Variable effectiveness of raw rice, cooked rice, and water in recapitulating dietary inorganic and organic arsenic exposure in the health effects of arsenic longitudinal study

The interaction between water and rice arsenic (As) has long raised questions about what type of As exposure monitoring is both feasible and effective. Here, we leverage matched samples of raw rice, cooked rice, cooking water, and spot urine from 396 participants in the Health Effects of Arsenic Longitudinal Study (HEALS) in Araihazar, Bangladesh, to determine which sample type is most informative of overall As exposure. First, we find that the difference in cooked and raw rice As is most pronounced when cooking water As exceeds the Bangladeshi 50 μg As/L safety limit. Second, we find that cooking rice in high As water can push rice with "safe" levels of As while uncooked into the unsafe range (>200 ng inorganic iAs/g dry rice) after cooking. Third, we find that cooked rice is a more reliable indicator of dietary As exposure across a wide range of water arsenic levels, but also find that neither rice or water As explain much variation in urinary As when As is low in both water and rice. Rice explains moderate urinary variance only when water As is low. Lastly, we find that mono- and di-methylated forms of As (MMA and DMA, respectively) from cooked rice minimally contribute to urinary MMA and DMA levels and do not bias common measures of As methylation efficiency in this population. These data suggest that methylation efficiency measures derived from urinary MMA and DMA levels are robust to dietary influences and support their continued use in epidemiological studies.