Pages within the "support" section:
1.2 Relationship to the BRAFO tiered approach
The related EU project BRAFO has developed a tiered approach for risk-benefit assessment (Hoekstra et al., submitted). This comprises 4 Tiers, which differ principally in the way risks and benefits are integrated. At Tier1, risks and benefits are assessed separately, while in Tiers 2-4 they are integrated using increasingly sophisticated approaches. Consequently in each tier the net health impact is assessed with increasing precision. As with all tiered assessment approaches, the aim is to refine the assessment only as far as is necessary to reach a decision, in this case on whether the net health impact of a dietary change is beneficial or adverse. The 4 Tiers are as follows:
· In Tier 1, each risk and benefit is assessed independently. These assessments will often use standard screening methods, but it may be worth using more refined methods if this avoids the need to proceed to Tier 2. Thus Tier 1 comprises separate assessments of risks and benefits, each as refined as may be needed.
· In Tier 2, risks and benefits are compared in a qualitative way; no common metric is used at this tier. However, the assessment of each individual risk or benefit may be quantitative or even probabilistic.
· In Tier 3, risks and benefits are integrated quantitatively in a common metric, using deterministic methods.
· In Tier 4, risks and benefits are integrated quantitatively in a common metric, using probabilistic methods.
In practice, there is a continuum between tiers 3 and 4. Initially all parts of the assessment are treated deterministically (i.e. as fixed values), after which progressively more parameters are treated probabilistically (i.e. using probability distributions), until the net health impact is sufficiently well characterised for decision-making.
The Qalibra project focussed on developing detailed methodology for the quantitative integration of risks and benefits, i.e. Tiers 3 and 4 of the BRAFO approach. Consistent with the BRAFO approach, the Qalibra framework and software tool allows progressive refinement from Tier 3 to Tier 4, as each input may be treated deterministically or probabilistically. If one or more of the inputs is probabilistic, then the output will also be probabilistic.
Quantitative integration of risk and benefit is technically complex and requires additional types of data or assumptions (e.g. severity of effects and age of onset) that are not required in conventional assessments of risk and benefit. Therefore, Qalibra recommends that practitioners follow the BRAFO tiered approach, resolving risk-benefit questions without quantitative integration where possible, and reserve the Qalibra framework for those cases where assessment at Tiers 3 and 4 is necessary.