The process for the assessment of scientific support for claims on food.

The concerted action "The process for the assessment of the scientific support for claims on foods", PASSCLAIM, proposed criteria that could provide an international yardstick for the harmonised transparent assessment of evidence submitted to support a claim for a food or food component. The evidence would be systematically appraised against specific criteria: namely, (1) a characterisation of the food or food component to which the claimed effect is attributed; (2) human data, primarily from intervention studies that represent the target populations for the claim; (3) a dose response relationship; (4) evidence allowing for confounders such as lifestyle, consumption patterns, background diet and food matrix etc.; (5) an appropriate duration for the study; (6) a measure of compliance; (7) adequate statistical power to test the hypothesis. Validated and quality assured markers of intermediate or final outcomes could be used when ideal endpoints are not easily accessible for measurement as long as their relationship to the development of the principal outcome relevant to the claim is well characterised and substantiated. The overall coherence and totality of published and unpublished evidence should be considered in the process. Assessments for substantiation claims need expert judgement, weighting of the strength of the claim, and intelligent use of the criteria applied on an individual basis with respect both to gaps in the knowledge and to any need for new knowledge and data.