Ingen diastitel

1
A world in transition Dan
Meyrowitsch Dept. of Epidemiology Institute of
Public Health University of Copenhagen, Denmark
FSV February 3, 2004
2
A world in transition

• Objectives of session
• To present and discuss the concept of summary
  measures of population health with specific focus
  on disability-adjusted life years (DALY)
• To present and discuss the theory on
  epidemiological transition (Omran, 1971)
• To give a general introduction to the global
  disease burden

3
Traditional epidemiological measurs for disease
burden

• Life expectancy (e.g. LE at birth, healthy LE)
• Mortality (e.g. enfant mortality, disease
  specific mortality)
• Morbidity (e.g. disease specific
  prevalence/incidence/number)
• Severity of disease (biological, self-reported)
• Disease impact on activity (ADL)
• Health related life quality assessment (e.g.
  SF-26, EuroQol)
• A need for one indicator which combines measures
  of mortality and morbidity

4
Summary measures of population health

• Definition
• Measures that combine information on mortality
  and non-fatal health outcomes to represent the
  health of a particular population as a single
  number

5
Use of summary health measures

• Comparison of populations
• Measure changes in population health over time
• Measure inequities in health between populations
• Assess the impact of disease among live
  individuals on overall health in population
• Allow discussion on prioritisation regarding
  health services
• Allow discussion on prioritisation regarding
  future health R D
• Improve curriculum in public health education
• Assess effects/benefits of interventions

(from Murray, Salomon Mathers, 2000)
6
Two types of summary measures
Health expectancies (life time expectations) DFLE,
disability-free life expectancy (lost years with
good health) HLE, life expectancy in good health
(definition of good health?) QALE,
quality-adjusted life expectancy ALE, active
life expectancy DALE, disability-adjusted life
expectancy Health gaps (measured against a
theoretical total-population-healthy
scenario) YLL, years of life lost YLD, years of
life lost to disability QALY, quality-adjusted
life years HALY, health-adjusted life years
DALY, disability-adjusted life years
7
Healthy, sick or dead
(Murray Lopez, The global burden of disease,
1996)
8
(Iburg, 2002, Danmark Statistik 1998-99)
9
Disability-adjusted life years (DALY)
DALY combines two health gap measures Years of
life lost due to premature death (YLL) Years of
life lost to disability (YLD) DALY YLL YLD,
One DALY is one year lost of a healthy life YLL
No. lost life years measured against a decided
upper norm In Global Burden of Disease Survey
Japan 80.0 years for men 82.5 years for
women YLD Years lost while living with
disease For each disease (incidence) x
(severity) x (duration) Algorithms include for x
(age weights) x (discounting) YLD estimates
based on registers, surveys, studies and expert
panels
10
(Murray Lopez, The global burden of disease,
1996)
11
Person 1Fatal car-accident at age 45
Disability
35 YLL
0.5
1
20
10
30
40
50
60
70
80
90
45
Age
12
Person 2Non-fatal car-accident at 50
0
300.4 12 YLD
Disability
0.4
0.5
1
50
80
Age
13
Person 3 Non-fatal car-accident at 20
premature death at 60
0
400.4 16 YLD
Disability
0.4
20 YLL
0.5
1
20
60
80
Age
14
Person 1 2 3
35 YLL (1) 12 YLD (2) 16 YLD (3)
20 YLL (3) 83 DALYs
15
Years of life lost years of life lost to
disability
(Murray Lopez, The global burden of disease,
1996)
16
Death years of life lost
(Murray Lopez, The global burden of disease,
1996)
17
Data needed for construction of summary measures

• Death (specific regarding age, sex and geo area)
• Epidemiology of major diseases (e.g. incidence,
  prevalence, duration)
• Life expectancy norm for health gaps
• Pathological history of diseases (e.g. Qol
  instruments, HUI, EuroQol, SF-36)
• Severity weights (e.g. VAS, SG, TTO, PTO)
• Other values for adjustment (discounting, age
  weights)

18
Age weights
Lost life years are perceived as having different
relative values in different age groups Example
2-year vs 22-year with meningitis The
importance of good health are perceived as
different in different age groups Each
individual represent a potential for future
production. Older individuals regarded as having
a lower potential
(Murray Lopez, The global burden of disease,
1996)
19
Discounting
Future healthy life years are regarded as being
of lower value than health here and now - i.e.
most individuals would like good health this year
instead of good health next year Same principle
as used in economical interest rates In
GBD-survey 3 discount rate
Men pappa, du pratar liksom om 25 år. Alltså,
tyvärr, men jag är hellre glad nu än om 25 år!
(Agnes in the movieFucking Åmål, 1998)
With age-weighting included, this furthermore
reduces the years lost when a child dies compared
to the death of 25 years old
20
Limitations (I) Who knows the real burden

• MDs, public health experts
• Individuals with disease
• Friends or relatives of individual with disease
• ?

21
Limitations (II)Proportion of ill-defined causes
of death
(Global burden of disease, 2000)
22
Limitations (III)
Limited data on incidence, prevalence and
duration of numerous illnesses . and existing
data are of low quality Low quality of
diagnostic methods, problems in recording and
reporting data
23
Limitations (IV)

• No adjustments/weights for
• Gender specific differences in impact of
  disease/death
• Socio-economic differences (rich/poor, ethnicity
  etc.)

24
Summary on summary measures Advantages/disadvantag
es

• Positive
• Evidence based, objective information
• Rank the relative importance of health problems
• Tool in planning, political priority making
  economical resource allocation
• Negative
• Not suited to monitor effect of
  disease-specific intervention (vertical approach)
• Large variation in culture-specific perceived
  severity of disease
• The socio-related values used in weighting (age
  and discounting) used in DALY calculations are
  ethically problematic
• The epidemiological inputs comprise the weakest
  part in construction of summary measures (of
  specific importance in low-income countries)

25
Summary on DALY
DALY combines two health gap measures Years of
life lost (YLL) Years of life lost to disability
(YLD) For each disease YLD is adjusted re
(incidence) x (severity) x (duration) Algorithms
include x (age weights) x (discounting) DALY
YLL YLD
26
Weight of same disease measured with different
methods
(WHO Bull 2002, The European Disability Weight
Project)
1
VAS TTO PTO
0.9
0.8
0.7
0.6
Disability weigts
0.5
0.4
VD
BC
LBP
DM
AST
AP
STK
DEP
QP
Disease stages