To the Editor: Dr Navas-Acien and colleagues1 extended the observation of a diabetogenic effect of arsenic exposure from drinking water containing arsenic levels greater than 100 μg/L in Taiwan, Bangladesh, and Mexico to a fairly low exposure of approximately 0.1 μg/kg per day in the United States. This low exposure dosage is approximately 55 times lower than that observed in the arseniasis-endemic areas (assuming that a 55-kg adult drinks 3 L of water containing an arsenic level of 100 μg/L per day). The odds ratios comparing the 80th vs the 20th percentile of total urine levels of arsenic and the trend test for the tertile stratification were significant only in the models considering seafood intake (ie, with additional adjustment for arsenobetaine and blood mercury levels).
Tseng C. Arsenic Exposure and Diabetes Mellitus in the United States. JAMA. 2008;300(23):2728-2729. doi:10.1001/jama.2008.812