Carrageenan (E407): What Regulators and Human Studies Show

Food-grade carrageenan (E407) and poligeenan are different materials. EFSA’s 2018 assessment found no genotoxicity or carcinogenicity concern for the food additive it evaluated, but retained a temporary group acceptable daily intake because important uncertainties remained. Human studies have tested food-grade carrageenan directly, including small trials in ulcerative colitis and healthy adults. The material distinction matters; it does not settle every question about human health.^[1][2][3]

Food-grade carrageenan and poligeenan: why the distinction matters

Carrageenan, listed as E407, is a thickener pulled from red seaweed. It is what keeps the almond milk from separating in the carton and gives sliced deli turkey its clean texture. Chemically, it is a long sulfated polysaccharide chain, intact and high molecular weight.

Poligeenan is a degraded product made by acid hydrolysis of carrageenan, with an average molecular weight of 10,000 to 20,000 daltons. EFSA’s 2018 assessment states that it has not been authorised as a food additive. Findings from degraded material must be identified as such; they cannot automatically be assigned to food-grade E407.^[1]

A 2010 study by Bénard and colleagues in PLoS ONE showed exactly how that size difference matters. In rats, a 40 kDa degraded carrageenan fraction triggered colitis, while a 10 kDa fraction from the same degradation process did not. Both fragment sizes activated inflammatory signaling (NF-kB, TNF-alpha, ICAM-1) in monocytes in culture.^[4] That is a study of the degraded breakdown product, not the intact food-grade polysaccharide on your ingredient panel.

Different materials, with different evidence

Data from EFSA's 2018 re-evaluation of carrageenan (E 407).

What EFSA’s 2018 review found

EFSA did not wave carrageenan through without looking. Its 2018 opinion lays out the toxicology in specific numbers:

• Subchronic toxicity in rats: no adverse effects up to 3,400 to 3,900 mg per kilogram of bodyweight per day, the highest dose tested.
• Chronic toxicity in rats: no adverse effects up to 7,500 mg per kilogram of bodyweight per day, again the highest dose tested.
• Carcinogenicity and genotoxicity: no concern identified for either.^[1]

75 mg/kg the daily intake limit EFSA kept for carrageenan, marked temporary pending more data PMID 32625873

What “temporary” meant in 2018: EFSA retained the existing group acceptable daily intake of 75 mg per kilogram of bodyweight per day for E407 and E407a, and made it temporary because of uncertainties in chemistry, exposure, and biological and toxicological data.^[1]

The panel requested an improved database within five years. Its uncertainty discussion included the characterisation of test materials, low-molecular-weight fractions and biological findings needing clarification. The absence of a genotoxicity or carcinogenicity concern in that assessment is not a finding that every health question was resolved. These are the conclusions of the 2018 opinion, not a claim that no later regulatory work exists.^[1]

Why one researcher’s name keeps coming up

Much of the public case against carrageenan traces back to one scientist. Joanne Tobacman’s widely cited 2001 review in Environmental Health Perspectives argued that even undegraded, food-grade carrageenan was associated with intestinal ulcerations and tumors in animal models, and called for its dietary use to be reconsidered.^[5]

The 2001 review and the 2018 regulatory assessment serve different purposes. Assessing their implications requires matching each experiment’s material, route, dose and outcome to the claim being made. A high-dose animal result does not establish a human dietary risk, and a regulatory assessment does not replace later human research. This article does not count all animal studies or establish that most adverse findings involved poligeenan.^[5][1]

The small ulcerative-colitis trial was reused in later reviews

A 2017 pilot trial put all participants on a carrageenan-free diet, then compared food-grade carrageenan capsules with placebo capsules. Fifteen people enrolled, three declined further participation after randomisation, and twelve followed the study diet: five in the carrageenan group and seven in the placebo group. This was a small challenge study in ulcerative colitis remission, not a trial showing that ordinary food products cause disease in the general population.^[6]

The original report observed three relapses in the carrageenan group and none in the placebo group. Recruitment stopped after a significant time-to-relapse analysis; the paper also reports that a different statistical comparison was not significant. The small sample, withdrawals and analysis choices limit how confidently this finding can be generalised.^[6]

12 participants followed the study diet in the 2017 ulcerative-colitis pilot; five received carrageenan and seven placebo PMID 28447072

The 2019 Cochrane review rated its carrageenan-free-diet evidence very low certainty. That assessment is consistent with substantial uncertainty; it is not a definitive recommendation for or against avoidance.^[7]

The 2023 review included fourteen trials across several dietary interventions. Its Table 1 identifies the same 2017 Bhattacharyya trial as the carrageenan study, and Table 3 lists one carrageenan-free-diet RCT. The fourteen trials were not fourteen independent carrageenan trials. The review’s non-significant carrageenan-specific estimate is a re-analysis of a small evidence base, not a larger independent replication disproving the pilot.^[8]

How to read the reviews: do not count a trial again when it appears in a later review, or confuse pooled results across different diets with carrageenan-specific evidence. The reviews use different denominators and outcome handling; their percentages should not be presented as separate confirmatory experiments.^[6][8]

Two later human trials add context, with important limits

In a 2023 randomised crossover study, seven women with quiescent ulcerative colitis consumed food-grade carrageenan at 2 g/day or an oat beta-glucan preparation during seven-day periods. There was no significant difference in the primary SCCAI disease-activity score or measured biochemical markers. The comparator was biologically active, and the tiny sample and short duration limit conclusions about longer-term relapse or safety.^[2]

A 2024 randomised crossover trial enrolled twenty young, healthy men and compared two-week periods of food-supplier E407 at 250 mg twice daily with placebo. Overall insulin sensitivity did not differ on the primary OGTT assessment or the co-primary clamp assessment; nineteen participants had paired clamp data. Increased intestinal permeability and BMI-related exploratory findings were also reported. These findings do not establish diabetes incidence or the effects of ordinary food products, and the supplier sheet did not fully characterise the carrageenan preparation.^[3]

What the ultra-processed food research does and doesn’t show

A 2024 dose-response meta-analysis by Babaei and colleagues, covering more than 4 million participants across 24 studies, found ultra-processed food intake associated with a higher risk of inflammatory bowel disease overall, with a clearer dose-response relationship for Crohn’s disease specifically.^[9]

That is a real finding, and it gets cited constantly in carrageenan scare content. But a study built that way tracks ultra-processed food as a whole, a category that includes dozens of co-occurring additives, refined carbohydrates and altered textures. It cannot isolate carrageenan’s individual contribution, any more than it could isolate any single ingredient in a food category defined by dozens of them.

Has anyone measured carrageenan intake in IBD patients directly?

Here is the genuinely surprising part. Katsoudas and colleagues’ 2024 Australian cohort study did exactly the direct measurement the scare content assumes has never been done. Analyzing food records from 367 participants (176 with inflammatory bowel disease, 191 healthy controls), the researchers tracked exposure to carrageenan and five other emulsifiers implicated in gut inflammation research by examining ingredient lists.^[10]

Reported exposure to the six selected emulsifiers was low or absent in these food records. This describes the sampled participants and foods; it does not measure new disease incidence, establish exposure in every country, or demonstrate that carrageenan is harmless.^[10]

What this means when reading a food label

A carrageenan entry on a label identifies an ingredient, not a diagnosis or an individual risk prediction. EFSA’s 2018 findings on genotoxicity and carcinogenicity should be read alongside the uncertainties recorded in the same opinion. Rat toxicology doses cannot be described as thousands of times higher than anyone eats without a defined human-exposure comparison.^[1]

For inflammatory bowel disease, the studies discussed here leave important limits:

• The 2017 pilot raised a relapse question in a small group with ulcerative colitis in remission; later reviews of that trial do not create independent participants.^[6][8]
• A separate 2023 crossover trial found no significant difference in its primary disease-activity endpoint over seven-day periods in seven women with quiescent ulcerative colitis.^[2]
• Short experiments in selected populations cannot establish long-term safety for everyone or show that avoidance prevents relapse. The 2024 metabolic trial adds human evidence outside IBD, with overall primary results and exploratory findings that must be kept separate.^[2][3]

If you have IBD, discuss dietary questions with your treating clinician or dietitian rather than using this article to start an elimination experiment or alter treatment. The EatMatrix app can help identify carrageenan on a product’s ingredient panel; ingredient identification is separate from personalised medical advice.

For ingredient names and the selected regulatory and research records, see the E407 carrageenan guide. The material tested, population and outcome all matter when interpreting a headline.

This article summarises publicly available regulatory assessments and peer-reviewed research about food ingredients. It is general information and is not intended to diagnose, treat, cure, or prevent any disease, and it is not personalised dietary or medical advice. Regulatory positions change; the EatMatrix app reflects the current assessments in our database. If you have an allergy, a medical condition, or are pregnant, talk to a qualified clinician or dietitian before changing what you eat.