Health inequalities

1
Health inequalities

• John Powles
• Public Health Topics, Block 4
• MSt in Public Health 2004

2
Understanding determinants of health
Understanding determinants of health inequalities
3
Outline

• Nature of social inequality
• Emergence of health inequality by rank in the C20
• C19
• Gradients in child health at the turn of the C20
  USA and GB
• Adult mortality in the UK
• Adult mortality in transition countries
• Explanations
• materialist, cultural/knowledge,
  psychosocial
• Solutions?

4
Nature of social inequality (in relation to
health)

• Inequalities by rank vs inequalities contingent
  on other things
• Eg living in a marshy area
• Dimensions of social inequality (ranking)
• Class / Status / Power

5
Dimensions of inequality analytical
capitalist
celebrity
politician
6
Dimensions of inequality empirical
Income / wealth
Occupation / social class /SEC
Schooling
7
The Registrar-Generals Social classes

• Developed by Stevenson to analyse health
  inequalities around 1911 census
• I Professional, etc occupations
• II Managerial and Technical occupations
• III Skilled occupations
• (N) Non-manual
• (M) Manual
• IV Partly skilled occupations
• V Unskilled occupations.

8
Pros .and Cons

• Produces nice gradients
• Easy to use only needs (detailed) occupational
  codes coding manual

• Circular
• Ie designed to produce gradients of health status
  and it does!
• Atheoretical
• Doesnt locate individual in any theoretically
  specified space

9
(No Transcript)
10
Theoretical basis of NS-SEC

• Based on Goldthorpe schema
• Makes employment relations central
• Mode of payment
• Promotion prospects
• Autonomy

11
Employment relations

• Labour contract
• Relatively short-term and specific exchange of
  money for effort
• Service relationship
• Rewards expected in longer as well as shorter
  term
• More autonomy

12
Information items needed

• Detailed occupational description (to allow 3
  digit coding)
• Employer / own account / employee
• Whether others are formally supervised
• Number of employees in establishment

13
Does anyone have any experience of working with
2001 census data?
14
Inequality by place in the C19
15
C19 England life expectancy and stature
16
Overcoming the urban penalty, Sweden, 1860s
1920s
17
Inequality by rank before the C20?

• In Britain the aristocracy had no survival
  advantage before the c18
• Then aristocratic males started to gain (but at
  end of C19 still not ahead of (low wage) healthy
  districts)
• Aristocratic females did worse til C19 higher
  maternal mortality (?iatrogenic)
• In China Imperial family also had no survival
  advantage over general population

18
(No Transcript)
19
1900 US census and 1911 GB census

• Both asked fertility questions (inc children
  ever born, number still alive)
• Allows indirect estimation of U5MR for aggregates
  of individuals
• Can be linked to other census data for
  individuals / families
• Relative mortality for chosen aggregates can then
  be expressed as a ratio of observed to expected
• First chance to test independent importance of
  residential environment vs social rank

20
U5MR by fathers occupation US, 1900
U5MR for children of physicians only 6 below
average
Source Preston, 1985, p 379
21
U5MR by fathers occupationUS 1890s to 1920s
Source Ewbank Preston, 1990, p 131
22
Interpretation of emergence of inequal-ities in
child survival by social rank

• When knowledge of effective means of protecting
  childrens health was developed and actively
  disseminated those best able to make use of it
  did so and their childrens prospects improved
  (relative to national averages)
• Consistent with inequalities today in low
  income countries (see next)

After Preston et al, circa 1990
23
U5MR by fathers occupation US, 1900 and
composite of developing countries, 1980
Source Preston, 1985, p 379
24
(No Transcript)
25
(No Transcript)
26
U5MR index (O/E) by social class of head of
household, 1911 census, Britain (Model 2)
Garrett et al, 2001
27
U5MR index (O/E) by area of residence, 1911
census, Britain (Model 1)
Garrett et al, 2001
28
U5MR index (O/E) by area of residence, patern-al
class and other variables, 1911 census, Britain
Garrett et al, 2001
29
Child mortality at the beginning of the C20

• Was only weakly related to class once area of
  residence was controlled
• Much of the class effect was accounted for by
  fertility differences
• The children of agricultural labourers and East
  European immigrants (Jewish but poor) did very
  well

30
Increase in adult male mortality differentials
through the middle half of the C20, England
Wales
Ratios of directly age-standardised rates to
those for total population in 1921 and 1971
31
Conclusion

• Increased inequalities mainly result from
  greater gains in the upper social ranks.
• (Some part of the apparent deterioration of the
  lower ranks will be due to re-classification ie
  becoming a more extreme group because
  representing a smaller proportion of the total
  distribution.)

32
Explanation?
33
(No Transcript)
34
(No Transcript)
35
Social transfers as of GDP, high income
countries 1995
Welfare States
Private market economies
welfare unemployment pensions health
housing
Source Lindert, Growing public, 2004
36

• Little apparent association between
  redistributive social policies and magnitude of
  health inequalities
• Lindert finds a weak association between level of
  social transfers and absolute adult mortality
  levels (lower in welfare states)

37
Excess mortality in male manual workers Hungary
compared to England and Wales
E W / Hungary Data are 20q45 (probability of
dying between 45 and 65)
Proportions non-manual 43 / 28
Period 1981-9 / 1988-92
Type longitudinal /
cross-sectional
Kunst et al, 1998
38
Manual / non-manual rate ratios by cause, Hungary
and England and Wales
39
Theories of inequalities

• Materialist
• Cant explain increased differentials since
  early c20
• highly variable relations of
  mortality levels to
  real income eg rural Crete
• Psychosocial
• Assumed neuro-humoral pathway doesnt fit
  so easily with heterogeneity of underlying
  causes
• Kunst et al
• Higher strata show a generalised ability to
  better avoid premature death

40
The meaning of poverty

• Conventional view
• Absolute in the space of commodities (poverty
  lines)
• Relative in social space (relative poverty)
• Sen
• Relative in the space of commodities
• Absolute in the space of capabilities
• Ie need for commodities is defined by what one
  needs to develop ones capabilities within a
  given social context

41
What is poverty? (2)

• Adam Smith
• Recognised that the need not to be ashamed to
  appear in public required different commodities
  in different cultural and historical contexts
• (in mid c18 Scotland - a linen shirt, shoes
  etc)
• A social functioning in Sens terminology

42
Amartya Sen

• The value of the living standard lies in the
  living, and not in the possessing of commodities,
  which has derivative and varying relevance.

The standard of living, 1987
43
(No Transcript)