Findings and scope
This is a letter in the Environmental Health Perspectives correspondence section, which the journal prints as “not peer-reviewed” (PDF p.1). The writer, employed by a consulting firm that supports “the regulated community” (PDF p.2), criticises the Amster et al. (2007) case report linking a kelp supplement to arsenic toxicosis. The letter argues that the case report measured total arsenic only and did not consider that arsenic in seaweed is mostly organic, mainly arsenosugars, which the writer describes as much less toxic than inorganic arsenic. It cites Rose et al. (2007) as finding inorganic arsenic undetectable in most seaweed varieties, and argues that the patient’s urinary arsenic and symptoms do not indicate arsenic toxicity. The letter reports no measurements of its own.
Key numbers
The letter prints no arsenic concentration for any kelp supplement or seaweed. One secondhand measurement is relayed and was captured as context only.
| Relayed figure | As printed (PDF p.2) | Primary source |
|---|---|---|
| Patient urinary arsenic | 85.5 µg/g creatinine | Amster et al. 2007 |
The letter also cites reference values that are not measurements: normal urinary arsenic of 50 µg/g creatinine (as stated in the case report) and 50 µg/L (ATSDR 2000), urinary arsenic up to 1,000 µg/L after seafood (Vahter 1994), the FDA 2 ppm arsenic standard for tissues of animals treated with veterinary drugs, and an FDA level of concern of 86 ppm total arsenic in crustaceans (FDA 1993). It states that this level is “10 times higher than the amount found in the kelp supplement” without printing that amount.
Methods and Evidence Fitness
There are no methods; the letter is argument supported by citations. The arsenic species claims rest on cited literature, not on speciation of the supplement in question.
Evidence Fitness is Context only. The letter documents one side of a published exchange about arsenic species in kelp supplements and points to the primary case report. It cannot supply any concentration or inorganic fraction for kelp supplements.
Assessment limits
The letter appears in a four-page correspondence spread (PDF pp.1 to 4) that also contains a letter from a natural products trade association (doi:10.1289/ehp.10360; PDF p.1), a letter from a herbal products trade association (doi:10.1289/ehp.10393; PDF pp.2 to 3), and a reply from the case-report authors (doi:10.1289/ehp.10393R; PDF pp.3 to 4). The reply disputes this letter’s premise. It cites unpublished data (Krishna et al.) reporting that 90 to 95% of the arsenic in ethanolic kelp and bladderwrack extracts was inorganic, and argues that most arsenic in kelp supplements may therefore be inorganic (PDF p.4). Neither position is supported by measurements printed in this PDF. The kelp supplement concentrations from the case report (1.59 to 65.5 ppm dry weight in eight of nine products) appear only in the co-printed letter on PDF p.1, not in this letter, and were not captured here. Each writer declares employment or a role relevant to the question.
Provenance notes
The PDF (four pages, A574 to A577) was delivered in batch 03 of order HMI-GAP-INGEST-1 on 22 October 2026, filed under the manifest’s kelp folder. The SHA-256 matches the batch manifest. The letter on PDF p.2 prints doi:10.1289/ehp.10472 and the title “Organic versus Inorganic Arsenic in Herbal Kelp Supplements”, both matching the manifest. The same correspondence spread is held in the corpus under a different file hash as the source page for the co-printed letter doi:10.1289/ehp.10360.
Related evidence
- Arsenic in Herbal Kelp Supplements: Concentration, Regulations, and Labeling
- Seaweed
- Ingredient class — algae/seaweed-based
- Seaweed/kelp foods (nori, wakame, kombu, dulse — as food products)
- Arsenic, Inorganic
- Arsenic, Total
Update history
22 October 2026: Ingested from the batch-03 PDF as correspondence. One secondhand urinary arsenic figure was added to the evidence register as context only.
Update history
No substantive edit history is available in this build. The full commit record is available in git.