Is gluten bad for you?

The objection

“Vegan diets lean on bread, pasta and seitan, so they’re full of gluten. Gluten loosens the lining of the gut even in people without coeliac disease, it’s inflammatory, and plenty of people feel better the moment they give it up.”

The answer

For people with coeliac disease, gluten is genuinely harmful: it makes the immune system attack the lining of the small intestine, and the only treatment is a strict gluten-free diet for life [3][5]. A 2026 review pooling 87 papers covering 394,274 people around the world put the share affected at 1.5% by antibody blood test and 0.8% when confirmed by biopsy, about 1 in 120 [1], and the NHS (2023) says at least 1 in 100 people in the UK have it [2]. A vegan diet can be gluten-free, since rice, potatoes, lentils, fruit and vegetables contain none [6], and neither do flours made from rice, corn, soya or potato [3]. Labels still need checking, because many processed foods contain gluten in additives such as malt flavouring [3]. Oats contain no gluten but can be contaminated with it, so people with coeliac disease who want them are advised to talk to their healthcare team and choose only oats labelled gluten-free, and a few react even to those [3]. Anyone diagnosed on the NHS is referred to a dietitian to make sure the new diet is balanced [3].

The step from there to “bad for everyone” is where the evidence gives out. Long-term studies of people without coeliac disease have mostly found no link between gluten and disease [7][15][16][17][18], and wheat eaten as wholegrain is linked to better health [10].

Does gluten harm people who don’t have coeliac disease?

In large studies that followed adults without coeliac disease for years, eating more gluten has mostly not been linked to heart disease, digestive cancers, inflammatory bowel disease or earlier death, though one UK study found a link with deaths from heart disease [7][15][16][17][18]. The earliest, published in the BMJ in 2017 and described by its authors as the first long-term study of the question, followed 64,714 women in the Nurses’ Health Study and 45,303 men in the Health Professionals Follow-up Study, none with a diagnosis of coeliac disease or with heart disease at the start, for 26 years, during which 6,529 had a heart attack [7]. Before any adjustment, the fifth eating the most gluten, averaging 7.5 g a day for women and 10 g for men at the start, had fewer heart attacks than the fifth eating the least, 277 against 352 per 100,000 people a year [7]. Once smoking, weight, diabetes, blood pressure, cholesterol, exercise and other parts of the diet such as meat, fruit and vegetables were allowed for, the gap was no longer statistically significant, with a hazard ratio of 0.95 (95% confidence interval 0.88 to 1.02) [7].

The researchers then looked at which grains the gluten came with. After allowing for refined grains, so that what was left of gluten intake tracked wholegrains, eating more gluten went with about 15% lower risk (hazard ratio 0.85, 0.77 to 0.93), and after allowing for wholegrains instead, gluten intake was not linked to risk either way (1.00, 0.92 to 1.09) [7]. The authors concluded that avoiding gluten may mean eating fewer of the wholegrains linked to heart health, and that gluten-free diets should not be promoted to people without coeliac disease to prevent heart disease [7]. Being observational, the study can show links but cannot prove cause. Gluten was estimated from food questionnaires, the participants were mostly white health professionals, and the study could not pick out people without coeliac disease who kept to a very low-gluten or gluten-free diet, though the follow-up, from 1986 to 2012, largely came before gluten became a popular health concern [7].

Later studies mostly point the same way. In 159,265 UK Biobank adults followed for about 11 years, gluten intake was not linked to death from any cause or from cancer, but it was linked to more deaths from ischaemic heart disease, a result that held after the authors (2021) corrected for multiple comparisons and that they said needs further study [15]. In 29,079 Japanese adults followed for 16 years, the quarter eating the most gluten had fewer deaths from cardiovascular disease than the quarter eating the least (hazard ratio 0.73, 0.62 to 0.86), a 2025 study found [16]. In three Harvard cohorts of more than 200,000 US adults, gluten intake was not linked to digestive cancers (hazard ratio 0.94, 0.87 to 1.02, most against least, in 2022) [17] or to inflammatory bowel disease (2022) [18].

The case that gluten harms everyone draws on mouse studies, a snapshot study linking gluten intake to a marker of inflammation in young adults, and the high glycaemic index of many foods that contain it, as the 2017 authors describe [7]. The laboratory evidence on the gut lining is real too: in a 2015 study, small-intestine biopsies bathed in digested gliadin, a part of gluten, became more permeable over two hours than tissue left in culture medium alone, in all four groups tested, including five people without coeliac disease who were having an endoscopy for indigestion [8]. A two-hour test on tissue from 23 people shows what gluten fragments can do to the gut lining in the laboratory, but it cannot tell us whether eating wheat for decades causes disease, and the long-term studies above found no consistent link with heart disease [7][15][16].

The amount of gluten may matter for children who carry gene types linked to coeliac disease. In the TEDDY study (2019), which tracked gluten intake in 6,605 such children in Finland, Germany, Sweden and the US, each extra gram of gluten a day in the first five years went with a higher chance of developing coeliac disease (hazard ratio 1.50, 1.35 to 1.66), an absolute difference the authors put at 7.2 percentage points by the age of three [19]. A Norwegian study of 86,306 children (2020) linked more gluten at 18 months with type 1 diabetes (hazard ratio 1.46 for each extra 10 g a day, 1.06 to 2.01), but found no link with how much gluten mothers ate in pregnancy [20]. Both are observational and neither tested a gluten-free diet, so if coeliac disease runs in the family, talk to a GP before changing a child’s diet [4].

Are wheat and other wholegrains good for you?

For the general population, wholegrains, wheat included, are linked to lower rates of heart disease and early death. A 2016 BMJ meta-analysis of 45 cohort studies found that each extra 90 g of wholegrains a day, about three servings such as two slices of wholegrain bread and a bowl of wholegrain cereal, went with 19% lower risk of coronary heart disease (relative risk 0.81, 0.75 to 0.87) and 17% lower risk of death from any cause (0.83, 0.77 to 0.90) [10]. Wholegrain bread, wholegrain breakfast cereals and added bran were each linked to lower risk, and refined grains showed little sign of any link [10]. People who eat more wholegrains may differ in other ways, though the links held in studies that allowed for smoking, alcohol, exercise, weight and other foods [10].

A 2019 series of Lancet meta-analyses, funded in part by the WHO, pooled 185 cohort studies and 58 trials and found in the cohort data that people eating the most fibre had 15 to 30% lower rates of death, heart disease, stroke, type 2 diabetes and bowel cancer than those eating the least, with similar findings for wholegrains [9]. It graded the certainty of the evidence as moderate for fibre and low to moderate for wholegrains, and the authors judged that the striking dose-response pattern meant the links could be causal [9].

A 2023 meta-analysis of 46 cohort studies agrees: each extra 30 g of wholegrains a day went with 6% lower risk of coronary heart disease and of death from any cause (relative risk 0.94, 0.92 to 0.97, for both), and its authors rated the evidence moderate and high quality respectively [21]. Trials are thinner. A 2017 Cochrane review found nine, with 1,414 people, none longer than 16 weeks and none reporting heart attacks, strokes or cardiovascular deaths, judged all of them at unclear or high risk of bias, and found too little trial evidence to say whether wholegrains change heart disease, cholesterol or blood pressure [22]. Four of the trials were funded by independent national or government bodies, and the other five had some funding from organisations with commercial interests in cereals [22].

Seitan is a different thing from the wholegrain bread and cereals in the 2016 review, a concentrated wheat protein with almost no fibre left, so it is a useful protein food rather than a wholegrain. It is made from a bread-like wheat dough washed under running water, which removes most of the fat and carbohydrate [11], and the figures for dried wheat gluten show what that does.

Per 100 g	Protein (g)	Fibre (g)
Wholegrain wheat flour [12]	13.2	10.7
Dried wheat gluten, sold as vital wheat gluten [12]	75.2	0.6

Why do some people feel better without wheat?

Some people react to wheat without having coeliac disease, through a wheat allergy or what is called non-coeliac gluten sensitivity [13]. A wheat allergy is a separate condition in which the immune system overreacts to wheat, and the NHS (2026) says a GP can refer you to a specialist for skin-prick or blood tests, with treatment depending on how severe it is, from antihistamines to emergency adrenaline [23]. Gluten sensitivity has no test and remains a diagnosis made by ruling other conditions out [13]. The 2025 Lancet review estimates that about 10% of adults worldwide say they are sensitive to gluten or wheat, but in controlled challenge studies only 16 to 30% of them have symptoms triggered by gluten itself [13]. It adds that fermentable carbohydrates in some of the test preparations make those studies harder to interpret, and concludes that fermentable carbohydrates and the nocebo effect, feeling worse because you expect to, contribute considerably to symptoms in many cases [13]. It also notes that the size of the gluten-free market raises questions about how commercial and media interests shape the way gluten sensitivity is portrayed [13].

Cutting gluten can ease bloating even in healthy people. In a 2018 Danish crossover trial, 60 healthy middle-aged adults reported less bloating after eight weeks eating 2 g of gluten a day than after eight weeks eating 18 g, and the researchers suggested that most of the effects of the low-gluten diet may have come from changes in the kinds of fibre eaten [24].

One trial, run in Oslo and published in 2018, gave 59 people who had cut out gluten themselves, and had coeliac disease ruled out, muesli bars hiding gluten, fructans (a fermentable carbohydrate that often comes with gluten in foods) or neither, for a week each without knowing which [14]. Fructans produced higher gut symptom scores than gluten, gluten did no worse than the dummy bar, and 13 people felt worst after gluten, 24 after fructans and 22 after the dummy [14]. A larger trial in the UK and the Netherlands (2024), part-funded by bakery and wheat-industry bodies, gave 84 people with self-reported gluten sensitivity bread with or without gluten, and told them at random that it did or did not contain gluten [25]. Symptoms were worst in those who both expected gluten and got it, and people given gluten while told the bread was gluten-free did no worse than those who had none, which the authors read mainly as a nocebo effect, though they said an additional effect of gluten could not be ruled out [25].

If gluten seems to disagree with you, see a GP before cutting it out. The blood test and biopsy are only accurate while you are still eating gluten, and the NHS advises against starting a gluten-free diet until a specialist has confirmed the diagnosis [4]. Around two in three people in the UK with coeliac disease have not yet been diagnosed, since the charity Coeliac UK’s current estimate is that only 36% have been [5]. Having no symptoms does not rule it out either, since about 41% of people with biopsy-confirmed coeliac disease in the 2026 review had none [1]. A 2024 review of 34 studies found that about 1 in 14 parents, children, brothers and sisters of people with coeliac disease had it confirmed by biopsy, and its authors suggested routine screening for them [26], so close relatives should ask a GP about testing [4]. If cutting out foods or worrying about what you eat starts to take over, eating-disorder charities can help: Beat in the UK, Bodywhys in Ireland, ANAD in the US, NEDIC in Canada, the Butterfly Foundation in Australia, and EDANZ, which supports families and carers, in New Zealand.