Table B.20.

Assessment of Confidence intervals (when well applied) against evaluation criteria

Criteria	Evidence of current acceptance	Expertise needed to conduct	Time needed	Theoretical basis	Degree/extent of subjectivity	Method of propagation	Treatment of uncertainty and variability	Meaning of output	Transparency and reproducibility	Ease of understanding for non‐specialist
Stronger characteristics	International guidelines or standard scientific method	No specialist knowledge required	Hours	Well established, coherent basis for all aspects	Judgement used only to choose method of analysis	Calculation based on appropriate theory	Different types of uncertainty & variability. quantified separately	Range and probability of possible answers	All aspects of process and reasoning fully documented	All aspects fully understandable
	EU level guidelines or widespread in practice	Can be used with guidelines or literature	Days	Most but not all aspects supported by theory	Combination of data and expert judgement	Formal expert judgement	Uncertainty and variability quantified separately	Range and relative possibility of answers	Most aspects of process and reasoning well documented	Outputs and most of process understandable
	National guidelines, or well established in practice or literature	Training course needed	Weeks	Some aspects supported by theory	Expert judgement on defined quantitative scales	Informal expert judgement	Uncertainty and variability distinguished qualitatively	Range of answers but no weighting	Process well documented but limited explanation of reasoning	Outputs and principles of process understandable
	Some publications and/or regulatory practice	Substantial expertise or experience needed	A few months	Limited theoretical basis	Expert judgement on defined ordinal scales	Calculation or matrices without theoretical basis		Quantitative measure of degree of uncertainty	Limited explanation of process and/or basis for conclusions	Outputs understandable but not process
Weaker characteristics	Newly developed	Professional statistician needed	Many months	Pragmatic approach without theoretical basis	Verbal description, no defined scale	No propagation	No distinction between variability and uncertainty	Ordinal scale or narrative description for degree of uncertainty	No explanation of process or basis for conclusions	Process and outputs only understandable for specialists