. 2020 Apr 29;18(4):e06091. doi: 10.2903/j.efsa.2020.6091

Table D.1.

Sources of uncertainty identified in the hazard identification and qualitative assessment of the impact that these uncertainties could have on the conclusions

Source or location of the uncertainty	Nature or cause of the uncertainty	Impact of the uncertainty on the conclusions (e.g. over/underestimation)
Hazard identification (literature search and screening)	Evidence on (1) hazards relevance for human illness associated with FFP or on (2) survival (no change or reduction) or growth of the hazard in FFP within the range −3°C to 7°C may have been missed or may be absent due to: underreporting, cases or outbreaks not being published or not being associated with FFP as a causative vehicle; and unavailability of experimental data for the combination hazard/FFP within the relevant temperature range.	Potential underestimation of the relevance of a hazard. However, given the use of extensive literature reviews, uncertainty associated with missing evidence in this step has been reduced as much as possible In particular, in consideration of the search extent, exclusion of a relevant hazard due to protracted non‐reporting of human disease was considered unlikely For hazards excluded from the assessment due to absence of evidence of growth or histamine production within the range −3°C to 7°C, other available information based on microbial ecology principles was considered as a scientific support for exclusion
Hazard identification (data extraction)	Uncertainty in the description of the growth/survival behaviour of the hazards due to inter‐strain variability, heterogeneous distribution of the microorganisms on the FFP and variability of the performance of the enumeration methods A systematic appraisal analysis for quality of the studies, particularly with regard to survival or growth of the selected hazards at temperatures relevant for the assessment, was not performed	Potential over‐ or underestimation of the relevance of a hazard The adoption of a criterion for the assessment of relevant growth/survival (threshold of 0.5 log₁₀ in the increase/decrease of microorganisms) limited the uncertainty in the interpretation of the hazards’ behaviour To avoid overestimation of growth/survival, particularly of mesophilic microorganisms, critical appraisal and exclusion of results was applied for publications in which inconsistencies in the experimental design or the analytical methods were detected

Source or location of the uncertainty

Nature or cause of the uncertainty

Impact of the uncertainty on the conclusions (e.g. over/underestimation)

Hazard identification (literature search and screening)

Evidence on (1) hazards relevance for human illness associated with FFP or on (2) survival (no change or reduction) or growth of the hazard in FFP within the range −3°C to 7°C may have been missed or may be absent due to:

underreporting, cases or outbreaks not being published or not being associated with FFP as a causative vehicle; and
unavailability of experimental data for the combination hazard/FFP within the relevant temperature range.

Potential underestimation of the relevance of a hazard. However, given the use of extensive literature reviews, uncertainty associated with missing evidence in this step has been reduced as much as possible

In particular, in consideration of the search extent, exclusion of a relevant hazard due to protracted non‐reporting of human disease was considered unlikely

For hazards excluded from the assessment due to absence of evidence of growth or histamine production within the range −3°C to 7°C, other available information based on microbial ecology principles was considered as a scientific support for exclusion

Hazard identification (data extraction)

Uncertainty in the description of the growth/survival behaviour of the hazards due to inter‐strain variability, heterogeneous distribution of the microorganisms on the FFP and variability of the performance of the enumeration methods

A systematic appraisal analysis for quality of the studies, particularly with regard to survival or growth of the selected hazards at temperatures relevant for the assessment, was not performed

Potential over‐ or underestimation of the relevance of a hazard

The adoption of a criterion for the assessment of relevant growth/survival (threshold of 0.5 log₁₀ in the increase/decrease of microorganisms) limited the uncertainty in the interpretation of the hazards’ behaviour

To avoid overestimation of growth/survival, particularly of mesophilic microorganisms, critical appraisal and exclusion of results was applied for publications in which inconsistencies in the experimental design or the analytical methods were detected

FFP: fresh fishery products.