievement Published 1965
29
A newly effective intervention for a common cause
of death

• Late 1940s to early 1950s
• Streptomycin, isoniazid and PAS shown to be
  effective in treating tuberculosis and brought
  into widespread use

30
TB mortality, England Wales1900-1971
extrapolation of linear trend for 1921 to 1946
(after McKeown)
prevented deaths
31
Estimated number of deaths from TB prevented by
chemotherapy
32
Tuberculosis trends lengthening the time frame
33
Estimated number of deaths from tuberculosis
prevented by chemotherapy
34
The nutrition hypothesis
Increased knowledge and changed outlook
?
Better health
35

• Economic development
• Increased personal incomes
• Increased command over food
• Better nutrition
• Better health
• (especially because of increased resistance
  against infection)
• Recently elaborated by Robert Fogel

36
Net nutrition (growth) and survival in infants

• Many studies in Third World populations since the
  1950s

37
Relative risk of dying in the next 6 months by
of the Harvard weight for age norm
Indian infants aged 1 to 36 months
38
Net nutrition and survival in adults

• Height summarises net nutrition in childhood
• Weight for height summarises recent adequacy (or
  excess) of dietary energy (relative to
  expenditure)

39
Waaler surface Relationship between height,
weight and risk of death based on follow-up of
309000 Norwegian males
40
Waaler surface Relationship between height,
weight and risk of death based on follow-up of
309000 Norwegian males
French mean 1785 163cm, 50 kg
.
41
Waaler surface Relationship between height,
weight and risk of death based on follow-up of
309000 Norwegian males
42
Some problems with generalising Waalers findings
to explain historic declines in mortality

• Relationships vary with disease risks faced
• Taller is (usually) better for vascular disease
• but not for cancer (eg breast cancer)
• Fatter is better when TB is your main risk, but
  not when heart attack is

43
TB risk by height and weight 800,000 US Navy
recruits tested 1958 1967
Proportions developing tb over 3 to 12 years of
follow-up
44
Some problems with generalising Waalers findings
to explain historic declines in mortality

• Transitional populations with abundant food
  didnt have low mortalilty
• Eg
• Late C18 French Canadian settlers
• Late C19 Argentinians

45
Some problems with generalising Waalers findings
to explain historic declines in mortality

• The relationship between levels of economic
  development and levels of mortality changed
  profoundly through the 20th century

46
The changing relationship between life expectancy
and income (Preston et al)
47
The nutrition hypothesis
Increased knowledge and changed outlook
?
?
Better health
48

• Historical evidence against major health benefits
  from economic development per se

49
Economic development and mortality in the USA
Source Fogel, and Costa, 1997, p 61
50
The 'Antebellum puzzle' in the United States

• Across around 1000 US counties in the 1840s,
• Mortality was higher in those that were more
  developed
• Eg More urbanised
• Richer
• With canal connections

Haines MR, Craig LA, and Weiss T. Development,
health, nutrition , and mortality The case of
the 'Antebellum puzzle' in the United States.
Cambridge, MA National Bureau of Economic
Research. 2000 NBER Working paper series on
historical factors in long run growth.
51
The 'Antebellum puzzle' in the United States

• Rapid economic development in the United
  States in the three decades prior to the Civil
  War were characterized by fast urban growth,
  significant migration from abroad, considerable
  internal mobility, great changes in the
  transportation infrastructure, and increased
  commercialization, including in agriculture.
  These all contributed to a worsening mortality
  environment which had adverse consequences for
  human growth. This may be characterized as
  "Smithian" growth via extension of markets with
  negative externalities.

Haines MR, Craig LA, and Weiss T. Development,
health, nutrition , and mortality The case of
the 'Antebellum puzzle' in the United States.
Cambridge, MA National Bureau of Economic
Research. 2000 NBER Working paper series on
historical factors in long run growth.
52
C19 England life expectancy and stature
53
The intervention hypothesis pt 2
Increased knowledge and changed outlook
Including public health activities
Better health
54
Overcoming the urban penalty, Sweden, 1860s
1920s
55
Role of the sanitary reforms

• 1842
• Report to House of Lords on 'Sanitary Condition
  of the Labouring Population..'
• 1848
• Public refuse removal in London
• 1865
• London Metropolitan Board of Works develops new
  network of sewers
• 1869
• Extension of franchise for local gov't
• 1875
• Filtered water supplies
• 1900
• Chlorination of water supplies

56

• So do the C19 public health reforms
• concentrated on clean water and sewerage
• deserve most of the credit?

57
Objection 1

• The biggest absolute declines in the late C19
  were in deaths from air-borne infection (McKeown)
• not likely to be due to sanitation?

58
Sewerage and water supplies not directly
connected to declines in respiratory infection
59
But it was noted at the time

• . that the purification of polluted public
  water-supplies reduces the general death-rate
  much more than it would be reduced by the saving
  of deaths from the commonly recognized
  water-borne disease, typhoid fever and Asiatic
  cholera alone.
• Sedgwick WT, MacNutt JS. On the Mills-Reincke
  phenomenon and Hazen's theorem concerning the
  decrease in mortality from diseases other than
  typhoid fever following the purification of
  public water-supplies. J.Infect.Dis. 1910 7
  489-564.
• Which is plausible given the harmful effects of
  recurrent bouts of diarrhoea.

60

• Objection 2
• Mortality decline was actually most rapid in
  the first decades of the C20
• when public health programmes had new
  priorities

61
Life expectancy in England since the C17
From family reconstitution (mostly rural)
From vital registration
62
The drop in child mortality in the first 3
decades of the C20
In US from 1895 to 1930 in E W from 1901 to
1931
Source (secondary) Ewbank and Preston, 1990
63

• It becomes clear that the problem of infant
  mortality is not one of sanitation alone, or
  housing, or indeed of poverty as such, but is
  mainly a question of motherhood.
• Newman, 1906
• Typhoid fever and malaria can often be routed
  on a large scale by the engineer but infant
  mortality must be met and conquered in the home.
• Winslow, 1909

64
Ewbank and Preston, 1990

• What changes in childrearing practices were
  experts seeking?
• Did mortality differentials increase consistent
  with greater uptake by those with more schooling?

65
Advancing knowledge Causes of diarrhoea
according to Diseases of children
66
Changes recommended

• Breast feed (in effect, may have slowed decline)
• Boil cows milk, sterilise bottles
• Protect infants from persons known to be
  infective
• Control flies
• Wash hands
• etc

67
Mass reach and participation (US)

• Infant care (US Childrens Bureau)
• 12 m copies
• Baby care columns in leading newspapers
• Nearly all by 1912 (though not a few years
  earlier)
• Baby weeks
• By 1919 17,000 local committees
• 11 m directly involved
• Also in England and New Zealand etc

68
Winslow (1929)

• ...the discovery of the possibilities of
  wide-spread social organization as a means of
  controlling disease was one which may almost be
  placed alongside the discovery of the germ theory
  of disease itself as a factor in the evolution of
  the modern public health campaign.

69

• Did those with more schooling benefit more?

70
US under 5 mortality rates by paternal
occupation, early C20
71

• ie
• A complex mix of institutional and behavioural
  changes fed by new knowledge, with those with
  most schooling benefiting most

72

• More recent examples of complex institutional
  change leading to health improvement
• (Success without interventions)
• 1 Smoking cessation in the 50s 60s

73
Smoking cessation in the 50s 60s
74
(No Transcript)
75
(No Transcript)
76
HIV transmission in English homosexual men in
1983/84
Source of data D. de Angelis and colleagues
77
HIV transmission in English homosexual men in
1983/84
78
Sudden infant death rates, England and Wales,
1985-1994
79
Sudden infant death rates, England and Wales,
1985-1994
Beal, Lancet letter, Aug 88
Back to sleep Campaign, Dec 91-
80
Beyond the intervention and nutrition
hypotheses
Increased knowledge and changed outlook
Changed public under- standing behaviour
Other roles of professionals
Better health
81
Beyond the intervention and nutrition
hypotheses
Increased knowledge and changed outlook
Changed public under- standing behaviour
Institutional modernisation
Other roles of professionals
Better health
82
Beyond the intervention and nutrition
hypotheses
Increased knowledge and changed outlook
Changed public under- standing behaviour
Institutional modernisation
Other roles of professionals
Better health
83
(No Transcript)
84
Non-medical
Medical
Commodities
Institutions
Health
Knowledge
85
Summary

• Marked increase in survival chances associated
  with modernisation - with major reversals in C19
  cities
• Idea that this was due to the cumulative effects
  of successful interventions is clearly wrong
• Paths by which new knowledge has been put to work
  to improve health have been complex and varied

86
Summary - 2

• An alternative (simple) explanation that relies
  heavily on the role of increased private incomes
  and improved nutrition is also inadequate
• Contribution of economic and social
  transformations has been more complex
• (also shown clearly in recent 3rd world
  experience)
• Inequalities in premature mortality have
  increased since the early C18, esp during C20