Transcript shah_ebrahim_11oct

CADENZA Symposium 2008
Successful Ageing
Influence of socio-economic factors,
gender and health service provision
Shah Ebrahim
London School of Hygiene &
Tropical Medicine
Outline
• Socio-economic development and life
expectancy
• Social class, survival and disability
• Gender, survival and disability
• Life-course influences on disability
• Health services
Successful ageing requires survival
Socio-economic
position
Gender
Survival
Health
services
Socio-economic development
and ageing
• Life expectancy: comparisons between
countries by income levels
Life expectancy and GDP
Lynch et al. BMJ 2000;320:1200
Preston’s curves: explanations for better
health
1900’s
Preston, S. H Int. J. Epidemiol. 2007 36:484-490; doi:10.1093/ije/dym075
Preston’s conclusion
• Improvements in survival are not all
explained by economic growth
• Nutrition and education have had only a
small role.
• Global diffusion of medical and health
technologies:
– innovations in hygiene and sanitation
– maternal and child services
– specific vaccines and drugs for treatment
of bacterial infections
Paradoxes of
Costa Rica,
Cuba, Sri
Lanka: high
life
expectancy
but low GDP
Marmot M, Clinical
Medicine, 2006
Social class (an English view)
I’m upper
class. I
look down
on both of
them
I’m middle
class. I look
up to him but
I look down
on him
John Cleese
Ronnie Barker
Ronnie Corbett
I know
my
place
Social class and life expectancy:age 65
Days of life expectancy
1972-6
7000
6500
6000
5500
5000
4500
4000
3500
3000
6388
6132
1997-9
5950
5512
I
II
IIInm
IIIm
Social class
5037
4891
IV
V
Locomotor disability and social class:
British Regional Heart Study men
5
Odds ratio
4
3
2
1
0
I
12.3%
II
19.2%
IIInm
21.5%
Source: Ebrahim et al, Int J Epidemiology (2000)
IIIm
28.5%
IV
33.7%
V
40.1%
Social class and disability:
possible explanations
Social class
Disability
Chronic diseases:
Arthritis, CVD
Risk factors: inactivity,
smoking, BMI etc
Locomotor disability and social class:
British Regional Heart Study men
5
Odds ratio
4
Adjusted
for
smoking,
BMI,
activity and
alcohol
3
2
1
Excluding men with CVD,
arthritis and respiratory
0
I
12.3%
II
19.2%
disease
IIInm
IIIm
21.5% 28.5%
Source: Ebrahim et al, Int J Epidemiology (2000)
IV
33.7%
V
40.1%
Material and psycho-social
models of causation
Poverty
MATERIAL
CONDITIONS
Inadequate diet
Smoking
Poor housing
PSYCHO-SOCIAL
CONDITIONS
Lack of
health &
social
services
Lack of control
Increased stress
Low social capital
Reduced survival
A metaphor: air travel: differences
in a neo-material and psychosocial
theory
First class
Cattle class
Lynch & Davey Smith BMJ 2000;320;1200-1204
Material vs. psychosocial
explanations
Compare air travellers in first and
economy class. Travellers in economy
have worse health because they sat in a
cramped space and couldn't sleep not
because they could see the bigger seats in
first class
Lynch & Davey Smith. BMJ 2000;320:1200
Implications for intervention
• psychosocial interpretation: health
inequalities would be reduced by
abolishing first class, or mass
psychotherapy to alter perceptions of
relative disadvantage.
• neomaterial viewpoint: health
inequalities can be reduced by
upgrading conditions in economy class
Lynch & Davey Smith BMJ 2000;320;1200-1204
Social inequalities and survival
• Growing wider
• Not fully explained by smoking,
diet, exercise
• Potentially avoidable
Gender, survival and disability
Life expectancy at age 65
Office of National Statistics, UK
Percentage of life expectancy spent
able to get outdoors, 1991
90
85
Men
80
Women
Percentage
75
70
65
60
55
50
8.6
4.8
11.2
6.1
75-79
85+
75-79
85+
45
40
Source: Bone et al Health Expectancy, 1995
Distribution of walking time
12 Time to walk 6m.
11.5
2% increase per
11
single year increase
in age, p<0.001)
10.5
10
9.5
9
8.5
8
1 (64, 66)
2 (66,69)
3 (69,72)
4 (72,75)
Age quintiles (range)
5 (75,82)
Adult social class, 2002/3
11
Time to walk 6m.
4.9%
increase in
walking time
per category
increase in
social class,
p=0.02
10.5
10
9.5
9
8.5
8
7.5
7
I&II
III
IV&V
Adult occupational social class, 2002/3
Household income, 1937/9 and
walking speed in 2002/3
10.5
Time to walk 6m.
3.2%
reduction in
walking time
per category
increase in
income,
p=0.04
10
9.5
9
8.5
8
7.5
7
high
>HK$120
low
HK$90-120
HK$60-89
<HK$60
Weekly household income, 1937/9
You need to
walk at 0.8 m/s
to cross a Hong
Kong road
The youngest
participants
(aged 64-66)
only walked at
0.7 m/s!
Guardian 9 September 2004
Inner-Age? Pharmanex? Isolagen?
Health services for older people
• Complex interventions - combinations of
interdisciplinary teamwork for health and
social problems
• Do they work?
MRC trial of multidimensional
assessment and management
• 40,000 older people randomized to
different care: death and institutional care
• Comparisons of geriatric service vs.
primary care service
• Comparison of targeted service vs.
universal service
• After 10 years work – geriatric service
slightly worse than primary care and
universal no better than targeted service
Components of complex
interventions
• Assessment
• Primary prevention
Physical activity
Environment, home safety
Self care, immunisation
Social network
• Secondary prevention
Treatment of chronic conditions
• Tertiary prevention
Medication review, rehabilitation
Meta-analysis of 45 trials
Fav ours interv ention
0.2
0.5
Fav ours control
1
2
5
Test f or ov erall ef f ect: Z = 3.10 (P = 0.002)
Test f or heterogeneity : Chi² = 80.83, df = 44 (P = 0.0006), I² = 45.6%
Total ev ents: 2079 (Interv ention), 2259 (Control)
Total (95% CI)
13278
12462
Relative risk of not living in own home
Newcomer 2004
Leung 2004
Kono 2004
Nikolaus 2003
Cunlif f e 2004
Shapiro 2002
Newbury 2001
Hebert 2001
Hogan 2001
By les 2004
Rockwood 2000
Eekhof 2000
Dalby 2000
Caplan 2004
Reuben 1999
Nikolaus 2 1999
Nikolaus 1 1999
Lev eille 1998
Kerse 1999
Close 1999
Wallace 1998
Engelhardt 1996
Beck 1997
Tinetti 1999
Trentini 2001
German 1995
Hansen 1995
Wagner 2 1994
Wagner 1 1994
Fabacher 1994
Dunn 1994
Martin 1994
Burton 1995
Stuck 1995
Silv erman 1995
Melin 1992
Hansen 1992
Carpenter 1990
Hall 1992
Rubin 1993
Epstein 1990
Clarke 1992
Williams 1987
Vetter urban 1984
Vetter rural 1984
Tulloch 1979
or sub-category
Study
Outcome:
Comparison:
Rev iew:
141/1338
14/130
9/58
33/173
63/185
2/34
3/46
17/245
4/70
145/862
26/95
40/679
7/66
87/369
0/177
65/169
63/173
1/96
1/130
37/178
45/45
9/80
6/118
8/148
9/77
204/1777
24/96
14/590
18/618
4/104
28/102
9/29
442/1824
30/215
9/239
63/150
48/168
66/247
21/81
30/92
15/185
68/260
12/54
49/296
57/281
33/129
107/1374
10/130
14/59
30/166
65/185
15/34
7/49
23/249
6/79
69/553
15/87
54/652
4/58
81/369
5/180
74/173
74/173
2/95
4/123
45/208
45/45
7/80
11/151
11/156
17/65
272/1796
27/97
23/317
23/317
4/95
35/102
13/25
514/1894
33/199
13/203
53/99
72/181
57/243
30/86
30/87
20/205
71/262
15/57
71/298
57/273
31/133
n/N
Interv ention
Favours
intervention
Favours
control
0.87 (95% CI 0.79, 0.94)
n/N
Control
01 Not liv ing at home (death and nursing home)
03 Not liv ing at home
Interv entions to prev ent disability in the elderly
Beswick A et al, Lancet 2007
95% CI
RR (random)
Health care and social support
• Effective services
– evidence base patchy in LMICs
• Affordability
– privatisation of long-term care
• Accessibility
– waiting lists, local treatment
• Appropriateness
– growing private anti-ageing sector
Number of admissions to hospitals in
the three years before death, England,
1999-2000.
Dixon, T. et al. BMJ 2004;328:1288
Projections of long-term care costs
30
£ billions
25
20
£19.9
£14.7
£11.1
5
10.9
NHS care
15
10
£28.0
Private
expenditure
Social services
7.8
10.1
5.7
4
7.2
5.5
4.5
2.6
3.5
1995
1.6%
2010
1.5%
4.9
7
0
GDP
With Respect to Old Age, Cm 4129, 1999
2021
1.6%
2031
1.8%
Summary
• Socio-economic factors play a major role
in determining survival and disability
• Women do better than men in terms of
survival but not in terms of disability
• Health services do improve survival and
reduce institutionalisation
• But too much health service use is a
problem for many