Inorganic arsenic in milk and milk chains: pathways, exposure levels and mitigation

Milk and dairy products may contain arsenic from environmental and feed sources, but most field evidence concerns total arsenic rather than inorganic arsenic (iAs). This critical review synthesises farm-to-consumer evidence to characterise water-soil-feed-milk pathways, distinguish total arsenic from species-resolved evidence, and prioritise feasible monitoring and mitigation options across production systems. Controlled studies report milk total As near or below low-µg kg^-1 concentrations despite higher intake, whereas some field surveys report higher total As in milk under chronic mixed-source exposure. Paired field studies report associations between arsenic in livestock water or roughage and total arsenic in milk, but species-resolved water-to-milk transfer coefficients are not yet available. An illustrative intake scenario assumes 5 µg L^-1 iAs in milk; a 7 kg infant consuming 0.6 L day^-1 would receive approximately 0.43 µg kg^-1 bw day^-1. This scenario is not a transfer estimate or a concentration band derived from total-As surveys. The evidence supports risk-based surveillance and identifies water treatment and feed substitution as candidate input controls whose effects on milk iAs require paired, species-resolved validation. Where binding maximum levels for iAs in fluid milk are not defined, clearly labelled non-binding action or trigger levels or buyer specifications can support consistent decision-making.