Social determinants of health

1
Social determinants of health

• Martin Bobak
• University College London

OSI conference on Graduate Public Health
Education, Kiev, July 1-3, 2004
2
Social determinants of health

• Ubiquitous
• Powerful influence on health
• Underlie
• Differences between individuals within countries
• Differences between groups, populations and
  countries
• Changes in health over time
• Complex pathways

3
Mortality at Titanic by passenger class, men and
women
Women
Children
From Broom L Selznick P, 1968
4
Global Burden of DiseaseFrom Ezzati et al,
Lancet 2002

• Childhood and maternal undernutrition
• Underweight
• Iron deficiency
• Vitamin A deficiency
• Zinc deficiency
• Other nutrition-related risk factors and physical
  activity
• High blood pressure
• High cholesterol
• High BMI (overweight and obesity)
• Low fruit and vegetable intake
• Physical inactivity
• Sexual and reproductive health
• Unsafe sex
• Lack of contraception

• Addictive substances
• Tobacco
• Alcohol
• Illicit drugs
• Environmental risks
• Unsafe water, sanitation, and hygiene
• Urban outdoor air pollution
• Indoor smoke from solid fuels
• Lead
• Global climate change
• Occupational risks
• Risk factors for injuries
• Carcinogens
• Airborne particulates
• Ergonomic stressors
• Noise
• Other selected risks
• Unsafe health-care injections
• Childhood sexual abuse

5
Global Burden of Disease attributable to risk
factors ( of DALY)
From Ezzati et al, Lancet 2002
6
Odds ratio for mammography uptake in London women
65 (from Harris et al 2002)
7
Differences between populations
8
(No Transcript)
9
Mean score of self-rated health in 25 European
countries 1990-91 (higher score means worse
health), men and women 35-64, from Carlson 1998.
10
LIFE EXPECTANCY AND INCOME FOR SELECTED
COUNTRIES AND TIME PERIODS
World Bank Development Report 1993
11
Mortality ratios East/West and Low/High social
class in the UK, men
Chr Rhemat
Stroke
Bronchitis
CHD
Suicide
Ca rectum
Ca larynx
RTA
Ca lung
Melanoma
Ca lymphatic
Ca colon
Ca prostrate
12
Mortality ratios East/West and Low/High social
class in the UK, women
TB
Stroke
Ca stomach
CHD
Ca rectum
Bronchitis
Ca larynx
Ca uterus
Suicide
Melanoma
Ca lung
Ca breast
13
Differences between individuals (within
populations)
14
Mortality rate ratio comparing lower to higher
educational group for major causes of death, in
Czech Republic, Hungary, Estonia and USA, men
45-59 years.
From Kunst, 1997
15
Lower vs higher education mortality rate ratios,
Russia, 1989, from Shkolnikov et al 1998.
16
Figure 17. CVD mortality by education and marital
status in Warsaw MONICA, 10-yr follow-up.
Men
Women
17
Mean depression (CES-D) score by education in
Czech Rep, Russia and Poland in 2000.
18
Mean depression (CES-D) score by material
deprivation in Czech Rep, Russia and Poland in
2000.
19
Changes over time
20
Life expectancy at age 15 in Europe, men,
1970-2001.
21
Life expectancy at age 15 in Europe, women,
1970-2001.
22
Figure 16. Changes in mortality (SMR) in Russia
by education between 1988-89 and 1993-94 (from
Shkolnikov).
57
MEN
WOMEN
30
35
8
SMR in 1988-89 100
23
Increase in educational differentials in
mortality between 1980s and 1990s in St
Petersburg men
From Plavinski et al 2003
24
Trends in probability of survival in men by
education (relatives study)
45 p20 probability of living to 65 yrs when
aged 20 yrs
25
Trends in birthweight by maternal education in
the Czech Republic 1989-1996 (differences from
primary educated)
26
Mortality by education in Czech men and women,
1990/91 and 1995
Men
Women
From Dzurova et al.
27
Death rates in Russia 1980-2002both genders, per
100,000
Coup against Gorbatchev, breakup of USSR
Rouble crisis
Gorbatchev elected
28
Proportional changes in age-specific mortality in
Russia
Shkolnikov et al, 2001
29
(No Transcript)
30
Russian mortality from all causes, 35-69 yrs,
rate ratios
31
(No Transcript)
32
RR of CVD death by drinking frequency and mean
dose per occasion in Novosibirsk men
5 of the men
From Malyutina et al, Lancet 2003
33
Relative risk of all cause mortality by drinking
frequency and bingeing in Russian men (n8616)
Adj. for age, smoking, calendar period of birth
and relative
34
Relative risk of all cause mortality by drinking
frequency and bingeing in Russian women (n2730)
23/216 only
Adj. for age, smoking, calendar period of birth
35
Trends in probability of survival in men by
education (relatives study)
45 p20 probability of living to 65 yrs when
aged 20 yrs
36
Reaction to economic change
From Rose 2001
37
Change in GDP after 1989 by country
Index , 1989 100
From Unicef, Social Monitor 2003
38
Change in fertility after 1989
Crude birth rate live births / 1000 population
From Unicef, Social Monitor 2003
39
Gini coefficient, 1989 and 2000
OECD
UNICEF 2003
40
CHANGE IN LE BY SOCIAL STRESS IN 12 RUSSIAN
REGIONS, 1989-94
STRESS CAUSED BY UNEXPECTED SITUATIONS
CHANGE IN MALE LIFE EXPECTANCY
UNEMPLOYMENT, LABOUR TURNOVER, CHANGE IN
MARRIAGE AND DIVORCE RATE
(Cornia 1997)
41
CHANGE IN LIFE EXPECTANCY BY INCREASE IN INCOME
INEQUALITY, 1989-95
Cze
Slo
Svk
Pol
Hun
Bul
Rom
Mol
Lit
Est
Bel
Rus
Marmot Bobak, BMJ 2001
42
Mechanisms (pathways)
43
Material factors
Social structure
(1)
(4)
Work
Brain Neuro- endocrine and immune
Psychosocial / psychological
Social Environment
(5)
(2)
Health Behaviours
(3)
(6)
Patho-physiological changes Organ impairment
Early Life
Well-being Mortality Morbidity
Genes
Culture
44
Societal factors
45
Perceived control over own health
and perceived health
)
30

(

h
t
l
25
a
Rus
Est-Rus
e
Lat-Rus
h
Lit-Rus

Lat-Lat
r
o
Kal
20
o
Lit-Lit
p
Kau

f
o
Est-Est
15

e
c
Kr
n
e
10
l
War
a
v
Cz
r-0.82 (p0.001)
e
r
5
P
4.0
3.8
3.6
3.4
3.2
3.0
2.8
2.6
2.4
2.2
Perceived control over health
46
PERCEIVED CONTROL IN NATIONAL SAMPLES AND ALL
CAUSE MORTALITY
RU
LA
ES
SDR all causes (per 100,000)
HU
PO
LI
CZ
CONTROL (AGE-SEX ADJUSTED)
Pikhart, 2000
47
INCOME INEQUALITY MORTALITY IN CANADA USA
Ross et al BMJ 2000
48
Hostility and mortality in US metropolitan cities
49
Policy response

• More difficult to quantify
• Central European countries responded by policy
  measures to limit the impact of transition
• Baltic countries responded later
• FSU - the response was inadequate
• Why has the policy response been so different?

50
Classical risk factors health behaviours
51
Contribution to risk factors to educational
gradient in mortality in the St Petersburg
Men
Women
From Vagero et al, unpublished
52
Psychosocial factors
53
(No Transcript)
54
Depression and effort-reward imbalance in Russia,
Poland and Czech Rep.
Quartiles of effort-reward ratio
Adjusted for age, sex and centre
(Pikhart et al. Soc.Sci. Med. In press 2003)
55
Odds ratio for new CHD in Whitehall II by
Employment Grade - Men
Marmot et al, Lancet 1997
56

• By necessaries I understand not only the
  commodities which are indispensably necessary for
  the support of life, but what ever the customs of
  the country renders it indecent for creditable
  people, even the lowest order to be without
  the poorest person would be ashamed to appear in
  public without them.
• Adam Smith, Wealth of Nations 1776

57
POOR HEALTH IN HUNGARY BYBASIC HOUSEHOLD ITEMS
Odds ratio of poor health
Age-sex adjusted Multivariate
adjusted
washing machine, fridge, freezer, microwave,
phone age, sex, educ, marital, material
deprivation Pikhart 2000
58
POOR HEALTH IN HUNGARY BYSOCIALLY ORIENTED ITEMS
Odds ratio of poor health
Age-sex adjusted Multivariate
adjusted
colour tv, stereo system, motorbike, car, car
radio age, sex, educ, marital, material
deprivation Pikhart 2000
59
POOR HEALTH IN HUNGARY BY LUXURY ITEMS
Odds ratio of poor health
Age-sex adjusted Multivariate
adjusted
cable tv, satellite, video record/cam, CD, PC,
dishwasher, dacha age, sex, educ, marital,
material deprivation Pikhart 2000
60
Conclusions

• Socioeconomic factors make major contribution to
  differentials between and within countries
  worldwide, included Central and Eastern Europe
• Play major role in temporal changes
• Their effect is mediated by a combination of
  material, behavioural and psychosocial factors