EFSA suggests harmonised default values for use in risk assessment

We have some criticisms concerning the harmonised default values suggested by EFSA:

 First, the extrapolation of the exposure time. Using an uncertainty factor out of two factors implies that the effects are simply regarded as duplicated from a subchronic to a chronic study. This operation does not take into account the specific effects of chronic periods such as hormonal effects that may appear in the range of nanomolar or even less. Here, no factor is applicable because we can not foresee what may happen. Secondly, reducing the uncertainty factor of 100 to specific factors when knowing the toxicokinetic or toxicodynamic characteristics of a compound seems unjustifed. It is less cautious. Thirdly, there is only one adult category. It still lacks some subpopulation such as pregnant women who at present are "treated" like men weighing an average of 70 kg. Factors to convert a dietary dose to a physiological dose appear to be correct and the idea of using Benchmark doses to be more accurate than a LOAEL seems interesting. It avoids the uncertainty associated with the choice of experimental doses, but it must not fall into the trap of linear dose response paradigms.