Overview
Carey and colleagues measured arsenic species in 1,180 polished market-rice samples from 29 sampling zones on six continents. The source reports global medians of 66 µg/kg inorganic arsenic and 21 µg/kg dimethylarsinic acid (DMA). The inorganic-arsenic range was <2–399 µg/kg and the DMA range <2–690 µg/kg. Regional variation was substantial, especially for DMA. These are sample-based historical distributions, not population-weighted estimates or current supplier guarantees.
Key findings and source correction
The medians above correct the previous source summary’s description of them as means. The original reports low inorganic-arsenic medians for Malawi and Tanzania and for some southern Indonesian samples; the previous summary’s reference to Mali and Ghana as East Africa was incorrect. Exact country values previously transcribed here are withheld pending reconciliation with the corrected Table 1.
The publication’s 31 January 2020 correction states that the atmospheric-arsenic column in Table 1 was included by mistake and was not used in the text. The earlier summary’s atmospheric-deposition interpretation is withdrawn.
Methods and Evidence Fitness
The study purchased rice from accessible market stalls and supermarkets. Geographic reach and sample density varied. Grain was freeze-dried and milled; arsenic species were measured by chromatography coupled to ICP-MS, with blanks and rice-flour reference materials.
The work supports species-specific occurrence comparisons within its sampled polished-rice materials. It does not isolate cultivar, soil or irrigation effects, validate current lots, or justify assigning a constant inorganic-arsenic fraction to all total-arsenic results. Historical regulatory comparisons in the paper require a separate current applicability check.
Original locators: Abstract; Rice Sample Collection; Sample Preparation; Results; Change history, 31 January 2020. Corrected primary article.
Related evidence
HMI reference use
The 29 geographic-group medians in Table 1 can contribute to HMI’s cross-study dry-grain inorganic arsenic (iAs) and DMA references. Their reported counts sum to 1,093, whereas the article reports 1,180 overall; the table is therefore used as a summary layer, not a fully reconciled individual-sample register. Primary weights are equal within the sampling family. The atmospheric column removed by the correction is excluded. Values at or below 3 µg/kg remain bounded because the abstract and methods give differing low-result limits.
Update history
No substantive edit history is available in this build. The full commit record is available in git.