The European Food Safety Authority is developing practical guidance for its expert risk assessors on how to identify, characterise, document and explain all types of uncertainty arising within an individual assessment for all areas of EFSA’s remit. The draft Guidance does not prescribe which methods should be used but provides a harmonised and flexible framework within which different qualitative and quantitative methods may be selected from a toolbox of described methods according to the needs of each assessment. The intention is that each scientific assessment should say clearly and unambiguously what sources of uncertainty have been identified and what is their combined impact on the final outcome; what range of outcomes is possible, and how probable they are.
The steps of the general analytical framework include planning the assessment strategy (problem formulation), identifying the sources of uncertainty, evaluating individual sources of uncertainty, evaluating the combined uncertainty, investigating influence and refining the assessment (if needed) and reporting the uncertainty analysis.
Expert judgement plays a key role in uncertainty analysis, as in other aspects of scientific assessment. Assessors should be systematic in identifying sources of uncertainty, checking each part of their assessment to minimise the risk of overlooking important uncertainties. Uncertainty may be expressed qualitatively or quantitatively. It is not necessary or possible to quantify separately every individual source of uncertainty affecting an assessment. However, assessors should express in quantitative terms the combined effect of as many as possible of the identified sources of uncertainty. Practical approaches to facilitate this are described.
Uncertainty analysis should be conducted in a flexible, iterative manner, starting at a level appropriate to the assessment in hand and then refining the analysis as far as is needed or possible within the time available. Some steps may be reduced or omitted in emergency situations and in routine assessments with standardised provision for uncertainty. Sensitivity analysis and other methods for investigating influence are used to target refinement on those sources of uncertainty where it will contribute most.
The draft guidance document was published for public consultation and will be finally revised after the end of a 12 month trial period during which EFSA’s Panels and Units are applying the guidance to examples of scientific assessments in their sector of food and feed safety, animal and plant health.