The quality of science for policy depends as much on the robustness of available scientific knowledge as it does on the procedural settings and working procedures in safety agencies. Using a report on Bisphenol A as a case study, and a set of original criteria, we provide an understanding of procedural influences on the results of scientific advisory committees and about literature reviews for chemical hazard characterization. Expert elicitation revealed that three aspects are critically important for the results of the advisory activity and for the selected case study: the method used to combine different studies, the interpretation of the review results in terms of level of evidence and conclusiveness, and the choice of uncertainty factors. Our results also show how procedural settings and working procedures can promote the invisible influence of values and policy on scientific advisory activities.