Epidemiological study outcomes provides the most direct evidence of environmental toxicant effects on human body, rather than experimental animal studies. However, up to date, almost all the current regulatory level of tolerable daily intake (TDI) (or reference dose, RfD) was established based on toxicological data, with the exception of methylmercury. One main difficulty might be exposure uncertainties as well as lack of suitable large-scale observational studies. In this talk, I will briefly explain the derivation of benchmark dose (BMD) for the RfD of methylmercury by US EPA IRIS in 1995 based on Faroe Islands data. I will then demonstrate the BMD we derived for melamine based on 13 nephrolithiasis cases from 932 children who enrolled in three Department of Health hospitals in Taiwan during the melamine outbreak in 2008. Finally, for the phthalate incident in 2011, we derived the corresponding BMD for di-2-ethylhexyl phthalate (DEHP) for pregnant women based on 122 mother-child pairs for the association between prenatal DEHP exposure and behavioral problems in children.