§ Claim under review · Study
"A study THIS YEAR detected PFAS 'forever chemicals' in 92% dog and cat food brands." (Claim text as submitted: "A 2026 study tested 100 commercial dog and cat foods for 34 different PFAS forever chemicals and found PFAS in 92% of the pet foods tested")
Verdict
Partially accurate but misleading
Confidence
MediumSummary
The study behind this post is real. Researchers at Ehime University published a paper in Environmental Pollution in 2026 that tested 100 commercial dog and cat food products for 34 PFAS compounds and found PFAS widely present, with fish-based foods showing the highest levels. But the post leaves out the most important detail: every product tested was bought in Japan, so this does not tell you what is in the specific brands sold where you live. The post also says "92% of brands" when the study tested products, not brands, and I could not confirm the 92% figure in the abstract, the university's own press release, or any news coverage of the study. The measured amounts were mostly very small, in the range of fractions of a nanogram per gram, and the study did not examine whether any pet actually got sick. The researchers said more evaluation is needed rather than concluding there is a proven risk. If you want a practical takeaway from the actual paper, it is that fish-based formulas tested highest, which is the one useful detail the post does not mention.
The readings
key figures from the evidencecommercial dog and cat food products tested
target PFAS compounds quantified in the study
claimed PFAS detection rate, unconfirmed in sources
Why this verdict
Evidence
The study is real and the structural numbers in the claim are correct. The researchers conducted a comprehensive analysis of 100 commercially available pet food products for dogs and cats in Japan, encompassing both dry and wet types, to quantify 34 target PFAS compounds.
The paper reports that PFAS were frequently detected, with concentrations varying by food type, ingredients, and country of origin.
On composition and findings: the sample set comprised 48 dog food products (dry: n = 26; wet: n = 22) and 52 cat food products (dry: n = 26; wet: n = 26), selected to represent products commonly available in the Japanese market . The analysis used LC-MS/MS, and PFAS were widespread, with PFOS, PFUnDA, and PFTrDA high in fish-based foods.
On magnitude, which the post omits entirely: PFAS concentrations in dog dry food ranged from below the method detection limit to 1.7 ng/g food (as fed), with a median of 0.2 ng/g and a mean of 0.4 ng/g, while cat dry food ranged from below the detection limit to 16 ng/g (median 0.18 ng/g; mean 1.0 ng/g), with certain products exhibiting high levels. The explicit presence of results "below the method detection limit" confirms that detection was not universal, so a figure below 100% is structurally plausible.
The institutional summary frames the conclusion cautiously: the authors state they analyzed 100 dog and cat food products in Japan and found PFAS widely, with fish-derived ingredients tending to contain higher long-chain PFAS , and that together the studies suggest pet food may be one contributor to long-chain PFAS exposure, particularly in cats .
Findings
✓ What's accurate 6
- A real, peer-reviewed 2026 study exists. It is not fabricated, and the "2026" date is genuine, not a red flag.
- It tested exactly 100 commercial dog and cat food products. Correct.
- It quantified exactly 34 target PFAS compounds. Correct.
- PFAS were found across a wide range of the products tested. The authors themselves use the word "widespread."
- Both dry and wet foods, and both dog and cat foods, were included and contaminated.
- Fish-based products showed the highest levels, which is a real and reproducible finding.
≈ What's misleading 6
- Omitted qualifier, geographic: This is the single largest distortion. Every product tested was purchased on the Japanese market. The post presents it to a general audience as if it describes the pet food on their own shelves. A reader in the US, UK, or elsewhere would reasonably conclude their brands were tested. They were not.
- Category error, "brands" versus "products": The post says "92% of dog and cat food brands." The study tested 100 products, not 100 brands. Multiple products can come from one brand. The claim's own text as submitted says "pet foods tested," which is correct; the post's image and caption change it to "brands," which is not.
- Unverified specific figure: I could not locate the 92% number in the abstract, the Ehime University release, or any of the news coverage retrieved. Coverage consistently uses qualitative language such as "frequently detected," "widespread," and "across many products." One legal-industry summary states that 52 cat food products and 48 dog food products contained PFAS, but those are the exact sample sizes of each group, which suggests that summary is restating the study design rather than a detection rate. The 92% may well appear in a detection-frequency table in the full text, which I could not access. But it is not confirmed by any source I retrieved.
- Omitted magnitude: The post conveys contamination without any concentration figures. The measured levels were in the sub-nanogram to low-nanogram per gram range for most products, with medians near 0.2 ng/g. "Detected" at these levels is a statement about analytical sensitivity, not automatically about harm. Modern LC-MS/MS can detect vanishingly small quantities.
- Implied harm without support: The framing "How sad" and "check what you're feeding your pets" implies established injury to pets. The study did not measure health outcomes. The authors explicitly called for further evaluation rather than declaring a risk.
- Missing actionable finding: The most useful result for an actual pet owner, that fish-based formulas were consistently the most contaminated, is absent from the post. The post's implied action is generalized alarm rather than the specific, evidence-supported distinction the study actually found.
? What's uncertain 4
- The exact provenance of the 92% figure. It is plausible as a detection frequency for "at least one of 34 PFAS" but I could not confirm it in any retrieved source. My search tool quota was exhausted before I could obtain the full-text detection frequency table.
- The contents of the published corrigendum, and whether any reported figure was revised.
- Whether Japanese market findings generalize to other markets. Ingredient supply chains are partly international, and the study did note country-of-origin patterns, but the paper does not test non-Japanese retail products.
- Whether these exposure levels produce any health effect in dogs or cats. The study did not test this and said species-specific guidance values do not yet exist.
Sources
6 of 7 linked to recordsNomiyama K, Sato A, Tanoue R, Saeki K, Ikenaka Y, Mizukawa H. "Widespread PFAS contamination in pet food: Dietary sources and health risks to companion animals." Environmental Pollution, Vol. 395, 15 April 2026, 127779 (epub 6 Feb 2026)
Corrigendum to the above, Environ Pollut Vol. 396, 1 May 2026, 127844 (epub 24 Feb 2026)
Ehime University press release via EurekAlert, 21 Aug 2026, "PFAS in pet food: tracing exposure in dogs and cats"
Phys.org, 28 Feb 2026, "PFAS exposure greater in wet pet food, study suggests"
Frontier Group, 12 Aug 2026
MG+M law firm summary
EWG, 2022 (pet food packaging, not food)