PFAS ("Forever Chemicals") and Diet: What's the Evidence?

Last reviewed: April 28, 2026

Summary

PFAS are persistent industrial chemicals that bioaccumulate up the food chain — and seafood, eggs, and meat are the dominant dietary sources for most adults. A plant-forward diet meaningfully reduces PFAS body burden, but plants aren't PFAS-free; filtered water and avoiding grease-resistant packaging matter too.

Supported by 6 cited sources

Key Points

1PFAS (per- and polyfluoroalkyl substances) are a family of more than 12,000 synthetic chemicals built around extremely stable carbon-fluorine bonds. The long-chain compounds that dominate health regulation — PFOA, PFOS, PFNA, and PFHxS — have human elimination half-lives ranging from roughly 2.7 years (PFOA) to 5-9 years (PFHxS), which is why they are nicknamed forever chemicals. The European Food Safety Authority (EFSA, 2020) set a group tolerable weekly intake of 4.4 ng/kg body weight per week for the sum of these four PFAS, and reported that mean upper-bound dietary exposure exceeds this threshold in toddlers and parts of the adult population.
2Established health effects at currently observed exposures include reduced antibody response to vaccines (the most sensitive endpoint and the basis of the EFSA TWI), elevated total and LDL cholesterol, elevated ALT, and modestly lower birth weight. In 2023 IARC classified PFOA as a Group 1 (known) human carcinogen, with the strongest evidence for kidney and testicular cancer. Thyroid cancer associations remain inconsistent — a 2024 meta-analysis by Van Gerwen and colleagues found no significant pooled effect.
3Dietary exposure ranking from EFSA's 2020 assessment is unambiguous for adults: fish and seafood are by far the largest contributor, especially predatory and freshwater species (perch, pike, eel, tuna, smelt) and bivalves. Eggs come next — protein-rich yolks bind PFAS, and backyard or free-range eggs near contaminated land can be extreme outliers. Meat and offal (especially liver and kidney, which concentrate PFAS) are meaningful contributors. Fruit appears prominent in some intake studies because consumption volume is high, but per-gram concentrations are low. Drinking water is highly variable but can dominate exposure in PFAS-impacted areas. Dairy is generally low.
4The most direct dietary comparison comes from Menzel et al. (2021), who measured plasma PFAS in 36 vegans and 36 omnivores in Berlin. Vegans had significantly lower PFOS (median 2.31 vs 3.57 ng/mL) and lower PFNA than omnivores. PFOA and PFHxS did not differ — both track drinking water more strongly than diet. Meat consumption was the strongest dietary correlate of PFOS and PFNA in the omnivore group. The FDA Total Diet Study (2019-2024) found no detectable PFAS in 95% of 1,352 retail samples, with positives concentrated in cod, shrimp, clams, salmon, tilapia, and catfish — consistent with the seafood signal.
5Plants are not PFAS-free, and the claim that they are is wrong. Roughly 70 million US acres may have received PFAS via biosolids (sewage-sludge fertilizer), and contaminated irrigation water adds further risk. Leafy greens (kale, lettuce, spinach) show the highest plant uptake; strawberries and apples show measurable but lower uptake. Typical levels in commercial plant foods remain well below those in seafood and contaminated eggs, but local hotspots exist and are real.
6Mitigation steps with the largest leverage, in order: shift protein away from fish and eggs (the single biggest dietary lever); filter drinking water — reverse osmosis removes more than 90% of long- and short-chain PFAS, certified granular activated carbon (GAC) removes long-chain well but is unreliable for short-chain replacements; avoid grease-resistant fast-food and bakery packaging (the FDA confirmed a US phase-out of fluorinated food-contact paper in February 2024, but legacy stock remains in circulation); replace damaged or overheated nonstick cookware; if you eat fish, prefer small forage species (sardines, anchovies) over large predators and freshwater eels.
7The April 2024 US EPA National Primary Drinking Water Regulation set legally enforceable maximum contaminant levels for PFOA and PFOS at 4 parts per trillion and individual MCLs for PFNA, PFHxS, and HFPO-DA (GenX) at 10 parts per trillion. Public water systems must complete monitoring by 2027 and achieve compliance by 2029. Until then, point-of-use filtration remains the most reliable individual-level safeguard.

Evidence Summary

The strongest single document on PFAS in food is the EFSA CONTAM Panel's 2020 scientific opinion (EFSA Journal 18(9):6223), which integrated toxicokinetics, dose-response on antibody response and other endpoints, and pan-European dietary intake modeling to derive the 4.4 ng/kg bw/week group TWI. EFSA's exposure assessment ranks fish and other seafood as the dominant dietary contributor for adults across most European populations, followed by eggs, meat, and (where consumption is high) fruit.

...

Several caveats apply. The Menzel et al. (2021) comparison is observational and relatively small (n=72), so residual confounding from drinking-water source, geography, and lifestyle cannot be fully excluded. Regulatory thresholds are evolving rapidly — the EFSA TWI dropped roughly 2,000-fold from earlier values, and further revisions are plausible as the evidence base grows. The phase-out of long-chain PFAS has been accompanied by a rise in short-chain replacements (such as GenX/HFPO-DA and PFBS) whose long-term health and exposure profiles are less characterised; some short-chain compounds are poorly removed by GAC filters. Plant-food contamination from biosolids and irrigation water is highly site-specific, so national averages can both understate and overstate individual exposure. Finally, most dietary studies measure four to a few dozen PFAS out of thousands in commerce, so reported totals are systematic underestimates.

Supporting Evidence

Based on the EFSA CONTAM Panel's 2020 scientific opinion, which ranked dietary contributors to summed PFOA+PFOS+PFNA+PFHxS exposure across European intake surveys. Fish, especially predatory and freshwater species, dominated; eggs, meat, and offal followed. The FDA Total Diet Study (2019-2024) is consistent: detectable PFAS in retail food was concentrated in seafood (cod, shrimp, clams, salmon, tilapia, catfish), with no detectable PFAS in 95% of 1,352 samples overall.

Sources:

Menzel et al. (2021) measured plasma PFAS in 36 vegans and 36 omnivores in Berlin. Median PFOS was 2.31 ng/mL in vegans versus 3.57 ng/mL in omnivores; PFNA was also significantly lower. PFOA and PFHxS did not differ between groups, consistent with these compounds tracking drinking water more than diet. In multivariable models, meat consumption was the strongest dietary correlate of PFOS and PFNA.

Sources:

Menzel J, Abraham K, Dietrich S, et al.. Internal exposure to perfluoroalkyl substances (PFAS) in vegans and omnivores (2021)

The EFSA CONTAM Panel (2020) derived a group TWI of 4.4 ng/kg body weight per week for the sum of PFOA, PFOS, PFNA, and PFHxS, anchored on reduced antibody response to childhood vaccines as the most sensitive endpoint. Mean upper-bound dietary intake estimated from EU consumption surveys exceeded this TWI for toddlers and substantial fractions of adolescent and adult populations, indicating that current exposures are not negligible.

Sources:

Multiple agronomic and exposure reviews, including Sunderland et al. (2019), document PFAS uptake into edible plant tissues from contaminated soil and irrigation water. Leafy greens such as kale, lettuce, and spinach show the highest uptake; strawberries and apples show measurable uptake. Roughly 70 million US acres may have received PFAS via biosolids (sewage-sludge fertilizer). Despite this, typical commercial plant-food PFAS levels remain materially lower than seafood and contaminated eggs.

Sources:

US EPA technical guidance and the April 2024 PFAS National Primary Drinking Water Regulation identify reverse osmosis (RO) and certified granular activated carbon (GAC) as the principal proven point-of-use and centralised treatment technologies for PFAS. RO consistently achieves greater than 90% removal across long- and short-chain PFAS in independent testing; GAC removes long-chain compounds (PFOA, PFOS, PFNA, PFHxS) effectively but is less reliable for short-chain replacements such as PFBS and GenX, and requires media replacement on a defined schedule.

Sources:

The Bottom Line

On the best available evidence, a plant-forward diet meaningfully reduces PFAS body burden, primarily by displacing fish, eggs, and meat — which are the dominant food sources of the most-studied PFAS. Plants are not PFAS-free, and a vegan diet does not exempt anyone from filtered water, sensible packaging choices, and the broader policy fight to phase these chemicals out of commerce. The dietary lever is real, well-supported by both intake modelling and direct biomonitoring, and additive with point-of-use water filtration.

Practical Takeaways

Concrete actions, ordered by impact for a typical adult: (1) reduce or eliminate fish and seafood — especially predatory and freshwater fish (tuna, swordfish, perch, pike, eel) and bivalves; if you eat fish, prefer sardines and anchovies; (2) reduce or eliminate eggs from unknown or backyard sources near agricultural land or known PFAS hotspots; (3) install a reverse-osmosis system at the kitchen tap (>90% removal of long- and short-chain PFAS) or a certified GAC filter (good for long-chain only) — check your local water utility's PFAS reports; (4) avoid grease-resistant takeout containers, microwave popcorn bags, and bakery wrappers when possible; (5) replace damaged or pre-2013 nonstick cookware and never preheat empty nonstick pans on high; (6) treat backyard-egg gifts from contaminated areas with caution and rotate produce sources if you are near known PFAS plumes.