Making sense of evidence Burden of Disease, Projections of Burden and Health Expenditure Opportuniti

1
Making sense of evidence Burden of Disease,
Projections of Burden and Health Expenditure
Opportunities for Prevention

• Prof Theo Vos
• Centre for Burden of Disease and
  Cost-Effectiveness
• Case-mix conference Adelaide
• 17 November 2008

University of Queensland School of Population
Health
2
Background
Finite resources in health choices
inevitable Increasing pressure to justify choices
in expenditure

• Costs health services are rising as GDP due to
• new expensive technologies
• ageing
• greater demands from consumers

3
Evidence for health policy

• size and distribution of health problems burden
  of disease projections
• costs and impact of health interventions disease
  costing cost-effectiveness

Essential is a universal method to measure health
outcomes
4
Burden of disease
5
DALY

• Disability-adjusted Life Year
• future stream of life lost due to premature
  mortality based on life expectancy
• Years of Life Lost (YLL)
• future loss of 'healthy' life arising from new
  cases of disabling conditions
• Years Lived with Disability (YLD)

6
DALYs by grouped conditions, Australia 2003
Other
Neonatal conditions
Congenital anomalies
Infections
Mortality
Digestive disorders
Disability
Genitourinary disease
Musculoskeletal disease
Diabetes
Chronic respiratory diseases
Injuries
Neurological and sense disorders
Mental disorders
Cardiovascular diseases
Cancer
0
5
10
15
20
of total DALYs
Begg SJ, Vos T, Barker B, Stanley L, Lopez AD
(2008). The burden of disease and injury in
Australia in the new millennium measuring health
loss from diseases, injuries and risk factors.
Medical Journal of Australia, 1883640.
7
Life expectancy and years lost due to disability
by jurisdiction, remoteness and socioeconomic
status, Australia 2003
State/Territory
SES quintile
Remoteness
High
ACT
82
82
82
WA
Mod high
Major cities
Vic
Qld
SA
NSW
Average
Mod low
Regional
80
80
80
Low
Tas
Life expectancy at birth (years)
Remote
78
78
78
76
76
76
NT
8
9
10
11
8
9
10
11
8
9
10
11
Proportion of life expectancy at birth lost due
to disability ()
Begg S, Vos T, Barker B, Stanley L, Lopez AD
(2007). The burden of disease and injury in
Australia 2003. PHE 82. Canberra AIHW.
8
Life expectancy and years lost due to disability,
Indigenous and total Australian population, 2003
Total Australian population
80
75
Life expectancy at birth (years)
Non-remote Indigenous population
70
65
Remote Indigenous population
60
8
9
10
11
12
13
Proportion of life expectancy at birth lost due
to disability ()
Vos T, Barker B, Stanley L, Lopez AD (2007). The
burden of disease and injury in Aboriginal and
Torres Strait Islander people 2003. Brisbane
University of Queensland.
9
DALY rates for leading broad cause groups,
Indigenous and total Australian population 2003
Age standardised to the total Indigenous
Australian population, 2003
Vos T, Barker B, Stanley L, Lopez AD (2007). The
burden of disease and injury in Aboriginal and
Torres Strait Islander people 2003. Brisbane
University of Queensland.
10
Age-standardised death rates for major causes of
death, Australia, 19222000
Begg S, Vos T, Barker B, Stanley L, Lopez AD
(2007). The burden of disease and injury in
Australia 2003. PHE 82. Canberra AIHW.
11
Change in mortality and disability 19932023
Mortality
Disability
400
300
300
200
200
Rate per 1,000
100
100
0
0
60
70
80
90
100
60
70
80
90
100
Age
1993
2023
Begg S, Vos T, Barker B, Stanley L, Lopez AD
(2007). The burden of disease and injury in
Australia 2003. PHE 82. Canberra AIHW.
12
Change in disease burden 19932023
Begg S, Vos T, Barker B, Stanley L, Lopez AD
(2007). The burden of disease and injury in
Australia 2003. PHE 82. Canberra AIHW.
13
Conclusions

• Major reductions in mortality and disability
  from
• Tobacco-related disease
• Cardiovascular disease
• Injuries
• Much less change in chronic disabling conditions
• Increase in diabetes as consequence of steady
  weight gain
• Increased life span coupled with healthier life
  lt80 downside gt80 accumulation of disability

14
Projections of health expenditure
15
Australian case study

• Report for UN flagship publication World
  Economic Survey 2007 with theme on ageing
• We have good disease projections and health
  expenditure by same disease categories
• Only Netherlands has been able to do same
  disease-by-disease health expenditure projections
• Does this make difference?

16
Demographic changes 1925-2045
17
Per capita health expenditure by age and sex,
Australia, 200001
Data from AIHW (2005) Health system expenditure
on disease and injury in Australia, 2000-01. AIHW
Cat No HWE 28
18
Non-demographic drivers of change in health
expenditure

• Technological change diagnostics, drugs,
  procedures
• Changing medical practice and policy
• Organisation and financing of health system
• Intensity or coverage of health services
• Excess health inflation
• Changes in population health status

19
Health expenditure, Australia 2003-2033 as of
projected GDP
GDP
12.0
10.8
9.9
10.1
9.4
10.0
9.2
9.0
7.9
9.9
8.0
6.8
5.7
7.9
6.0
6.2
5.1
4.0
4.7
2.0
0.0
2003
2013
2023
2033
2043
SPH/AIHW (all health expenditure, including
private/out-of-pocket
Inter-Generational Report Federal Govt
expenditure only
PC (State and Federal Govt expenditure)
Begg S, Vos T, Goss J, Mann N (2008). An
alternative approach to projecting health
expenditure in Australia. Aust Health Rev,
32148155.
20
Drivers of growth in health expenditure,
Australia 2003-2033
million
100,000
0
0
Change in total expenditure
80,000
Price
60,000
0
Treatment proportion
40,000
Volume per case
0
Disease rate
20,000
Population
0
Ageing
-20,000
2012-13
2022-23
2032-33
Begg S, Vos T, Goss J, Mann N (2008). An
alternative approach to projecting health
expenditure in Australia. Aust Health Rev,
32148155.
21
Growth in health expenditure, by selected disease
groups, Australia 2003-2033
Begg S, Vos T, Goss J, Mann N (2008). An
alternative approach to projecting health
expenditure in Australia. Aust Health Rev,
32148155.
22
Projected growth in health expenditure, by type,
Australia 2003-2033
Cardiovascular 7,905 16,178 105 99
133 Neurological 3,982 15,126
280 235 294 Respiratory 5,925
12,621 113 108 279 Dental
5,097 12,436 144 144 Musculoskeleta
l 3,743 9,864 164 144
259 Injuries 5,592 9,359 67
65 153 Mental 4,304 8,483
97 89 167 Diabetes 1,392
6,971 401 399
431 Genitourinary 3,060 6,798 122
121 223 Cancer 2,808 5,167
84 83 194 Maternal neonatal
2,305 3,252 41 41 Other
21,220 46,401 119 114 Total
71,376 162,319 127 114 242
Begg S, Vos T, Goss J, Mann N (2008). An
alternative approach to projecting health
expenditure in Australia. Aust Health Rev,
32148155.
23
Conclusions

• Impact of changes in disease epidemiology has
  only small impact on overall projected health
  expenditure estimates (favourable trends CVD,
  tobacco-related disease, injuries partly
  compensated by increase in diabetes)
• . but our disease-specific estimates are much
  less alarming than Treasury/Productivity
  Commission projections
• Ageing, population change, increased volume per
  case and health inflation main drivers of growth
• Disease specific estimates indicate big shifts in
  type of services required

24
Risk factors and opportunities for prevention
25
Disease burden attributable to major risk
factors, Australia 2003
Females
Males
2.3
intimate partner violence
0.1
osteoporosis
0.3
1.5
child sexual abuse
0.3
unsafe sex
0.7
0.5
air pollution
0.8
0.7
occupational exposures
2.6
1.3
illicit drugs
2.7
1.2
2.7
1.5
low intake fruit veg
alcohol
3.8
0.7
high cholesterol
6.6
5.8
physical inactivity
6.4
6.8
7.7
high blood pressure
7.3
high body mass
7.8
7.3
tobacco
9.6
5.8
of total burden
26
Risk factors

• Proportion of disease groups explained by
  combination of 14 risk factors
• Cardiovascular disease 69
• IHD 85
• Stroke 70
• Cancer 33
• Mental disorders 27
• Neurological disorders lt1
• Injuries 32

27
Prevention of cardiovascular disease and diabetes

• Major risk factors and interventions
• Tobacco
• lots of cost-effective interventions but still
  2.9 million smokers
• continued expansion of tobacco control measures
• need for new quit interventions vaccine?, snus?,
  ...?
• Blood pressure cholesterol
• Under-utilisation of cost-effective interventions
• Inefficient use of current resources
• Body mass, diet physical activity
• Weak evidence for (cost-)effectiveness of
  interventions
• No change or worsening over time ? not
  contributed to decline in CVD
• Massive increase in incidence of diabetes
  predicted while case fatality drops ? even bigger
  increase in prevalence

28
Optimal mix of blood pressure and
cholesterol-lowering interventions in primary
prevention
35,000
30,000
25,000
20,000
Lifetime Costs (million AUS)
15,000
Current practice
Statin
10,000
Ezetimibe
ß-blocker
Dietitian
5,000
Diuretic Aspirin
Phytosterol
0
0
100
200
300
400
500
600
700
community heart health program
Lifetime DALYs averted ('000)
29
Reasons for inefficiency of current practice

• Single risk factor thresholds (i.e. high vs
  normal blood pressure) vs combined absolute
  risk
• Not enough attention to lifestyle and public
  health interventions
• CHHP
• Dietary counselling
• Phytosterol supplementation
• Current resources directed at less efficient
  classes of BP lowering drugs
• ACE inhibitors

30
Optimal mix of physical activity interventions
Life time DALYs averted
0
10,000
20,000
30,000
40,000
50,000
60,000
70,000
0
-200
Pedometers
GP referral
-400
Life time costs A Millions
TravelSmart
Mass media
GP prescription
-600
Internet
-800
Including time travel costs
Excluding time travel costs
-1,000
31
Comments on physical activity interventions

• weak evidence
• Pedometers tested in volunteers
• Different units of measurement of effect
• Varies from meta-analyses to single studies with
  before-after design
• but 5 of the 6 interventions could address up to
  a third of burden from physical inactivity

32
Prevention of cancer

• Most change over time from tobacco interventions
• Smaller risk factors alcohol, air pollution, sun
  exposure, HPV
• Screening for breast, cervix, bowel, prostate
  modest health gain, relatively high cost
• A lot of unexplained cancer
• Treatments have at best modest impact on course
  of most cancers

33
Prevention of mental disorders

• Substance use
• Some reduction in alcohol dependence but not
  binge drinking
• Plausible evidence for link heavy cannabis use
  schizophrenia
• Up to a third of alcohol burden can be reduced
  with a package of six preventive measures
• Mental disorders
• Limited knowledge about prevention of major
  mental disorders but some ideas being tested for
  prevention of depression and childhood disorders
• No evidence for interventions to reduce exposure
  to child sexual abuse or intimate partner violence

34
Optimal mix of alcohol interventions
35
Prevention of neurological disorders

• Neurological disorders
• Some prospects of treatment/prevention with
  drugs/vaccine for dementia and Parkinsons but
  current ability to prevent is very limited

36
Trends in road accident mortality, Australia
1950-2003
Seat belts
Seat belts
RBT
RBT
37
Trends in homicide violence mortality,
Australia 1950-2003
Port Arthur massacre
Port Arthur massacre
38
Conclusions

• Burden of disease (how big are health problems,
  who gets it, and are things changing?) and
  cost-effectiveness (how can we achieve most
  health gain given a certain budget) form a
  powerful set of tools to inform resource
  allocation decision making in health
• Requires intensive analytical effort but tools to
  do so are improving
• Most immediate additional health gain from past
  winners tobacco control, CVD prevention, injury
  control
• Large disease/risk factor areas not well
  addressed by prevention obesity, physical
  activity, neurological conditions, mental
  disorders, hearing loss, diabetes, asthma ?
  research priorities