Low-dose effects and nonmonotonic dose–responses of endocrine disrupting chemicals: Has the case been made?

Vandenberg et al. (2012) claim that "most if not all [endocrine-disrupting chemicals (EDCs)] are likely to have low-dose effects" and "nonmonotonicity is a common occurrence after exposures to hormones and EDCs in cell culture and animals and across human populations." They present examples as anecdotes without attempting to review all available pertinent data, selectively citing studies without evaluating most of them or examining whether their putative examples are consistent and coherent with other relevant information. They assume that any statistically significant association indicates causation of an adverse effect, and their limited evaluation of specific studies is not done uniformly (i.e., studies with positive results are evaluated differently than those with null results). They also do not evaluate whether exposures in studies are truly "low-dose" and relevant to humans. They propose a number of different nonmonotonic dose-response curves, but do not consider reasons for why they should be expected to apply generally across species. Many of their examples would be - and indeed have been - questioned by many scientists. Overall, Vandenberg et al. put forth many asserted illustrations of their two conclusions without providing sufficient evidence to make the case for either and while overlooking evidence that suggests the contrary.