Table 2 |
||||||
|
Illustrations of characterizations of uncertainties in environmental burden of disease assessments |
||||||
|
Source of uncertainty |
Nature Epistemic/Ontic (Process Variability/Normative Uncertainty) |
Range Statistical/Scenario |
Recognized ignorance |
Methodological unreliability |
Value diversity among analysts |
|
|
|
||||||
|
CONTEXTUAL UNCERTAINTY |
||||||
|
|
||||||
|
1 |
Multiple ways of defining the 'total environment' |
E/Nor |
Sc |
- |
+ |
++ |
|
|
||||||
|
2 |
Only including diseases that cause at least 1% of the global burden of disease |
Nor |
Sc |
-- |
-- |
+ |
|
|
||||||
|
MODEL STRUCTURE UNCERTAINTY |
||||||
|
|
||||||
|
3 |
Specific form of the exposure-response relationship is unknown |
E |
Sc |
+ |
+ |
+ |
|
|
||||||
|
4 |
Evidence for causality (environmental factor leading to health effect) is weak and contradicting |
E |
Sc |
++ |
++ |
+ |
|
|
||||||
|
5 |
Incomplete understanding of the joint effect of smoking and radon in relation to lung cancer |
E |
Sc |
+ |
+ |
+ |
|
|
||||||
|
6 |
Accounting for susceptible groups if the available relative risk is not representative for this group |
Pro/E |
St |
+ |
+ |
+ |
|
|
||||||
|
PARAMETER UNCERTAINTY |
||||||
|
|
||||||
|
7 |
Determining a relative risk (RR) for long-term exposure to PM10 |
E |
St |
+ |
+ |
- |
|
|
||||||
|
8 |
Applying an American RR for PM10 to the Netherlands |
E |
Sc |
++ |
+ |
+ |
|
|
||||||
|
9 |
Use of severity weights |
Nor/E |
Sc |
+ |
+ |
++ |
|
|
||||||
|
INPUT DATA UNCERTAINTY |
||||||
|
|
||||||
|
10 |
Extrapolating non-assessment-specific exposure measurements |
E |
Sc |
++ |
+ |
+ |
|
|
||||||
|
11 |
Measuring population exposure |
E |
St |
+ |
+ |
- |
|
|
||||||
|
Knol et al. Environmental Health 2009 8:21 doi:10.1186/1476-069X-8-21 |
||||||
