Using DALYs for estimating environmental burden of disease in the Netherlands

1
Using DALYs for estimating environmental burden
of disease in the Netherlands

• RIVM National Institute of Public Health the
  Environment
• IRAS Institute of Risk Assessment Sciences, UU
• Erik Lebret

2
The aim of the RIVM is to improve public health
and to safeguard a healthy environment.

• This is achieved by
• Research
• Policy support
• National coordination functions
• Specific intervention programs
• Providing targeted information on health and
  environment

3
political and research responsibilities
4
National Vaccin Institute
5
Content

• Why do burden of disease estimates?
• What is it
• How to do burden of disease estimation
• Some examples

6
A small poll what causes the highest disease
burden in modern society?

• Soil pollution
• Dioxins and PCBs in food
• Carcinogens in air
• Ambient air toxics (non-car.)
• Indoor air
• Radiation
• Non-ionising radiation
• Noise

7
What is the impact of environmental pollution on
disease burden?

• 0
• 1- 5
• 6 - 10
• 10 - 20
• gt 20

8
Why How bad is it?

• To summarise
• Multiple sources,
• Multiple pollutants
• Multiple health endpoints
• To prioritise environmental health problems
• To optimise policy solutions

9
What is health?

• a state of complete physical, mental and social
  well-being, and not merely the absence of disease
  or infirmity (WHO charter, 1946)
• that what the doctor can see
• the ability to cope with the demands of daily
  life (the Dunning Committee on Medical Cure and
  Care, 1991)

10
Survival curves through the ages(adapted from
Wills, 1996, Ruwaard Kramers, 1998)
11
Survival health curves NL
12
Definition of public health

• Public health is the science and art of
• preventing disease, prolonging life and
• promoting health through the organised
• efforts of society (Acheson, 1988)

13
Dimensions of public health

• quantity of life
• quality of life
• number of people affected
• time is unit of measurement

Global Burden of Disease Project (Murray Lopez,
1993) DALYsYLLYLD
14
The concept of DALYs
14
15
Estimation of environmental DALYs

• estimate number of people affected
• estimate average duration of the response
• attribute severity weight to responses
• calculate annual public health loss
• uncertainty analysis (Monte Carlo)

16
GBD-study
17
Presuppositions

• Prognosis must not be taken into the weighing
• Adaptation must not be included
• Medical standard treatment
• Weighing independent of duration

17
18
EuroQol-5D
19
Results weighing
20
You are here!
21
What is special in e-BoD

• Exposure oriented (exception clusters)
• Ubiquitous exposures
• Weak associations in the presence of strong
  confounding
• Strong regulatory context
• Strong societal interest and pressures, if not
  public outcry

22
Pyramid model (ATS, 1989)
23
Two approaches to estimate DALYs

• From health data, estimate the share of health
  determinants responsible for disease burden
• Take disease-specific health statistics and have
  experts estimate fractions for different
  determinants
• From exposure to determinants, estimate the
  attributable number of people affected

24
(No Transcript)
25
Epidemiological transition.
26
Finland in WHO Environmental BoD
27
Finland in WHO Environmental BoD
28
Number of people affected

• Population exposure distribution (p)
• Defining health outcomes
• Quantifying dose response relations

Population Attributive Fraction PAF
p(RR-1)/p(RR-1)1
Mortality/morbidity data (B)
Number of people
N B x PAF
29
What input data for exposure?

• Exposure characterisation in indicators and
  proxies that have the same dimensions as the
  exposure-response function
• Dichotomous (no/yes)
• Distance to road, traffic intensity
• Ambient concentration
• Personal exposure
• Dose
• ?Intake fraction?
• Reference or counterfactual exposure scenario

30
What data for exposure-response function

• Data from meta-analyses
• Data from systematic reviews
• Data from expert judgements
• But, how to
• Deal with mixtures and combined exposures
• Local data
• Combine epidemiological and toxicological data

31
What else?

• Background incidence and prevalences of relevant
  health endpoints
• Estimates of the duration of the effect
• mortality displacement
• duration of non-lethal effects (typically not
  studied in environment epi. or tox. studies)

32
National Environmental Outlook 1999Environmental
Disease burden in the Netherlands
33
Some trends
34
Trends in the EDB in the Netherlands 1980-2010
35
Policy intervention
36
Deaths, DALYs, and Dollars? 2003
37
DALYs vs Euros not very consistent!
38
Who wants to?

• Ministry of Health
• Ministry of Environment
• Ministry of Transport
• Food and Consumer Safety Authority
• Advisory Council Dangerous Substances
• Local health authorities
• Ministry of the Interior

Me too!!
39
Indicators used by different stakeholders
40
Risk perception

• Risk perception involves people's beliefs,
  attitudes, judgements and feelings as well as
• the wider social or cultural values and
  dispositions that people adopt towards hazards
  and their benefits.
• Pidgeon, et al

41
Risk perception and health

• Is (negative) risk perception part of (ill)
  health?
• Is (negative) risk perception a determinant of
  (ill) health?
• Is health, as an un-negotiable value, a
  bargaining chip for people with negative
  perception of the quality of their living
  environment?
• Does negative risk perception reduce quality of
  life in a similar way as ill health/diseases?

Probably all of the above
42
Risk perception is a psychosocial factor and as
such may affect physical health

• Review ..prospective cohort studies provide
  strong evidence that pychosocial factors. are
  independent aetiological and prognostic factors
  for coronary heart disease (Hemingway Marmot,
  BMJ 1999)
• Visual exposures influence recovery from surgery
  (Ulrich)
• Review Chronic psychological stress affects
  immune system and vaccination response (van
  Loveren et al. RIVM, 2000)

43
Risk aspects

• This risk is considered dreadful by society
• This risk is unknown to science
• Exposure to this risk is involuntary
• This risk is uncontrollable by experts
• The negative effects are unequally distributed
  across society
• This is a new risk, without history
• Negative effects are likely to occur
• The effects of this risk are immediate
• This risk affects a large number of people
• This risk is associated with severe effects

44
Perception of risk attributes of acceptability
45
(No Transcript)
46
Attribution of clusters to environmental factors
(van Poll Drijver99)
47
Is there a problem with EMF from power lines? If
so, how bad is it?

• There is no conclusive evidence for a causal
  relation between EMF and disease at levels in
  the environment
• Given the uncertainty NIMBY
• EU reference level of 100 microtesla not exceeded
  in NL
• In NL 23,000 houses above 0.4 microtesla
• Maximum Tolerable Risk (10-6) exceeded in NL
• In NL 0.5 extra childhood leukaemia cases, 0.15
  0.25
• In NL there is double the leukaemia risk for
  children within 0.4 microtesla area
• In NL 10-30 DALYs
• Monetary burden 0.15, 2.5 and 15 bill. , resp.
  or, 18, 126, 665 k per avoided house (depending
  on efficacy)
• Buy up all electric blankets to acquire emission
  rights for power lines

48
What ethical decision framework?
Virtual-certainty-required

• There is no evidence for a causal relation
  between EMF and disease
• Given the uncertainty NIMBY
• EU reference level of 100 microtesla not exceeded
  in NL
• In NL 23,000 houses above 0.4 microtesla
• Maximum Tolerable Risk (10-6) exceeded in NL
• In NL 0.5 extra childhood leukaemia cases, 0.15
  0.25
• In NL there is double the leukaemia risk for
  children within 0.4 microtesla area
• In NL 10-30 DALYs
• Monetary burden 0.15, 2.5 and 15 bill. , resp.
  or, 18, 126, 665 k per avoided house (depending
  on efficacy)
• Buy up all electric blankets to acquire emission
  rights for power lines

Precautionary principle
Rights-based
Rights-based
Public health/utilitarian
Equity/social justice
Utilitarian
49
Minister of Environment
We will launch a 16 M research programme on
electromagenetic fields.
., but we dont think that there are any risks
involved with EMF.
50
NRC Understanding risk 1996

• Getting the science right
• Getting the right science
• Getting the right participation
• Getting the participation right
• Developing accurate, balanced, and informative
  synthesis

IRGC Risk Governance Framework
51
Output multiple panels (e.g. in
dashboard/cockpit) that summarise info

• Policy panel
• Distance to stated policy targets
• Policy deficit
• Impact panel
• Health
• Listing of effects (number and nature of effects)
• Disease burden (e.g. DALY)
• Economy
• Full cost-benefit analyses
• Impact pathway methodology
• Public acceptability panel
• Psychometric paradigm aspects
• Environmental Equity
• Appraisal panel
• (0verview table)
• Multi-criteria analyses
• CBA

52
Framework
53
Extended Full Chain Approach
54
Protocols Guidelines step-by-step
55
(No Transcript)
56
So the source-dose relation summarised as Intake
Fraction links to a Bayesian derived ERF to
estimate DALYs and NUSAP approach-based
uncertainties..
Whats a Bayesian ERF?
Whats a DALY or NUSAP?
Whats an Intake Fraction?
57
In conclusion

• E-BoD is a good indicator to summarise the
  variety of health effects, e.g. to inform the
  policy debate
• E-BoD is only part of the equation
• E-BoD is hard to communicate
• E-BoD indicators are synthetical, cannot be
  directly measured for validation, e.g. like
  inflation

58
Where are we now?
59
Where are we now?
60
Is the DALY as single metric the Holy Grail?

• It depends on your ethical decision framework
• It will work for people with a utalitarian
  worldview
• It will not work sufficiently for people with
  worldviews like
• Libertarians
• Hierarchists
• Egalitatians
• Fatalists