Gluten-Free Diet

Diet Reference Archive: a medical diet that is also a popular one

Gluten-Free Diet

A way of eating that removes gluten, a protein in wheat, barley and rye. For people with celiac disease it is the medical treatment, lifelong and strict; many other people follow it for reasons the research has not borne out.

Family: Therapeutic and elimination diets Promoted by: gastroenterology guidelines, as the treatment for celiac disease; food manufacturers and the media, for popular use First appeared: as a treatment for celiac disease by the mid-1900s; popular use rose sharply from about 2009 GRADE: Moderate for the European guideline panel's recommendation of a strict, lifelong gluten-free diet, compared with a gluten-containing diet, for adults with diagnosed celiac disease; published guideline rating, European Society for the Study of Coeliac Disease (Al-Toma et al., 2026). A 2026 British guideline rates the evidence behind its matching recommendation Low. Not a rating for people without celiac disease.

ON THIS PAGE

1. In brief
2. Overview
3. In practice
4. For the coach
5. The evidence (opens on request)
6. Mechanisms (opens on request)
7. Risks, contraindications and myths (opens on request)
8. Key terms
9. Review questions
10. References
11. Evidence rating and review record

The summary and sections 1 to 3 are the short version and are open when the page loads. Sections 4 to 6 hold the research detail, the biology, and the cautions. Each of those opens when you select its heading, and you can read the page without opening them. Back to the Diet Reference Archive index.

In brief

What is it? A diet with no wheat, barley or rye, and nothing made from them. It means two different things depending on who is following it. For a person diagnosed with celiac disease, it is the treatment, kept for life and kept strictly. For everyone else, it is a choice, often made for digestive symptoms, energy or weight.

Does it work? For celiac disease, yes. The European panel strongly recommends a lifelong, strict gluten-free diet for adults with celiac disease and assigns Moderate certainty to the evidence behind that recommendation; a 2026 British guideline makes the same strong recommendation and rates its evidence Low. In one randomized trial, screen-detected people with celiac disease healed more of the intestinal lining in a year off gluten than people who kept eating it. When people who believe gluten upsets them are given gluten or a look-alike placebo without knowing which, only a minority met the studies' criteria for gluten-specific symptoms; symptoms also occurred with placebo. For irritable bowel syndrome the evidence is very low in certainty. Small, short studies have measured body weight and metabolic markers in people without celiac disease, including people with metabolic conditions; they do not establish that removing gluten improves long-term health in the general population.

Is it safe? A carefully planned gluten-free diet can be nutritionally adequate, but it takes planning. The two cautions that matter most come before the diet starts and while it runs. Going gluten-free before being tested for celiac disease can make the tests inaccurate, so testing comes first. And many packaged gluten-free products are lower in fiber, iron and some vitamins and higher in sugar and fat than the foods they replace, and they cost more.

Who should not do it without talking to a doctor or dietitian first? Anyone who suspects celiac disease and has not yet been tested. Anyone with a diagnosis of celiac disease, wheat allergy or irritable bowel syndrome, whose diet belongs with their clinician and a registered dietitian. Parents considering it for a child. Anyone who is pregnant. Anyone with type 1 diabetes or who takes thyroid medication. Anyone with a history of disordered eating.

Related pages: Low-FODMAP Diet and Elimination Diets. Full list on the archive index.

1. Overview

Keep two groups apart

This page describes one diet followed by two very different groups of people, and it keeps them separate throughout. For people with celiac disease, an immune condition in which gluten damages the lining of the small intestine, a strict gluten-free diet for life is the treatment that gastroenterology guidelines set out, decided and supervised by a physician and a registered dietitian (Al-Toma et al., 2026; Rubio-Tapia et al., 2023). For everyone else, including people who believe gluten upsets them, people trying to lose weight and people who think gluten-free eating is simply healthier, it is a personal choice whose benefits the research has mostly not confirmed. Two other conditions sit between these groups and are covered separately below: non-celiac gluten or wheat sensitivity, and wheat allergy.

What you eat

The diet is built on foods that contain no gluten to begin with. Meat, fish, eggs, dairy foods, fruits, vegetables, beans, rice and potatoes are naturally gluten-free when they have no additives or seasonings made from wheat, barley or rye, and flours made from rice, corn, potatoes, soy, nuts, cassava, amaranth, quinoa, buckwheat and beans are also safe, according to the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK, n.d.-c). A European guideline for adults with celiac disease describes a balanced version as one built on naturally gluten-free grains such as quinoa, brown rice, millet, amaranth, buckwheat and certified gluten-free oats, with fruits and vegetables, lean proteins, dairy foods or fortified alternatives and healthy fats, and with gluten-free breads and pastas used to keep meals familiar and social (Al-Toma et al., 2026).

Oats need a word of their own. Oats do not contain gluten, but ordinary oats are commonly contaminated with wheat, barley or rye during growing and processing. The European guideline judges certified gluten-free oats safe for most people with celiac disease from diagnosis, noting that a small number react to the oat protein itself, and the American College of Gastroenterology recommends gluten-free oats with monitoring for how well they are tolerated (Al-Toma et al., 2026; Rubio-Tapia et al., 2023).

What you avoid or limit

The diet removes wheat and its varieties, such as durum, emmer, semolina and spelt; barley, including malt, malt extract, malt vinegar and brewer's yeast; rye; and triticale, a cross between wheat and rye. It also removes foods made with them, such as most breads, baked goods, baking mixes, cereals and pastas, and it means watching for ingredients made from these grains, such as some colorings, flavorings, starches and thickeners, which turn up in candy, condiments, sausages, ice cream, salad dressings and soups (NIDDK, n.d.-c).

How strict the limit is depends on why the person is following it. For celiac disease it is a hard rule. The European guideline gives a generally accepted ceiling of no more than 10 milligrams of gluten a day, because higher intakes over time can injure the intestine in some people (Al-Toma et al., 2026). That is why a person with celiac disease also has to manage cross-contact, gluten reaching gluten-free food when foods are grown, processed, stored, prepared or served together, which is why the federal guidance advises storing and preparing gluten-free food separately from other family members' food (NIDDK, n.d.-c). For a person following the diet by choice, the limit is whatever they decide it is, and many follow it loosely.

Where it came from

The gluten-free diet began as medicine. A 1993 article in the journal Gut honored the Dutch pediatrician Willem-Karel Dicke, who lived from 1905 to 1962, as the pioneer of the gluten-free diet and marked more than 50 years of its use in treating celiac disease (van Berge-Henegouwen & Mulder, 1993). Since then the diet has remained the treatment. Testing changed around it: biopsy of the small intestine became the standard way to confirm the diagnosis, and blood tests for antibodies made it possible to find people with a wide range of symptoms (Rubio-Tapia et al., 2023).

The popular diet is much newer. A condition now called non-celiac gluten or wheat sensitivity was first described in the late 1970s, with the first published series in 1980, and research interest grew quickly after 2009 (Molina-Infante & Carroccio, 2017). In national United States survey data, the share of people aged 6 and older who avoided gluten without having celiac disease rose from 0.5 percent in 2009 and 2010 to 1.7 percent in 2013 and 2014, while the share with celiac disease stayed at about 0.7 percent (Choung et al., 2017). A review written for gastroenterologists attributes the growth to social and traditional media, marketing by manufacturers and stores aimed directly at consumers, and reports of the benefits of avoiding gluten (Niland & Cash, 2018).

What it claims to do

For celiac disease, the claim is a clinical one: that removing gluten stops the immune reaction that damages the small intestine, so symptoms settle, the intestine heals and later complications become less likely (Al-Toma et al., 2026; NIDDK, n.d.-e). Section 4 shows how well that is supported.

The popular claims are broader. A review of gluten avoidance in people without celiac disease lists the reasons people give: relief of digestive and other symptoms, and a belief that gluten is harmful, so that avoiding it is a healthy way to live (Niland & Cash, 2018). The federal health agency that covers digestive diseases describes people avoiding gluten because they believe the diet is healthier or will help them lose weight (NIDDK, n.d.-c). In an online survey of 910 athletes without celiac disease, 41 percent followed the diet at least half the time, most of them had diagnosed their own sensitivity to gluten, and the authors concluded that a perception of health benefits and an athletic edge, rather than a medical reason, drove most of that choice (Lis, Stellingwerff, Shing, et al., 2015). Section 4 takes each claim in turn.

What it gets right

For the people who need it, this diet is one of the clearest cases in nutrition of a diet that treats a disease. No medication replaces it for celiac disease (Al-Toma et al., 2026; Rubio-Tapia et al., 2023).

For people who do not need it, the diet can still point somewhere useful. A person who goes gluten-free by eating more naturally gluten-free foods, such as vegetables, fruits, beans, rice and potatoes, may end up eating fewer packaged baked goods; if that happens, the change in the overall pattern, rather than the missing gluten, is the part worth keeping. A person's report that wheat-based foods leave them bloated or uncomfortable is real, and it deserves a proper medical explanation even when gluten turns out not to be the cause. What the diet gets right for this group is the attention to how food makes them feel; what it gets wrong is the assumption that gluten is the reason.

How clients describe it, and what it gets confused with

Clients say "I'm gluten-free," "I'm off gluten," "gluten doesn't agree with me," "I'm gluten intolerant," "bread makes me bloated," "I'm cutting out wheat," "my doctor said I have celiac," or "I'm going gluten-free to lose weight." Each of these can mean something different, from a diagnosed disease to a hunch to a weight-loss plan. Ask before assuming.

The diet is most often confused with four others. A wheat-free diet is the main management of wheat allergy, a different condition from celiac disease, and is defined by the allergy rather than by gluten (Cianferoni, 2016). The low-FODMAP diet, which has its own archive page, Low-FODMAP Diet, removes wheat for a different reason, because wheat contains fructans, a type of fermentable carbohydrate, and the research on gluten sensitivity keeps pointing back to it (Skodje et al., 2018). Low-carbohydrate and ketogenic diets cut bread and pasta along with other carbohydrate, so a client who says "I don't eat bread anymore" may mean either; the Ketogenic Diet page is also in this archive. And gluten-free, casein-free eating, which removes milk protein as well, is a combination some families try for a child with autism (Quan et al., 2022).

2. In practice

A typical day of eating

For a person with celiac disease following the plan a dietitian set out: breakfast is certified gluten-free oatmeal with berries and milk, or eggs with corn tortillas and salsa. Lunch is a rice and bean bowl with roasted vegetables and chicken, or a salad with quinoa, chickpeas and a vinegar and oil dressing made without malt vinegar. A snack is fruit with peanut butter, or yogurt. Dinner is baked fish or a stir-fry made with gluten-free tamari rather than ordinary soy sauce, with potatoes or rice and a vegetable. Most of this is ordinary food that happens to contain no gluten. What makes the day work is the checking: labels, the toaster, the restaurant's fryer, and the question of what a sauce is thickened with.

A person following the diet by choice often eats much the same way, or swaps wheat bread, pasta and crackers for the gluten-free versions on the store shelf. The second approach changes the label on the food more than the food itself.

Cost, products, and who profits

Naturally gluten-free foods, such as rice, potatoes, corn, beans, eggs, fruits and vegetables, cost no more on this diet than on any other. The gluten-free versions of wheat foods cost much more. In a United States market basket study that priced 10 categories of foods, including bread, pasta, cereal, crackers, cookies, pizza and waffles, in 5 regions and online in late 2016, the gluten-free products cost on average 183 percent more per ounce than their wheat-based counterparts. The gap was 240 percent in the same study design 10 years earlier, after adjusting for inflation, and it varied by region, from 162 percent in Atlanta to 245 percent in Rapid City, South Dakota (Lee et al., 2019).

Availability varied too, and products were easiest to find online and on the coasts (Lee et al., 2019). In a trial in irritable bowel syndrome, participants found ordinary dietary advice cheaper and less time-consuming to shop for than either a gluten-free or a low-FODMAP diet (Rej et al., 2022).

The diet depends on label reading, time to cook from naturally gluten-free foods, and, for celiac disease, a separate set of kitchen tools or careful cleaning to avoid cross-contact. Gluten-free foods are a large retail category: a review reported more than 15.5 billion dollars in United States retail sales in 2016 (Niland & Cash, 2018). The companies that make gluten-free products profit when a person without celiac disease adopts the diet, which is worth knowing because those products, not naturally gluten-free foods, carry most of the marketing a client has seen.

For someone with limited money or store access, the cheaper route is the naturally gluten-free one: rice, corn tortillas, potatoes, beans, eggs and frozen vegetables. For a person with celiac disease, cost is part of the burden of treatment, which guideline authors and researchers both describe as substantial (Lee et al., 2019; Rubio-Tapia et al., 2023).

Environmental footprint

One Australian modeling study compared a gluten-free menu with the average Australian diet and other menus. Its emissions estimates depended on the particular foods selected, especially meat and dairy, and do not establish the footprint of every gluten-free diet. Removing gluten does not specify what replaces it (O'Brien et al., 2025).

The study built a 7-day gluten-free meal plan for a 71-year-old man from a national celiac society's sample menus and estimated its climate footprint with an Australian food emissions calculator: 2.51 kilograms of carbon dioxide equivalents a day, against 2.38 for the average Australian diet in the same analysis. Beef and dairy foods made up most of the gluten-free menu's footprint and grains less than 1 percent, and replacing beef with lamb and cow's milk with soy milk lowered the estimate by 29 percent. The calculator leaves out packaging and waste, and the authors note that the results apply only to the Australian food system and to that reference person (O'Brien et al., 2025). The searches for this page did not identify a study that estimated the land use or water use of a gluten-free diet.

Cultural fit

The diet fits most easily with food traditions whose staples are already gluten-free, such as those built on rice, corn, potatoes, beans or cassava. It asks more of traditions built around wheat bread, pasta, flour tortillas, dumplings, noodles or baked goods, where the familiar staple has to be replaced rather than kept. It can be adapted without losing what defines it by keeping the dish and changing the grain: corn instead of flour tortillas, rice noodles instead of wheat noodles, a gluten-free flour blend for a family recipe. For a person with celiac disease the harder part is often social rather than culinary. The European guideline notes that the lifelong restriction can bring frustration, social isolation and distress, particularly where gluten-free options are scarce or a person feels left out of shared meals and cultural traditions (Al-Toma et al., 2026).

What makes people stop

For people with celiac disease, the European guideline lists factors linked to lower adherence: younger age, lower socioeconomic status, eating outside the home, having no symptoms, and limited knowledge of the diet (Al-Toma et al., 2026). In the Finnish trial described in section 4, which found people with celiac disease through screening, social functioning scores favored the group that kept eating gluten, while digestive symptoms and anxiety improved more in the group that went gluten-free; after 2 years, 13 percent found the diet difficult and 92 percent reported sticking to it (Kurppa et al., 2014). In the trial in irritable bowel syndrome, ordinary dietary advice was easier to follow when eating out than a gluten-free diet (Rej et al., 2022).

Money matters too. In a Dutch survey of 548 adults with celiac disease or non-celiac gluten sensitivity who followed the diet, 23.2 percent were food insecure, and food insecurity went with poorer diet quality and with more difficulty around cooking skills, social situations, resources and access to naturally gluten-free foods (Smeets et al., 2024).

For people following the diet by choice without any gluten-related condition, the searches for this page did not identify a study of why they stop. What follows is observation rather than measured data: cost, the inconvenience of eating out and at other people's homes, and a sense that the diet did not deliver what was hoped are the reasons a coach is most likely to hear.

What happens when you stop

For a person with celiac disease, stopping brings the disease back. The diet has to be followed for life to keep symptoms and intestinal damage from returning, and symptoms can come back with small amounts of gluten (NIDDK, n.d.-c; NIDDK, n.d.-d). This is not a diet a person with celiac disease can take a break from.

For a person without a diagnosis who went gluten-free before being tested, stopping matters for a different reason. If celiac disease is still a question, the person needs to eat gluten again for a period, called a gluten challenge, before testing can be trusted. The European guideline suggests at least 3 grams of gluten a day for 6 weeks, and recommends a dietitian's help for anyone who finds the challenge hard (Al-Toma et al., 2025).

For people who believed gluten upset them, one small Finnish study followed up adults 25 years after celiac disease and wheat allergy had been ruled out and they had been advised to go back to a normal diet: 28 of the 76 took part, half of them were still avoiding gluten-containing cereals, none had been diagnosed with celiac disease or another digestive disease, and as a group they reported more digestive symptoms and poorer well-being than healthy comparison adults (Salmela et al., 2025). The searches for this page did not identify a study that followed people without any gluten-related complaint after they stopped the diet.

Common variations

The strict medical version for celiac disease keeps gluten below about 10 milligrams a day, includes only certified gluten-free oats, and manages cross-contact (Al-Toma et al., 2026). A wheat-free diet for wheat allergy removes wheat but is defined by the allergy rather than by gluten (Cianferoni, 2016). A low-gluten diet, which reduces gluten without removing it, has been tested in healthy adults (Hansen et al., 2018). The gluten-free, casein-free diet removes milk protein too and is tried mostly for children with autism (Quan et al., 2022). A gluten-reduced diet as part of a low-FODMAP plan happens because wheat is high in fructans; the Low-FODMAP Diet page is in this archive. And in the popular version, "gluten-free" can mean anything from a strict diet to swapping a sandwich for a salad at lunch.

3. For the coach

SCOPE OF PRACTICE

A health and wellness coach does not prescribe, adjust, endorse or discourage a gluten-free diet, and does not diagnose or treat. The coach helps the client understand what they have chosen, notice how it is going, and decide what to do next.

The medical use of this diet belongs to the physician and the registered dietitian. Celiac disease is diagnosed by blood tests and, for most people, a biopsy of the small intestine, taken while the person is still eating gluten; the diet that follows is prescribed and monitored by them (Al-Toma et al., 2025; Rubio-Tapia et al., 2023). Wheat allergy is diagnosed and managed by an allergist. Irritable bowel syndrome is diagnosed by a physician, and any diet for it is guided by a dietitian.

The point at which this diet most needs a clinician comes before it starts. A client who has digestive or other symptoms and is thinking of going gluten-free to see whether it helps is describing a decision that can make celiac testing inaccurate. Federal guidance is that testing comes first, and the coach does not suggest a trial without gluten (NIDDK, n.d.-b; NIDDK, n.d.-c).

Who does what

The physician

Usually a primary care physician or gastroenterologist. Tests for celiac disease while the person is still eating gluten, most often with a blood test for tissue transglutaminase antibodies and a check of total IgA, and confirms the diagnosis with a biopsy of the small intestine in most adults; orders a gene test for HLA-DQ2 and HLA-DQ8 when someone has already stopped eating gluten, since a negative result rules celiac disease out; and decides whether a gluten challenge is needed (Al-Toma et al., 2025; Rubio-Tapia et al., 2023). After diagnosis, follows antibody levels, nutrient levels, bone density, thyroid function and vaccinations, and works up symptoms that continue despite the diet (Al-Toma et al., 2026; Rubio-Tapia et al., 2023). An allergist diagnoses wheat allergy with skin or blood tests and food challenges and prescribes emergency epinephrine where needed (Cianferoni, 2016).

The registered dietitian

For celiac disease, a dietitian with expertise in the diet is a core part of care. The American College of Gastroenterology calls a visit with a dietitian after diagnosis mandatory, and the European guideline gives the dietitian the work of teaching the lifelong diet, assessing nutrition at diagnosis, correcting deficiencies, checking adherence, finding hidden gluten when symptoms continue, and supporting a gluten challenge when one is needed (Al-Toma et al., 2025; Al-Toma et al., 2026; Rubio-Tapia et al., 2023). For irritable bowel syndrome and suspected sensitivity, the dietitian guides any trial of a restricted diet, including a low-FODMAP plan, and the reintroduction that follows (Al-Toma et al., 2026; Rej et al., 2022).

The health and wellness coach

Helps the client see their own reasons for the diet and what they hope it will do. For a client with diagnosed celiac disease, works on the life around the prescription: planning meals and groceries on a budget, handling restaurants, travel, work events and family gatherings, keeping a shared kitchen workable, and noticing when the diet is taking a toll on mood, social life or the person's relationship with food. For a client without a diagnosis, explores what is driving the interest, and, with the client's permission, can share general guidance from recognized authorities, such as the federal advice to be tested before starting the diet, and explore what the client wants to do with it. Holds the referral line when a question belongs to a clinician.

What a client bringing this diet needs

Clients bring this diet for very different reasons, and the reason shapes the conversation. A client newly diagnosed with celiac disease is learning to live with a lifelong medical diet and may feel overwhelmed, grieving familiar foods, or anxious about eating anywhere they do not control. A client who has been told by a friend or a website that gluten causes their bloating, fatigue or brain fog is often hoping for a simple explanation of symptoms that have bothered them for a long time. A client trying to lose weight may have picked up gluten-free eating as one more rule.

What the coaching conversation can offer that a search engine cannot is room to sort these apart. For the first client, the coach supports a plan the medical team has already made. For the second, the most useful thing may be helping the client see that their symptoms deserve a proper medical look, and that the order of steps matters. For the third, it is a chance to explore what the client wants for their health more broadly, without endorsing or dismissing the diet.

What to ask

• What led you to think about going gluten-free?
• What are you hoping will be different if you do?
• Who else knows about the symptoms you have been having, and what have they suggested?

Referral red flags

A client who has digestive symptoms, unexplained tiredness, anemia, a rash, or a family member with celiac disease, and who is thinking of going gluten-free, should talk with a physician about testing before changing their diet, because a gluten-free diet can affect the test results (NIDDK, n.d.-b; NIDDK, n.d.-c). The same applies to a client who has already gone gluten-free and now wonders whether they have celiac disease: the physician decides whether a gene test or a gluten challenge is needed, and the coach does not suggest the client start eating gluten again on their own (Al-Toma et al., 2025).

A client with diagnosed celiac disease whose symptoms continue or come back despite following the diet goes back to the physician and dietitian, who look for hidden gluten and for other causes (Al-Toma et al., 2026; NIDDK, n.d.-e). A client who describes hives, swelling of the lips or throat, wheezing or trouble breathing after eating wheat, or after eating wheat and then exercising, is describing a possible allergic reaction; a severe reaction needs emergency help, and the client needs an allergist (Cianferoni, 2016). Weight a client loses without meaning to is a reason to involve a clinician, since a physician checking for celiac disease looks for signs of weight loss (NIDDK, n.d.-b). It is never a sign that the diet is working.

A client whose list of avoided foods keeps growing beyond gluten, who is frightened of eating anything they did not prepare, or who has stopped eating with other people is worth asking about directly: what each restriction is for, whether it causes distress, whether it is narrowing what they eat, and whether it is getting in the way of daily life. There are ordinary reasons for strict avoidance, and celiac disease is one of them. Where the answers raise a real concern about disordered eating, the coach follows their own referral process, or tells the client's clinician under an existing care plan.

Special populations

Pregnancy and breastfeeding

For women with celiac disease, the European guideline notes that, although evidence is limited, celiac disease may raise the risk of complications such as preterm birth and low birth weight, that keeping to the diet has been associated with better reproductive outcomes, and that coordinated care before and during pregnancy, with attention to folate and other nutrients, is good practice (Al-Toma et al., 2026). For women without celiac disease, the searches for this page did not identify a randomized trial in pregnancy. A Canadian observational pregnancy cohort included women reporting a gluten-free diet without celiac disease and found no clear differences in the outcomes it measured; this does not establish the diet's safety or benefit in all pregnancies. Of 2,359 pregnancies, 108 women reported the diet, 68 of them without celiac disease, and small size for gestational age, birth weight, gestational age at birth, neonatal intensive care admission and cesarean birth did not differ clearly; diet and outcomes were self-reported and the group was small (Srajer et al., 2026). A Finnish hospital-record study of 291 women avoiding gluten without a celiac diagnosis and 79 with celiac disease, compared with 456 other women, found no increase in pregnancy complications or in babies small or large for gestational age, although labor was induced more often and lasted longer in the gluten-free group (Reijonen et al., 2025).

One large Danish cohort found that higher gluten intake in pregnancy was associated with a higher rate of type 1 diabetes in the children: a hazard ratio of 1.31, from 1.001 to 1.72, for each additional 10 grams of gluten a day (Antvorskov et al., 2018). That is an association from a food questionnaire, not evidence that going gluten-free in pregnancy prevents anything. None of these studies recommends restricting gluten in pregnancy, and any change of this size in pregnancy belongs with the obstetric clinician and a dietitian.

Children and adolescents

Celiac disease is diagnosed in children with the same blood tests, and in selected children with very high antibody levels a diagnosis can sometimes be made without a biopsy (Rubio-Tapia et al., 2023). In children, untreated celiac disease can show up as slowed growth, delayed puberty, weight loss or failure to gain weight as expected (NIDDK, n.d.-d). The searches for this page did not identify a trial of a gluten-free diet in otherwise healthy children. A randomized trial in 59 young people, median age 11, with islet autoantibodies, a sign of high risk for type 1 diabetes, found no differences in glucose tolerance between a gluten-free and a normal diet over 17 months (Maziarz et al., 2025). For children with autism, a review combined 8 small randomized trials: 5 tested removal of gluten and casein, a milk protein, and 3 removed gluten alone. Its combined results cannot show the separate effect of either diet, only 1 trial stated that the children were blinded, and the authors called for larger, better trials (Quan et al., 2022). A child's diet belongs to the family and the pediatrician, with a dietitian for any restriction.

Older adults

Celiac disease can be diagnosed at any age. The European guideline notes that people diagnosed when older often have symptoms other than the classic ones, such as iron-deficiency anemia, tiredness and bloating, which can delay diagnosis; they carry a higher risk of complications such as osteoporosis; and although their response to the diet may be slower, it still brings substantial benefit (Al-Toma et al., 2026). The searches for this page did not identify a study of the gluten-free diet in older adults without celiac disease.

Athletes and people with high activity

Gluten-free eating is common among athletes. In an online survey of 910 athletes without celiac disease, 41 percent said they followed the diet at least half the time (Lis, Stellingwerff, Shing, et al., 2015). In a small blinded crossover trial, 13 competitive cyclists without celiac disease ate a gluten-free diet for 7 days with food bars that did or did not contain gluten; the trial did not demonstrate a clear difference in time-trial performance, digestive symptoms, well-being or blood markers between the 2 diet periods (Lis, Stellingwerff, Kitic, et al., 2015). A later 6-week pilot study that was not randomized also did not demonstrate a performance difference (Zdzieblik et al., 2025). Separately, wheat is a frequent trigger of a rare allergic reaction that occurs only when exercise follows eating, called wheat-dependent exercise-induced anaphylaxis (Cianferoni, 2016); an athlete who describes hives or breathing trouble during exercise after a meal needs medical attention.

People taking medication for diabetes, blood pressure, or blood thinning

For a person with both celiac disease and type 1 diabetes, the European guideline calls for a plan that fits the gluten-free diet together with carbohydrate counting, with care shared among the gastroenterologist, endocrinologist and dietitians (Al-Toma et al., 2026). Any client taking insulin, a sulfonylurea, blood pressure medication or warfarin who plans a substantial change in eating keeps the prescriber informed before making it, which is the client keeping the prescriber in the loop rather than asking permission for every food choice. A possible low blood sugar episode calls for immediate action under the client's own diabetes plan, and severe symptoms, or an episode the person cannot manage, call for emergency help; the client also tells the prescriber about the episode. Thyroid medication deserves its own mention: celiac disease can reduce how well levothyroxine is absorbed, and starting the diet often improves absorption, so thyroid levels and dosing need close monitoring by the prescriber (Saad & Maisch, 2026). That is a question for the prescriber, and section 6 says more.

People with a history of disordered eating

The European guideline reports strong evidence that people with celiac disease have a higher risk of eating disorders, a risk that is higher in women and in people diagnosed in adolescence or early adulthood, and it links this to the strict requirements of the diet, constant vigilance around food and anxiety about accidental gluten (Al-Toma et al., 2026). In a Swedish national study of women, celiac disease and anorexia nervosa were associated in both directions: anorexia was diagnosed more often after celiac disease, and a prior anorexia diagnosis was more common in women later found to have celiac disease (Mårild et al., 2017). For a person with a history of disordered eating who is choosing a gluten-free diet without a medical reason, the diet adds a rule-based restriction whose health benefit the studies reviewed here do not establish. Distress, narrowing and interference with daily life are what raise a concern, and a concern goes through the coach's own referral process or to the client's clinician under an existing care plan.

Client misconceptions

Three come up most. The first is that gluten-free eating is healthier for everyone or helps with weight loss. The federal agency for digestive diseases states that researchers have found no evidence that a gluten-free diet promotes better health or weight loss for the general population (NIDDK, n.d.-c).

The second is that feeling better without wheat proves a person has a problem with gluten. In blinded challenges, only a minority of people who believe they are sensitive to gluten met the trials' criteria for symptoms specific to gluten, symptoms also occurred with placebo, and in one trial fructans, a carbohydrate in wheat, raised symptoms more than gluten did (Molina-Infante & Carroccio, 2017; Skodje et al., 2018). The third is that it is fine to try the diet first and get tested later. Going gluten-free first can make celiac tests inaccurate and may mean eating gluten again for weeks before testing (Al-Toma et al., 2025; NIDDK, n.d.-c). Each is taken up in section 6.

🤝 COACHING CONNECTION

Megan has had bloating and loose stools on and off for a year, and a coworker told her that going gluten-free fixed the same problem. She tells the coach she plans to cut out gluten for a month to see. The coach asks what she has noticed about when the symptoms come and go, and who else knows about them. Megan says she has not mentioned them to her doctor because they did not seem serious enough. The coach asks whether she would like to hear what a federal health agency says about testing before starting a gluten-free diet, and she says yes.

The coach shares that the agency advises talking with a doctor about celiac testing first, because avoiding gluten beforehand can make the results inaccurate, and asks what she makes of that. Megan decides to book an appointment before changing anything, and to keep a simple record of her symptoms to bring with her. The coach did not suggest a diet, and did not guess at a cause.

4. The evidence

(select the heading to open or close this section)

The evidence has to be read one group at a time. For celiac disease, a strict gluten-free diet is the established treatment. A European guideline panel rated the evidence behind that recommendation moderate in certainty, and a British panel rated the evidence behind its matching recommendation low; both drew mainly on observational studies, and one randomized trial in people found by screening adds to them (Al-Toma et al., 2026; Kurppa et al., 2014; Penny et al., 2026).

For people without celiac disease, the trials are smaller and point the other way: in blinded tests, only a minority of people who believe they are sensitive to gluten met the trials' criteria for gluten-specific symptoms, and the evidence in irritable bowel syndrome is very low in certainty. The large cohort studies of gluten intake in people without celiac disease did not find that eating less gluten went with less heart disease or diabetes.

Evidence from randomized controlled trials

A trial assigns people to a diet by chance and compares them with people assigned to something else, which is the strongest way to test whether the diet itself causes a result. Trials of this diet fall into 3 groups that answer different questions: a trial in people with celiac disease, trials that test whether gluten itself causes symptoms in people who believe they are sensitive to it, and trials in other conditions.

Celiac disease

Because the diet is the accepted treatment, trials that withhold it from people diagnosed with celiac disease are rare. The one identified for this page used a group for whom the question was still open. In Finland, researchers screened 3,031 relatives of people with celiac disease for celiac antibodies. Of the 148 who tested positive, 40 adults who considered themselves free of symptoms were randomly assigned to go gluten-free or to keep eating gluten for 1 year, with a dietitian counseling everyone. The main outcome was the shape of the lining of the small intestine, measured on biopsies that were not read until the study ended, so the people reading them could not know who had eaten what (Kurppa et al., 2014).

After 1 year, the lining improved in the gluten-free group compared with the gluten group, antibody levels fell, and digestive symptoms, indigestion, reflux and anxiety improved more, as did people's own rating of their health. Social functioning scores favored the group that kept eating gluten. The paper describes the direction of social functioning inconsistently: one sentence in its results says the score fell in the gluten group, while its figure, abstract and discussion show the higher score in the group that kept eating gluten. Laboratory values, bone density and body mass index did not differ clearly between the groups. When the gluten group then went gluten-free for a year, most measures improved in them too (Kurppa et al., 2014). The participants could not be blinded to their diet, and the authors note that expectation may have shaped the symptom results; the biopsy result does not depend on what participants believed.

Two current guidelines make the same strong recommendation and attach different certainty labels to it. The European Society for the Study of Coeliac Disease panel rates the evidence behind its recommendation of a strict, lifelong gluten-free diet Moderate (Al-Toma et al., 2026). The British Society of Gastroenterology's 2026 guideline rates the evidence behind its recommendation that adults with celiac disease keep a lifelong gluten-free diet Low, and notes that no randomized placebo-controlled trial has evaluated the diet's long-term health impact (Penny et al., 2026).

Both rest mainly on observational studies. The British panel's supporting table rates each outcome it considered from non-randomized studies: quality of life, fractures, metabolic and liver outcomes, mucosal healing and nutrition each Low, with serious risk of bias and a strong association, and cancers of the lymphatic system Very low, with serious risk of bias and serious imprecision. The European panel's table gives symptom resolution Moderate and labels its other 3 outcomes Low, but for those 3 the circle symbols printed beside the labels disagree with the words. The 2 recommendation labels are reported as each panel published them and are not averaged.

Non-celiac gluten or wheat sensitivity

Non-celiac gluten or wheat sensitivity describes digestive and other symptoms after eating wheat-containing food, in a person who does not have celiac disease or wheat allergy, that improve without it and return with it. The European guideline now prefers the name non-celiac wheat sensitivity, because components of wheat other than gluten, including fructans and proteins called amylase-trypsin inhibitors, may cause the symptoms, and it cites a global prevalence estimate of 0.6 to 13 percent (Al-Toma et al., 2025).

A 2026 meta-analysis of 25 population surveys, with 49,476 people in 16 countries, put the share who report symptoms they attribute to gluten or wheat at 10.3 percent, from 7.0 to 14.0; about 40 percent of them followed a gluten-free diet, bloating was the most common symptom, and self-reported sensitivity went with psychological distress and irritable bowel syndrome (Shiha et al., 2026). That figure counts people who believe they react, not people in whom a reaction has been confirmed.

There is no blood test for non-celiac gluten or wheat sensitivity, so the condition is diagnosed by ruling out celiac disease and wheat allergy while the person is still eating gluten, and an expert group proposed a blinded challenge, gluten or placebo given in turn for a week each without the person knowing which, as the way to confirm it (Catassi et al., 2015).

When that test has been used in research, it has confirmed the condition in few people. A review of 10 blinded, placebo-controlled gluten challenge trials in 1,312 adults found that gluten raised average symptom scores more than placebo in 7 trials. Only 4 trials reported how individual people responded. In those 4 trials, the review classified 38 of 231 participants as having gluten-specific symptoms, symptoms with gluten but not with placebo; the share ranged from 8 to 34 percent between trials, and definitions and challenge methods varied. Symptoms with placebo were common: 94 of 231 reacted as much or more to the placebo. The trials differed widely in dose, length and what the placebo contained (Molina-Infante & Carroccio, 2017). The review gives one trial's result as 15 percent in its table and 17 percent in its discussion; this page uses the table.

Two trials show where the symptoms may come from. In Australia, 37 people with self-reported gluten sensitivity and irritable bowel syndrome improved on a diet low in FODMAPs, a group of fermentable carbohydrates, and then worsened to a similar degree whether gluten or a whey protein placebo was added; only 8 percent showed effects specific to gluten (Biesiekierski et al., 2013). In Norway, 59 people who had gone gluten-free on their own, with celiac disease ruled out, ate muesli bars containing gluten, fructans or neither for 7 days at a time in random order without knowing which. Overall symptom scores averaged 33.1 with gluten, 38.6 with fructans and 34.3 with placebo; fructans raised symptoms more than gluten did, and gluten did not differ from placebo. Thirteen people had their worst symptoms with gluten, 24 with fructans and 22 with placebo (Skodje et al., 2018).

Irritable bowel syndrome

Many people with irritable bowel syndrome feel that wheat-based foods make their symptoms worse. A meta-analysis found only 2 randomized trials of a gluten-free diet, with 111 participants, and in both, people who had already responded to the diet were randomized to stay on it or have gluten added back. Global symptoms were lower on the gluten-free diet, but the result was not statistically clear, a relative risk of symptoms remaining of 0.42, from 0.11 to 1.55, and the authors rated the overall quality of the evidence very low (Dionne et al., 2018). The British Society of Gastroenterology, drawing on that review, recommends against a gluten-free diet in irritable bowel syndrome, a weak recommendation based on very low quality evidence, and notes that the benefit people report may come mainly from eating fewer fructans when wheat is removed rather than from removing gluten (Vasant et al., 2021).

A later trial in the United Kingdom randomized adults with irritable bowel syndrome without constipation to traditional dietary advice, a low-FODMAP diet, or a gluten-free diet for 4 weeks; the gluten-free diet was less strict than a celiac diet and allowed small amounts of cross-contact. Of 101 who began, 99 completed the trial, 33 on each diet. Among those who completed it, a clinically meaningful improvement in symptom scores occurred in 14 of 33, 42 percent, with traditional advice, 18 of 33, 55 percent, with the low-FODMAP diet and 19 of 33, 58 percent, with the gluten-free diet, a difference that was not statistically clear. Participants found traditional advice cheaper, quicker to shop for and easier when eating out, and the authors recommended it as the first diet to try (Rej et al., 2022). The trial was designed to detect only a large difference between diets.

Cardiometabolic outcomes

The searches for this page did not identify a trial of a gluten-free diet that measured heart attacks, strokes or other cardiovascular events. A Cochrane review of gluten-reduced and gluten-free diets for preventing cardiovascular disease in the general population, searching to June 2021, found 1 randomized trial: 60 healthy adults in Italy assigned to a gluten-free or a usual diet for 6 months. In the review's summary table, the trial did not show a clear difference in systolic blood pressure, a mean difference of 6.9 mmHg lower, from 17.1 lower to 3.3 higher, which the review rated very low in certainty, and the review reports no clear difference in diastolic blood pressure or LDL cholesterol (Schmucker et al., 2022).

In the Danish crossover trial described under body weight below, an 8-week low-gluten diet in healthy adults did not clearly change measures of blood sugar or blood fats compared with a high-gluten diet (Hansen et al., 2018). In a 4-week study without a comparison group, 27 healthy volunteers on a strict gluten-free diet had lower white blood cell counts and C-reactive protein, but no overall improvement in cardiovascular measures or quality of life (Lange et al., 2024). The cohort studies in the next part of this section carry the longer-term evidence.

Cancer

The searches for this page did not identify a trial of a gluten-free diet that measured cancer. For people with celiac disease, the evidence on cancer is observational and is described below.

Body weight, as measured

This archive reports body weight as something a study measured, in that study's own numbers, and not as a goal or a measure of whether an eating pattern worked; see the standing rule in the review record at the foot of this page. The one exception is weight a person loses without meaning to, which can signal a health problem and is a reason to involve a clinician, never a sign that a diet worked.

In the Finnish trial in people with celiac disease, changes in body mass index and body fat did not differ clearly between the gluten-free and gluten groups over a year (Kurppa et al., 2014). In the 6-month Italian trial in healthy adults described above, body mass index did not differ clearly between the gluten-free and usual diets, a mean difference of 0.1 kg/m² lower, from 3.3 lower to 3.1 higher, as the Cochrane review reports (Schmucker et al., 2022).

Small, short studies have measured body weight and metabolic markers in people without celiac disease, including people with metabolic conditions. Their dietary changes, populations and methods differ, and they do not establish that removing gluten improves long-term health in the general population. The studies are described here as measurements, in each study's own numbers.

In a Danish crossover trial, 60 healthy middle-aged adults were randomized to eat a low-gluten diet, about 2 grams of gluten a day, and a high-gluten diet, about 18 grams a day, for 8 weeks each. Body weight was on average 0.8 kilograms lower after the low-gluten period than after the high-gluten period, with no difference in reported energy intake; the main purpose of the trial was to study gut bacteria, the low-gluten diet was not gluten-free, and the two diets also differed in the kinds of fiber they contained (Hansen et al., 2018).

In Iran, 50 adults with metabolic syndrome were randomized to a diet with less than 2 grams of gluten a day, supported with gluten-free bread, or to their usual diet for 8 weeks, and 45 finished. Change in body weight did not differ clearly between the groups, while fasting blood sugar and triglycerides fell more in the reduced-gluten group; participants knew which diet they were on. The paper reports a greater reduction in waist circumference, but its displayed baseline and follow-up means do not show that direction; the discrepancy is unresolved (Ehteshami et al., 2018). By this page's definition that diet was gluten-reduced rather than gluten-free.

In a second Iranian trial, 60 adults with nonalcoholic fatty liver disease received lifestyle advice, which in both groups meant a low-carbohydrate diet and exercise, with or without a gluten-free diet for 3 months, without blinding. The authors reported lower fasting blood sugar and triglycerides with the added diet, but no clear between-group difference in body mass index. Adherence was assessed through telephone self-report, and the report contains inconsistent entries in its outcome tables (Jazayeri et al., 2026).

In Pakistan, 108 adults described as obese and with type 2 diabetes were randomized to a gluten-free or a standard diet of the same calorie range for 12 weeks. Body mass index went from 30.8 to 27.2 in the gluten-free group and from 30.6 to 29.8 in the standard group, and blood pressure, blood sugar, cholesterol and liver enzymes improved within the gluten-free group. The paper compared each group with its own starting values rather than testing the difference between the groups, so it does not show that one diet did better than the other. The gluten-free group also had more education, more employment and more awareness of the diet at the start, which the authors note could explain part of the result, and the paper's text and table give different figures for that awareness (Naeem et al., 2026).

Mortality

The searches for this page did not identify a trial of a gluten-free diet that measured death from any cause or from a specific cause.

Other conditions people use it for

Schizophrenia. A small group of people with schizophrenia have raised antibodies to gliadin, a part of gluten. In a 5-week inpatient trial, 43 people with schizophrenia or a related disorder and raised antigliadin antibodies were randomized, and 39 were analyzed. All ate a gluten-free diet on a research unit, and twice a day they drank a protein shake mixed with identical-looking flour that was either gluten flour or certified gluten-free rice flour; participants, researchers and the clinical team did not know which (Kelly et al., 2026).

On the trial's main outcome, the motivation and pleasure component of negative symptoms (CAINS-MAP), scores improved modestly more in the gluten-free group over 5 weeks; broader negative-symptom and other symptom scales did not show clear differences. Two people in the gluten-free group were assessed after 4 weeks rather than 5. The authors called the results preliminary and asked for larger, longer trials (Kelly et al., 2026). This is a specific subgroup in a hospital setting, not a treatment for schizophrenia in general, and any such decision belongs with the psychiatric care team.

Autism. A review combined 8 small randomized trials in 297 children with autism: 5 tested removal of gluten and casein, a milk protein, and 3 removed gluten alone. The review reported pooled improvements in stereotyped behavior, a standardized mean difference of 0.41 in the direction of improvement, from 0.15 to 0.68, in 5 trials, and in cognition, 0.46, from 0.01 to 0.91, in 3 trials, with no clear change in other symptom groups. These are the review's estimates for this mixed set of diets, and they do not establish the separate effect of either diet. The cognition result depended on one trial that removed both proteins. Only 1 of the 8 trials stated that the children were blinded to their diet, and 4 trials lasted less than 4 months (Quan et al., 2022).

Hashimoto's thyroiditis. Some people with this autoimmune thyroid condition, without celiac disease, go gluten-free hoping to calm the thyroid. A 2025 meta-analysis pooled 3 randomized trials with 110 participants, all women. A gluten-free diet did not clearly change thyroid-stimulating hormone or thyroid hormone levels; one thyroid antibody fell and another rose; and the authors rated the certainty of the evidence very low for every outcome, because of high risk of bias in 2 of the 3 trials and very wide intervals (Araújo et al., 2025). Thyroid treatment belongs with the prescriber.

Athletic performance. In 13 competitive cyclists without celiac disease, a 7-day gluten-free diet, tested in a randomized, blinded crossover design against the same diet with added gluten, did not demonstrate a clear difference in time-trial performance, digestive symptoms, well-being or markers of intestinal injury and inflammation (Lis, Stellingwerff, Kitic, et al., 2015). A later 6-week pilot study in 20 male endurance athletes doing sprint interval training was not randomized: the athletes were matched on fitness and allocated alternately, and 15 finished, 6 of the 10 in the gluten-free group. Both groups covered more distance in a 60-minute run, with no clear difference between them; weight fell more in the gluten-free group, which the authors attribute to that group's higher starting weight and lower reported energy intake rather than to the diet (Zdzieblik et al., 2025). Neither study supports a claim that the diet improves athletic performance.

Eating behavior and wellbeing

In the Finnish trial, anxiety and people's own rating of their health improved more on the gluten-free diet, while social functioning scores favored continued gluten (Kurppa et al., 2014). The European guideline describes psychological distress, including anxiety and depression after diagnosis, as common in celiac disease, and judges that psychological support may help the people who are struggling, at low certainty (Al-Toma et al., 2026).

Evidence from long-term prospective cohort studies

A cohort study follows people who already eat a certain way and records what happens to them. It can show that a pattern goes with an outcome; it cannot by itself show that the pattern caused it. The two large cohort studies below measured gluten intake, estimated from food questionnaires, in people without celiac disease, rather than studying people on a gluten-free diet. The people who ate the least gluten were not avoiding it, so these studies say more about gluten in an ordinary diet than about the effects of going gluten-free.

Cardiometabolic outcomes

In 64,714 women and 45,303 men without celiac disease in 2 United States cohorts of health professionals, followed for 26 years, 6,529 developed coronary heart disease. After adjustment for known risk factors, people in the highest fifth of estimated gluten intake had a hazard ratio for coronary heart disease of 0.95, from 0.88 to 1.02, compared with the lowest fifth, so no clear association was detected. When the analysis was adjusted for refined grains, which leaves the remaining variation in gluten intake tied mostly to whole grains, higher gluten intake went with lower risk, a hazard ratio of 0.85, from 0.77 to 0.93. The authors concluded that avoiding gluten may reduce whole grain intake, and that gluten-free diets should not be promoted to people without celiac disease to prevent heart disease (Lebwohl et al., 2017).

The Cochrane review described above, which included these 2 cohorts, rated the evidence that lower compared with higher gluten intake is not associated with non-fatal heart attack low in certainty, a hazard ratio of 0.99, from 0.89 to 1.10, in its summary table (Schmucker et al., 2022).

In 3 of the same cohorts, 71,602 women, 88,604 women and 41,908 men followed from the 1980s and 1990s for more than 20 years, 15,947 developed type 2 diabetes. People in the highest fifth of gluten intake had a hazard ratio for type 2 diabetes of 0.80, from 0.76 to 0.84, compared with the lowest fifth, after adjustment; the association weakened but remained, 0.87, from 0.81 to 0.93, after adjustment for cereal fiber. Gluten intake tracked closely with intake of refined grains, starch and cereal fiber, so the authors suggest that limiting gluten may mean eating less cereal fiber and other beneficial nutrients (Zong et al., 2018). The Cochrane review rated the evidence low in certainty that lower gluten intake may be associated with a slightly higher risk of type 2 diabetes, in its summary table (Schmucker et al., 2022).

Cancer

In people with celiac disease, the European guideline judges, at low certainty, that the risk of cancer is only slightly increased, low in absolute terms, and lower with long-term strict adherence to the diet (Al-Toma et al., 2026). The searches for this page did not identify a cohort study of cancer in people without celiac disease who follow a gluten-free diet.

Body weight, as measured

In people with celiac disease, weight changes are common after the diet begins, as eating changes and the healing intestine absorbs food better. In a retrospective United States celiac-center cohort, 679 adults with biopsy-confirmed celiac disease had repeated body mass index measurements over an average of 39.5 months. Mean body mass index rose from 24.0 to 24.6; 15.8 percent of those initially in low or normal categories moved into overweight or obese categories. Among the 139 initially classified as overweight, 31, or 22.3 percent, had an increase of at least 2 body mass index points. The overall change did not differ clearly between adherence groups, and the first measurement was not always taken at diagnosis: 384 of the 679 had a first measurement within 90 days of it (Kabbani et al., 2012).

A review cited in the European guideline found metabolic syndrome in 4 percent and fatty liver disease in 18.2 percent of people with celiac disease at diagnosis, rising to 21.3 percent and 28.2 percent after the diet began, close to rates in the general population; the guideline recommends a balanced diet, activity and monitoring of metabolic risk factors (Al-Toma et al., 2026).

Mortality

The searches for this page did not identify a cohort study that followed people without celiac disease who chose a gluten-free diet and recorded deaths. Studies of gluten intake come closest. In the Cochrane review's summary table, lower compared with higher gluten intake showed no association with death from cardiovascular disease, a hazard ratio of 1.00, from 0.95 to 1.06, across cohorts from the United States and United Kingdom, rated low in certainty, and the association with death from any cause, 1.00, from 0.99 to 1.01, in 1 United Kingdom cohort, was rated very low in certainty (Schmucker et al., 2022). The review's abstract and detailed table differ in their cohort description; this page follows the detailed table.

For celiac disease, the observational studies discussed in the American College of Gastroenterology guideline compare people whose intestine healed with people whose intestine did not, which is a different question from whether the diet itself lowers death rates, and the guideline treats their results as uncertain (Rubio-Tapia et al., 2023).

Evidence from cross-sectional studies

A cross-sectional study measures diet and an outcome at the same time, so it cannot say which came first. Three kinds of cross-sectional evidence matter for this page.

Who follows the diet. In the United States national health survey, celiac disease affected about 0.7 percent of people aged 6 and older from 2009 to 2014, about 2.0 million people, while the share avoiding gluten without celiac disease rose to 1.7 percent by 2013 and 2014 (Choung et al., 2017). Wheat allergy is a separate condition: in a national survey of more than 40,000 United States adults, about 0.8 percent reported a wheat allergy with symptoms consistent with an immediate allergic reaction (Gupta et al., 2019).

Toxic metals. Rice, a common replacement for wheat, can contain arsenic. In the same national survey, 73 of 7,471 participants reported a gluten-free diet; after adjustment, one estimate of their urine arsenic was about 1.9 times higher, from 1.3 to 2.6, and their total blood mercury about 1.7 times higher, from 1.1 to 2.4, than in people not on the diet (Bulka et al., 2017). A second survey analysis found higher mercury, lead and urine arsenic in gluten-free participants without celiac disease; the cadmium difference in this subgroup was not statistically clear. The paper's abstract reverses the 2 group sizes for that urine arsenic comparison; its table gives 28 people on the diet and 3,632 not on it (Raehsler et al., 2018). The health effects of these levels are uncertain, and neither study can say whether the diet caused them.

Cost. The market basket study described in section 2 is also a cross-sectional comparison: it priced gluten-free products and their wheat-based counterparts at one point in time (Lee et al., 2019).

Highlighted studies

These 4 studies carry the most weight for this page: a randomized trial in celiac disease, a review of blinded gluten challenges, a blinded trial that separated gluten from fructans, and a large cohort study of gluten intake in people without celiac disease.

On a narrow screen, scroll the table sideways to see every column.

Highlighted studies for the Gluten-Free Diet

Study	Design	Sample size	Duration	Finding	The limitation that matters
Kurppa et al., 2014, Finland	Randomized controlled trial, gluten-free diet against continued gluten, in adults with celiac antibodies found by screening	40 adults, from 3,031 relatives screened	1 year	The lining of the small intestine improved on the gluten-free diet compared with continued gluten, antibodies fell, and digestive symptoms and anxiety improved more; no clear difference in laboratory values, bone density or body mass index	Small, in relatives who felt well, and participants knew their diet; the biopsy result was read blind, but the symptom results could reflect expectation
Molina-Infante & Carroccio, 2017	Review of blinded, placebo-controlled gluten challenge trials	10 trials, 1,312 adults with suspected gluten sensitivity; individual responses reported in 4 trials, 231 adults	Challenges of 1 day to 6 weeks	Gluten raised average symptom scores more than placebo in 7 trials; in the 4 trials reporting individual responses, the review classified 38 of 231 participants as having gluten-specific symptoms, and 94 of 231 reacted as much or more to placebo	The trials differed widely in dose, length and placebo, and some placebos contained fermentable carbohydrate
Skodje et al., 2018, Norway	Randomized, blinded, placebo-controlled crossover trial of gluten, fructans and placebo	59 adults on a self-chosen gluten-free diet, celiac disease ruled out	7 days per challenge	Symptom scores averaged 33.1 with gluten, 38.6 with fructans and 34.3 with placebo; fructans raised symptoms, gluten did not differ from placebo	Short challenges with a single dose of each, in one hospital
Lebwohl et al., 2017, 2 United States cohorts	Prospective cohort studies, analyzed together	110,017 adults without celiac disease, 6,529 cases of coronary heart disease	26 years	Highest against lowest fifth of gluten intake: hazard ratio for coronary heart disease 0.95, from 0.88 to 1.02; lower risk with more gluten once refined grains were accounted for	Observational: it measured gluten intake estimated from questionnaires, not people choosing a gluten-free diet, and participants were health professionals

Head to head

The one trial identified for this page that compared a gluten-free diet directly with other diets randomized adults with irritable bowel syndrome to traditional dietary advice, the low-FODMAP diet or the gluten-free diet for 4 weeks. Response rates among those who completed the trial were 42, 55 and 58 percent, with no clear difference between diets, and traditional advice was the cheapest and easiest to follow (Rej et al., 2022). The Low-FODMAP Diet page carries that diet's own evidence. In the Australian and Norwegian trials described above, reducing FODMAPs or adding fructans changed symptoms more than adding gluten did, which is why researchers increasingly look to fructans rather than gluten to explain many people's reactions to wheat (Biesiekierski et al., 2013; Skodje et al., 2018).

Where the evidence is absent

The searches for this page did not identify a trial of a gluten-free diet that measured heart attacks, strokes, cancer or death, in people with or without celiac disease. For people without a gluten-related condition, the trials identified are short, most from days to a few months, and the cardiometabolic cohorts summarized here measured gluten intake rather than following people who chose the diet, so the long-term effects of going gluten-free by choice are not established in the studies reviewed here. The evidence on body weight, pregnancy, children, athletes and the environmental footprint is described in its own part of this page, with what the searches did and did not find.

Non-celiac gluten or wheat sensitivity has no biomarker, and the blinded challenges used in research differ so much in method that the share of people they confirm varies from trial to trial. The schizophrenia, autism and thyroid trials are small and short. The searches did not identify a study of the diet in older adults without celiac disease. For celiac disease, the evidence that the diet works rests largely on observational studies, and the British guideline notes that no randomized placebo-controlled trial has evaluated the diet's long-term health impact (Penny et al., 2026).

5. Mechanisms

(optional reading; select the heading to open or close this section)

This section explains how the diet is thought to work in the body. You can skip it. Nothing in the coaching sections depends on it.

Established mechanisms

Celiac disease is an immune reaction that only happens in people who carry certain gene variants. Gluten proteins from wheat, barley and rye resist complete digestion, and in a person with the gene variants called HLA-DQ2 or HLA-DQ8, pieces of them are presented to the immune system in a way that triggers inflammation in the lining of the small intestine (Al-Toma et al., 2025; Hansen et al., 2018). Over time the finger-like projections of that lining, the villi, flatten, which is called villous atrophy, and the intestine absorbs nutrients less well (NIDDK, n.d.-a; Rubio-Tapia et al., 2023). The body also makes antibodies against one of its own enzymes, tissue transglutaminase, identified as the target of the autoimmune reaction, which is why a blood test for those antibodies can point to the disease (Caio et al., 2019; Rubio-Tapia et al., 2023).

About 30 percent of people carry DQ2 or DQ8, but only about 3 percent of them develop celiac disease, so the genes are necessary but not enough (NIDDK, n.d.-d). Removing gluten removes the trigger: antibody levels fall within weeks to months, and the lining heals over about 1 year or longer, though in some adults it heals incompletely (Al-Toma et al., 2026).

Wheat allergy works differently. It is a reaction of the allergic part of the immune system, in which antibodies of the IgE type recognize wheat proteins and set off a fast reaction, with symptoms that can include hives, swelling, asthma, vomiting and, at worst, anaphylaxis, usually within minutes to hours (Al-Toma et al., 2025; Cianferoni, 2016). It does not damage the small intestine the way celiac disease does (NIDDK, n.d.-a).

Proposed mechanisms

For non-celiac gluten or wheat sensitivity, the mechanism is not established. The European guideline lists several candidates: gluten itself, fructans, amylase-trypsin inhibitors and wheat germ agglutinins, acting through the innate immune system or through non-immune routes similar to those in irritable bowel syndrome, and it notes that a nocebo effect, symptoms produced by the expectation of harm, has been demonstrated (Al-Toma et al., 2025). The blinded trials in section 4 support fructans as one important trigger (Skodje et al., 2018).

Changes in gut bacteria are another proposed pathway. In the Danish low-gluten trial, the mix of gut bacteria shifted, including fewer of several species that ferment fiber, and breath hydrogen, a sign of fermentation, fell; the authors suggest these changes may come mainly from the different fibers in the two diets rather than from gluten (Hansen et al., 2018). In people with schizophrenia and raised antigliadin antibodies, researchers propose an inflammatory pathway involving tryptophan metabolites, which fell on the gluten-free diet in their trial (Kelly et al., 2026). Each of these is a reason to study the diet further, not evidence that it treats a condition.

6. Risks, contraindications and myths

(select the heading to open or close this section)

Warnings from professional or regulatory bodies

The searches for this page did not identify a regulatory warning against the diet itself. The federal health agency that covers digestive diseases gives two cautions in its own words. On testing: "Doctors don't recommend starting a gluten-free diet before diagnostic testing because a gluten-free diet can affect test results" (NIDDK, n.d.-b). On people without celiac disease: "researchers have found no evidence that a gluten-free diet promotes better health or weight loss for the general population" (NIDDK, n.d.-c). The same agency notes that a person who starts avoiding gluten without advice from a doctor or dietitian may not get enough of nutrients such as fiber, iron and calcium (NIDDK, n.d.-c).

Nutrient gaps

The gaps come mostly from the replacements. A review of the nutritional quality of the diet found it often low in fiber, because many gluten-free products are made from starches and refined flours, and low in vitamin D, vitamin B12 and folate and in iron, zinc, magnesium and calcium; it also found more saturated fat and a higher glycemic load (Vici et al., 2016). The European guideline makes the same point: many gluten-free substitutes are lower in protein, fiber, iron, vitamin B12, folate and vitamin D and higher in saturated fat, sodium and added sugars, and heavy reliance on them can keep deficiencies going. It recommends building the diet on naturally gluten-free foods, choosing fortified products where possible, and ongoing support from a dietitian (Al-Toma et al., 2026).

Whole grains deserve particular mention. Avoiding gluten can mean eating fewer whole grains, which are associated with lower cardiovascular risk, as the authors of the large cohort study of gluten and heart disease point out (Lebwohl et al., 2017). The naturally gluten-free whole grains, such as brown rice, quinoa, buckwheat, millet and certified gluten-free oats, are the usual way to fill that gap (Al-Toma et al., 2026).

A 2025 review of 46 observational studies found that people with celiac disease on a gluten-free diet had a lower risk of several deficiencies, including vitamin D, vitamin B12, calcium and iron, than people with untreated celiac disease; it rated the overall quality of the evidence very low (Russell et al., 2025).

A different kind of concern comes from rice. Two analyses of national survey data found higher levels of arsenic and mercury, and in one analysis lead and cadmium, in people on a gluten-free diet; the authors suggest rice-based products may contribute, and the health effects of these levels are not known (Bulka et al., 2017; Raehsler et al., 2018).

For a person with celiac disease, some deficiencies come from the disease itself rather than the diet. Iron deficiency, low vitamin D and other fat-soluble vitamins, and low zinc and copper are common at diagnosis because the damaged intestine absorbs poorly, which is why nutrition is assessed at diagnosis and followed afterward (Al-Toma et al., 2026).

Who should not follow it without medical supervision

Anyone who suspects celiac disease and has not been tested, because going gluten-free first can make testing inaccurate and may later require a gluten challenge (Al-Toma et al., 2025; NIDDK, n.d.-c). Anyone diagnosed with celiac disease, for whom the diet is a lifelong medical treatment followed with a physician and a dietitian (Rubio-Tapia et al., 2023). Anyone with a wheat allergy, which needs an allergist's diagnosis and an emergency plan (Cianferoni, 2016). Anyone with irritable bowel syndrome, for whom a national gastroenterology guideline does not recommend the diet, and a trial recommended traditional dietary advice as the first diet to try (Rej et al., 2022; Vasant et al., 2021).

For a child, the diet affects nutrition and growth (NIDDK, n.d.-d), and in pregnancy the European guideline stresses attention to folate and other nutrients for women with celiac disease (Al-Toma et al., 2026); a restriction this broad in either group belongs with a dietitian. People with type 1 diabetes need a plan that fits the diet to carbohydrate counting (Al-Toma et al., 2026). People taking thyroid medication need their prescriber involved (Saad & Maisch, 2026). And anyone with a history of disordered eating, for whom a strict rule-based diet adds risk (Al-Toma et al., 2026; Mårild et al., 2017).

Medication interactions

A coach never adjusts medication. The point of knowing these is to know when to send a client back to the prescriber or pharmacist.

Levothyroxine. Celiac disease can reduce absorption of levothyroxine, the usual thyroid hormone replacement, and starting a gluten-free diet often improves absorption, so the prescriber monitors thyroid levels and may change the dose or the formulation (Saad & Maisch, 2026). Thyroid disease is also more common in people with celiac disease, and the European guideline recommends checking thyroid function at diagnosis (Al-Toma et al., 2026).

Insulin and other diabetes medication. For a person with type 1 diabetes and celiac disease, the diet is fitted together with carbohydrate counting under the care team (Al-Toma et al., 2026). Any client on insulin or a sulfonylurea who makes a substantial change in eating keeps the prescriber informed before making it. A possible low blood sugar episode calls for immediate action under the client's own diabetes plan. Severe symptoms, or an episode the person cannot manage, call for emergency help; afterward the client tells the prescriber about the episode.

Gluten in medicines. For a person with celiac disease, the federal agency advises asking a pharmacist about the ingredients of prescription and over-the-counter medicines and supplements, and notes that medicines are rare sources of gluten and, when present, usually in very small amounts (NIDDK, n.d.-e). For a US medicine, the pharmacist can help check the specific manufacturer's ingredients; the FDA food-labeling rule does not by itself establish a medicine's status. The FDA has issued draft guidance on gluten statements in drug labeling, separate from its food rule, and states that most oral medicines contain no gluten or virtually none (U.S. Food and Drug Administration, n.d.-b).

Signs it is going wrong

For a person with celiac disease, digestive symptoms that continue or come back despite the diet are common, in roughly 10 to 20 percent of adults, and need a physician and dietitian to look for hidden gluten and other causes (Al-Toma et al., 2026; NIDDK, n.d.-e). Physical signs a coach or a family member might notice in anyone on the diet include ongoing tiredness, digestive changes, a rash, and weight a person loses without meaning to, which physicians look for when they assess celiac disease (NIDDK, n.d.-b; NIDDK, n.d.-d). In a child, slowed growth or failure to gain weight as expected is a reason to see the pediatrician (NIDDK, n.d.-d).

Hives, swelling or breathing trouble after eating wheat, or during exercise after eating it, are signs of a possible allergic reaction; a severe reaction needs emergency help (Cianferoni, 2016). Each is a reason to involve a clinician, not a diagnosis, and the clinician works out the cause.

Behavior matters as much and has to be read with care, because strict avoidance is exactly what celiac disease requires. What carries more meaning is a list of avoided foods that keeps growing beyond gluten, fear of eating anything the person did not prepare, or turning down meals with others. The European guideline describes how constant vigilance and anxiety about accidental gluten can feed restrictive or unhelpful eating patterns in celiac disease (Al-Toma et al., 2026). The questions are the same in every case: what the behavior is for, whether it causes distress, whether it is narrowing what the person eats, and whether it is getting in the way of daily life. Where the answers raise a real concern about disordered eating, the coach follows their own referral process.

Myths and where they came from

Myth: gluten-free food is healthier, and going gluten-free helps you lose weight

Where it came from: marketing aimed at consumers, social and traditional media, and reports of the benefits of avoiding gluten (Niland & Cash, 2018). What is true: the federal agency for digestive diseases states that researchers have found no evidence that a gluten-free diet promotes better health or weight loss for the general population (NIDDK, n.d.-c), and many gluten-free substitutes are lower in fiber and nutrients and higher in sugar, fat and sodium than the foods they replace (Al-Toma et al., 2026).

Myth: if I feel better without wheat, I must have a problem with gluten

Where it came from: people's real experience of feeling better after cutting out wheat, and the idea of gluten sensitivity. What is true: in blinded challenges, only a minority of people who believe they are sensitive to gluten reacted to gluten and not to a placebo, and in one trial fructans, a carbohydrate in wheat, raised symptoms where gluten did not (Molina-Infante & Carroccio, 2017; Skodje et al., 2018). Feeling better is real; the cause needs a proper medical look, starting with celiac testing while still eating gluten.

Myth: I can try the diet now and get tested later

Where it came from: the reasonable wish to see whether a simple change helps. What is true: avoiding gluten before testing can make celiac tests inaccurate, and confirming the diagnosis afterward may mean a gluten challenge of at least 3 grams a day for 6 weeks (Al-Toma et al., 2025; NIDDK, n.d.-c).

Myth: a little gluten now and then won't hurt someone with celiac disease

Where it came from: the fact that small exposures often cause no immediate symptoms. What is true: the European guideline gives a safe ceiling of no more than 10 milligrams of gluten a day, cites a 3-month study in which 50 milligrams a day damaged the intestine in several participants, and notes that people without symptoms can still develop complications (Al-Toma et al., 2026; NIDDK, n.d.-d).

Myth: a gluten-free label means there is no gluten at all

Where it came from: the wording of the label. What is true: in the United States, a food labeled gluten-free must contain less than 20 parts per million of gluten, an amount too small to cause problems for most people with celiac disease, under a federal rule issued in 2013 (NIDDK, n.d.-c; U.S. Food and Drug Administration, n.d.-a).

7. Key terms

Gluten

A group of proteins found in wheat, barley and rye.

Celiac disease

A lifelong immune condition, in people with certain gene variants, in which gluten damages the lining of the small intestine; the treatment is a strict gluten-free diet.

Non-celiac gluten or wheat sensitivity

Symptoms after eating wheat-containing food in a person who does not have celiac disease or wheat allergy; it has no diagnostic test, and the cause may be gluten, fructans or other parts of wheat.

Wheat allergy

A reaction of the allergic part of the immune system to wheat proteins, often fast and sometimes severe; it is managed by avoiding wheat and having an emergency plan.

Cross-contact

Gluten reaching a gluten-free food through shared equipment, surfaces or preparation.

Tissue transglutaminase antibody test

The main blood test for celiac disease, reliable only while the person is still eating gluten.

Villous atrophy

Flattening of the tiny finger-like projections that line the small intestine, the damage a biopsy looks for in celiac disease.

HLA-DQ2 and HLA-DQ8

Gene variants that almost everyone with celiac disease carries; a negative test makes celiac disease very unlikely.

Gluten challenge

A period of eating gluten again, under a physician's direction, so that celiac testing can be done in a person who had stopped eating it.

Fructans

A type of fermentable carbohydrate found in wheat, one of the group called FODMAPs.

Nocebo response

Symptoms caused by the expectation that something will cause harm, measured in trials as a reaction to a placebo.

Refractory celiac disease

A rare form of celiac disease in which symptoms and intestinal damage continue despite at least 12 months on a strict gluten-free diet.

8. Review questions

1. Explain why this page keeps people with celiac disease separate from everyone else who follows a gluten-free diet. Name the condition the diet treats, and name 2 other conditions, besides celiac disease, that can lead a person to avoid wheat.
2. For readers who open section 4: in the review of blinded gluten challenges, what share of people with suspected gluten sensitivity reacted to gluten and not to the placebo, and what share reacted as much or more to the placebo? What did the Norwegian trial suggest might explain many people's symptoms?
3. A client tells you she has had bloating for months and plans to go gluten-free for a few weeks to see whether it helps. Write what you would ask, and describe what, with her permission, you could share and why the order of steps matters. Say exactly where the referral line is.
4. A client with diagnosed celiac disease tells you he has started avoiding restaurants, friends' homes and anything he did not cook himself, and that his list of foods to avoid now goes well beyond gluten. What questions help you understand what is happening, and when does this become a referral?
5. Where did the belief that a gluten-free diet helps with weight loss come from, and what does the federal agency for digestive diseases say about it? Write the 2 sentences you would use with a client who raises it.

9. References

Al-Toma, A., Branchi, F., Zingone, F., Schiepatti, A., Malamut, G., Canova, C., Rosato, I., Ocagli, H., Trott, N., Elli, L., Popp, A., Gianfrani, C., Auricchio, R., Neefjes-Borst, A., Sanders, D. S., Cellier, C., Mulder, C. J., Bouma, G., Lundin, K. E. A., . . . Schumann, M. (2026). European Society for the Study of Coeliac Disease (ESsCD) 2025 updated guidelines on the diagnosis and management of coeliac disease in adults. Part 2: Management, follow-up, and complex disease courses. United European Gastroenterology Journal, 14(2), Article e70195. https://doi.org/10.1002/ueg2.70195

Al-Toma, A., Zingone, F., Branchi, F., Schiepatti, A., Malamut, G., Canova, C., Rosato, I., Ocagli, H., Trott, N., Elli, L., Popp, A., Gianfrani, C., Auricchio, R., Neefjes-Borst, A., Sanders, D. S., Cellier, C., Mulder, C. J., Bouma, G., Lundin, K. E. A., . . . Schumann, M. (2025). European Society for the Study of Coeliac Disease 2025 updated guidelines on the diagnosis and management of coeliac disease in adults. Part 1: Diagnostic approach. United European Gastroenterology Journal, 13(10), 1855-1886. https://doi.org/10.1002/ueg2.70119

Antvorskov, J. C., Halldorsson, T. I., Josefsen, K., Svensson, J., Granström, C., Roep, B. O., Olesen, T. H., Hrolfsdottir, L., Buschard, K., & Olsen, S. F. (2018). Association between maternal gluten intake and type 1 diabetes in offspring: National prospective cohort study in Denmark. BMJ, 362, Article k3547. https://doi.org/10.1136/bmj.k3547

Araújo, E. M. Q., Coutinho-Lima, C. R. O., Sousa, A. S., Souza, L. M. S., Ramos, H. E., Almeida-Pititto, B., Canto, G. L., & Trevisani, V. F. M. (2025). Effects of gluten-free diet in non-celiac Hashimoto's thyroiditis: A systematic review and meta-analysis. Nutrients, 17(21), Article 3437. https://doi.org/10.3390/nu17213437

Biesiekierski, J. R., Peters, S. L., Newnham, E. D., Rosella, O., Muir, J. G., & Gibson, P. R. (2013). No effects of gluten in patients with self-reported non-celiac gluten sensitivity after dietary reduction of fermentable, poorly absorbed, short-chain carbohydrates. Gastroenterology, 145(2), 320-328.e3. https://doi.org/10.1053/j.gastro.2013.04.051

Bulka, C. M., Davis, M. A., Karagas, M. R., Ahsan, H., & Argos, M. (2017). The unintended consequences of a gluten-free diet [Letter to the editor]. Epidemiology, 28(3), e24-e25. https://doi.org/10.1097/EDE.0000000000000640

Caio, G., Volta, U., Sapone, A., Leffler, D. A., De Giorgio, R., Catassi, C., & Fasano, A. (2019). Celiac disease: A comprehensive current review. BMC Medicine, 17(1), Article 142. https://doi.org/10.1186/s12916-019-1380-z

Catassi, C., Elli, L., Bonaz, B., Bouma, G., Carroccio, A., Castillejo, G., Cellier, C., Cristofori, F., de Magistris, L., Dolinsek, J., Dieterich, W., Francavilla, R., Hadjivassiliou, M., Holtmeier, W., Körner, U., Leffler, D. A., Lundin, K. E. A., Mazzarella, G., Mulder, C. J., . . . Fasano, A. (2015). Diagnosis of non-celiac gluten sensitivity (NCGS): The Salerno experts' criteria. Nutrients, 7(6), 4966-4977. https://doi.org/10.3390/nu7064966

Choung, R. S., Unalp-Arida, A., Ruhl, C. E., Brantner, T. L., Everhart, J. E., & Murray, J. A. (2017). Less hidden celiac disease but increased gluten avoidance without a diagnosis in the United States: Findings from the National Health and Nutrition Examination Surveys from 2009 to 2014. Mayo Clinic Proceedings, 92(1), 30-38. https://doi.org/10.1016/j.mayocp.2016.10.012

Cianferoni, A. (2016). Wheat allergy: Diagnosis and management. Journal of Asthma and Allergy, 9, 13-25. https://doi.org/10.2147/JAA.S81550

Dionne, J., Ford, A. C., Yuan, Y., Chey, W. D., Lacy, B. E., Saito, Y. A., Quigley, E. M. M., & Moayyedi, P. (2018). A systematic review and meta-analysis evaluating the efficacy of a gluten-free diet and a low FODMAPs diet in treating symptoms of irritable bowel syndrome. The American Journal of Gastroenterology, 113(9), 1290-1300. https://doi.org/10.1038/s41395-018-0195-4

Ehteshami, M., Shakerhosseini, R., Sedaghat, F., Hedayati, M., Eini-Zinab, H., & Hekmatdoost, A. (2018). The effect of gluten free diet on components of metabolic syndrome: A randomized clinical trial. Asian Pacific Journal of Cancer Prevention, 19(10), 2979-2984. https://doi.org/10.22034/APJCP.2018.19.10.2979

Gupta, R. S., Warren, C. M., Smith, B. M., Jiang, J., Blumenstock, J. A., Davis, M. M., Schleimer, R. P., & Nadeau, K. C. (2019). Prevalence and severity of food allergies among US adults. JAMA Network Open, 2(1), Article e185630. https://doi.org/10.1001/jamanetworkopen.2018.5630

Hansen, L. B. S., Roager, H. M., Søndertoft, N. B., Gøbel, R. J., Kristensen, M., Vallès-Colomer, M., Vieira-Silva, S., Ibrügger, S., Lind, M. V., Mærkedahl, R. B., Bahl, M. I., Madsen, M. L., Havelund, J., Falony, G., Tetens, I., Nielsen, T., Allin, K. H., Frandsen, H. L., Hartmann, B., . . . Pedersen, O. (2018). A low-gluten diet induces changes in the intestinal microbiome of healthy Danish adults. Nature Communications, 9(1), Article 4630. https://doi.org/10.1038/s41467-018-07019-x

Jazayeri, M., Zarrin, R., Mohammadi, A., Pashaei, M. R., & Ahmadpour, S. (2026). Effects of gluten-free diet on metabolic syndrome components and anthropometric indices in patients with nonalcoholic fatty liver disease: An open-label randomized clinical trial. Arab Journal of Gastroenterology, 27(1), 84-90. https://doi.org/10.1016/j.ajg.2025.10.001

Kabbani, T. A., Goldberg, A., Kelly, C. P., Pallav, K., Tariq, S., Peer, A., Hansen, J., Dennis, M., & Leffler, D. A. (2012). Body mass index and the risk of obesity in coeliac disease treated with the gluten-free diet. Alimentary Pharmacology & Therapeutics, 35(6), 723-729. https://doi.org/10.1111/j.1365-2036.2012.05001.x

Kelly, D. L., Lee, C. M., Vyas, G., Buchanan, R. W., Roche, D. J., Harrington, V., Rowland, L. M., Wijtenburg, S. A., Adhikari, B. M., Sathyasaikumar, K. V., Kochunov, P., Talor, M. V., Waltz, J., Liu, F., Hare, S., Adams, H. A., Schwarcz, R., Salem, D., Gold, J. M., . . . Eaton, W. W. (2026). Randomized double-blind inpatient study of a gluten-free diet for negative symptoms in people with schizophrenia. Schizophrenia Research, 289, 33-45. https://doi.org/10.1016/j.schres.2026.01.006

Kurppa, K., Paavola, A., Collin, P., Sievänen, H., Laurila, K., Huhtala, H., Saavalainen, P., Mäki, M., & Kaukinen, K. (2014). Benefits of a gluten-free diet for asymptomatic patients with serologic markers of celiac disease. Gastroenterology, 147(3), 610-617.e1. https://doi.org/10.1053/j.gastro.2014.05.003

Lange, S., Tsohataridis, S., Boland, N., Ngo, L., Hahad, O., Münzel, T., Wild, P., Daiber, A., Schuppan, D., Lurz, P., Keppeler, K., & Steven, S. (2024). Effects of short-term gluten-free diet on cardiovascular biomarkers and quality of life in healthy individuals: A prospective interventional study. Nutrients, 16(14), Article 2265. https://doi.org/10.3390/nu16142265

Lebwohl, B., Cao, Y., Zong, G., Hu, F. B., Green, P. H. R., Neugut, A. I., Rimm, E. B., Sampson, L., Dougherty, L. W., Giovannucci, E., Willett, W. C., Sun, Q., & Chan, A. T. (2017). Long term gluten consumption in adults without celiac disease and risk of coronary heart disease: Prospective cohort study. BMJ, 357, Article j1892. https://doi.org/10.1136/bmj.j1892

Lee, A. R., Wolf, R. L., Lebwohl, B., Ciaccio, E. J., & Green, P. H. R. (2019). Persistent economic burden of the gluten free diet. Nutrients, 11(2), Article 399. https://doi.org/10.3390/nu11020399

Lis, D., Stellingwerff, T., Kitic, C. M., Ahuja, K. D. K., & Fell, J. (2015). No effects of a short-term gluten-free diet on performance in nonceliac athletes. Medicine and Science in Sports and Exercise, 47(12), 2563-2570. https://doi.org/10.1249/MSS.0000000000000699

Lis, D. M., Stellingwerff, T., Shing, C. M., Ahuja, K. D. K., & Fell, J. W. (2015). Exploring the popularity, experiences, and beliefs surrounding gluten-free diets in nonceliac athletes. International Journal of Sport Nutrition and Exercise Metabolism, 25(1), 37-45. https://doi.org/10.1123/ijsnem.2013-0247

Mårild, K., Størdal, K., Bulik, C. M., Rewers, M., Ekbom, A., Liu, E., & Ludvigsson, J. F. (2017). Celiac disease and anorexia nervosa: A nationwide study. Pediatrics, 139(5), Article e20164367. https://doi.org/10.1542/peds.2016-4367

Maziarz, M., Koskenniemi, J. J., Martinez, M. M., Spiliopoulos, L., Salami, F., Toppari, J., Kero, J., Veijola, R., Tossavainen, P., Palmu, S., Aronsson, C. A., Lundgren, M., Borg, H., Katsarou, A., Elding Larsson, H., Knip, M., Lou, O., Dunne, J. L., Törn, C., & Lernmark, Å. (2025). β-Cell function and glucose tolerance in persons with multiple islet autoantibodies randomized to a gluten-free diet. Journal of the Endocrine Society, 9(8), Article bvaf073. https://doi.org/10.1210/jendso/bvaf073

Molina-Infante, J., & Carroccio, A. (2017). Suspected nonceliac gluten sensitivity confirmed in few patients after gluten challenge in double-blind, placebo-controlled trials. Clinical Gastroenterology and Hepatology, 15(3), 339-348. https://doi.org/10.1016/j.cgh.2016.08.007

Naeem, F., Noreen, S., Sultan, H., & Aja, P. M. (2026). Modulation of metabolic parameters among obese diabetic individuals through a gluten-free diet. BMC Nutrition, 12, Article 102. https://doi.org/10.1186/s40795-026-01316-1

National Institute of Diabetes and Digestive and Kidney Diseases. (n.d.-a). Definition & facts for celiac disease. National Institutes of Health. niddk.nih.gov/health-information/digestive-diseases/celiac-disease/definition-facts

National Institute of Diabetes and Digestive and Kidney Diseases. (n.d.-b). Diagnosis of celiac disease. National Institutes of Health. niddk.nih.gov/health-information/digestive-diseases/celiac-disease/diagnosis

National Institute of Diabetes and Digestive and Kidney Diseases. (n.d.-c). Eating, diet, & nutrition for celiac disease. National Institutes of Health. niddk.nih.gov/health-information/digestive-diseases/celiac-disease/eating-diet-nutrition

National Institute of Diabetes and Digestive and Kidney Diseases. (n.d.-d). Symptoms & causes of celiac disease. National Institutes of Health. niddk.nih.gov/health-information/digestive-diseases/celiac-disease/symptoms-causes

National Institute of Diabetes and Digestive and Kidney Diseases. (n.d.-e). Treatment for celiac disease. National Institutes of Health. niddk.nih.gov/health-information/digestive-diseases/celiac-disease/treatment

Niland, B., & Cash, B. D. (2018). Health benefits and adverse effects of a gluten-free diet in non-celiac disease patients. Gastroenterology & Hepatology, 14(2), 82-91. pmc.ncbi.nlm.nih.gov/articles/PMC5866307

O'Brien, R., Cosier, D., & Lambert, K. (2025). The climate footprint of diabetic and gluten-free diets in Australia. Dietetics, 4(2), Article 12. https://doi.org/10.3390/dietetics4020012

Penny, H. A., Shiha, M. G., Raju, S. A., Nandi, N., Trott, N., Costas-Batlle, C., Whelan, J., Schiepatti, A., Andrews, C., Preest, G., Currie, C., Mountford, N., Urwin, H., Cross, S. S., Bateman, A. C., Robins, G., Byrnes, V., Gillett, H., Gillett, P., . . . Sanders, D. S. (2026). The 2026 British Society of Gastroenterology guidelines on the diagnosis and management of adult coeliac disease. Gut. Advance online publication. https://doi.org/10.1136/gutjnl-2025-337747

Quan, L., Xu, X., Cui, Y., Han, H., Hendren, R. L., Zhao, L., & You, X. (2022). A systematic review and meta-analysis of the benefits of a gluten-free diet and/or casein-free diet for children with autism spectrum disorder. Nutrition Reviews, 80(5), 1237-1246. https://doi.org/10.1093/nutrit/nuab073

Raehsler, S. L., Choung, R. S., Marietta, E. V., & Murray, J. A. (2018). Accumulation of heavy metals in people on a gluten-free diet. Clinical Gastroenterology and Hepatology, 16(2), 244-251. https://doi.org/10.1016/j.cgh.2017.01.034

Reijonen, J. K., Tihtonen, K. M. H., Luukkaala, T. H., & Uotila, J. T. (2025). Gluten-free diet during pregnancy and pregnancy outcome: A retrospective cohort study. International Journal of Gynecology & Obstetrics, 169(3), 1184-1190. https://doi.org/10.1002/ijgo.16130

Rej, A., Sanders, D. S., Shaw, C. C., Buckle, R., Trott, N., Agrawal, A., & Aziz, I. (2022). Efficacy and acceptability of dietary therapies in non-constipated irritable bowel syndrome: A randomized trial of traditional dietary advice, the low FODMAP diet, and the gluten-free diet. Clinical Gastroenterology and Hepatology, 20(12), 2876-2887.e15. https://doi.org/10.1016/j.cgh.2022.02.045

Rubio-Tapia, A., Hill, I. D., Semrad, C., Kelly, C. P., Greer, K. B., Limketkai, B. N., & Lebwohl, B. (2023). American College of Gastroenterology guidelines update: Diagnosis and management of celiac disease. The American Journal of Gastroenterology, 118(1), 59-76. https://doi.org/10.14309/ajg.0000000000002075

Russell, L. A., Alliston, P., Armstrong, D., Verdu, E. F., Moayyedi, P., & Pinto-Sanchez, M. I. (2025). Micronutrient deficiencies associated with a gluten-free diet in patients with celiac disease and non-celiac gluten or wheat sensitivity: A systematic review and meta-analysis. Journal of Clinical Medicine, 14(14), Article 4848. https://doi.org/10.3390/jcm14144848

Saad, M., & Maisch, N. M. (2026). Levothyroxine absorption in patients with celiac disease: Challenges and therapeutic strategies. The Annals of Pharmacotherapy, 60(3), 303-305. https://doi.org/10.1177/10600280251361914

Salmela, E., Kurppa, K., Lindfors, K., Saavalainen, P., Huhtala, H., Kaukinen, K., & Taavela, J. (2025). Long-term health outcomes of people without celiac disease avoiding gluten consumption: A 25-year prospective cohort study. European Journal of Clinical Nutrition, 79(10), 1053-1059. https://doi.org/10.1038/s41430-025-01641-x

Schmucker, C., Eisele-Metzger, A., Meerpohl, J. J., Lehane, C., Kuellenberg de Gaudry, D., Lohner, S., & Schwingshackl, L. (2022). Effects of a gluten-reduced or gluten-free diet for the primary prevention of cardiovascular disease. Cochrane Database of Systematic Reviews, 2022(2), Article CD013556. https://doi.org/10.1002/14651858.CD013556.pub2

Shiha, M. G., Manza, F., Figueroa-Salcido, O. G., Ontiveros, N., Caio, G., Jansson-Knodell, C. L., Rubio-Tapia, A., Aziz, I., & Sanders, D. S. (2026). Global prevalence of self-reported non-coeliac gluten and wheat sensitivity: A systematic review and meta-analysis. Gut, 75(3), 502-510. https://doi.org/10.1136/gutjnl-2025-336304

Skodje, G. I., Sarna, V. K., Minelle, I. H., Rolfsen, K. L., Muir, J. G., Gibson, P. R., Veierød, M. B., Henriksen, C., & Lundin, K. E. A. (2018). Fructan, rather than gluten, induces symptoms in patients with self-reported non-celiac gluten sensitivity. Gastroenterology, 154(3), 529-539.e2. https://doi.org/10.1053/j.gastro.2017.10.040

Smeets, S. M., Kiefte-de Jong, J. C., & van der Velde, L. A. (2024). Food insecurity and other barriers to adherence to a gluten-free diet in individuals with coeliac disease and non-coeliac gluten sensitivity in the Netherlands: A mixed-methods study. BMJ Open, 14(10), Article e088069. https://doi.org/10.1136/bmjopen-2024-088069

Srajer, A., Harris, M., Santos, J., Dalere, J., King, J. A., Skiffington, J. C., Slater, D., Leijser, L., & Metcalfe, A. (2026). Obstetrical outcomes in pregnant patients following a gluten-free diet: A prospective cohort study. International Journal of Gynecology & Obstetrics, 174(1), 340-346. https://doi.org/10.1002/ijgo.70843

U.S. Food and Drug Administration. (n.d.-a). Gluten-free labeling of foods. fda.gov/food/nutrition-food-labeling-and-critical-foods/gluten-free-labeling-foods

U.S. Food and Drug Administration. (n.d.-b). Medications and gluten. fda.gov/drugs/buying-using-medicine-safely/medications-and-gluten

van Berge-Henegouwen, G. P., & Mulder, C. J. (1993). Pioneer in the gluten free diet: Willem-Karel Dicke 1905-1962, over 50 years of gluten free diet. Gut, 34(11), 1473-1475. https://doi.org/10.1136/gut.34.11.1473

Vasant, D. H., Paine, P. A., Black, C. J., Houghton, L. A., Everitt, H. A., Corsetti, M., Agrawal, A., Aziz, I., Farmer, A. D., Eugenicos, M. P., Moss-Morris, R., Yiannakou, Y., & Ford, A. C. (2021). British Society of Gastroenterology guidelines on the management of irritable bowel syndrome. Gut, 70(7), 1214-1240. https://doi.org/10.1136/gutjnl-2021-324598

Vici, G., Belli, L., Biondi, M., & Polzonetti, V. (2016). Gluten free diet and nutrient deficiencies: A review. Clinical Nutrition, 35(6), 1236-1241. https://doi.org/10.1016/j.clnu.2016.05.002

Zdzieblik, D., Zierke, A., Waldvogel, T., Gollhofer, A., & König, D. (2025). A gluten-free diet does not alter performance outcomes in nonceliac athletes undergoing sprint interval training: A pilot trial. Frontiers in Sports and Active Living, 7, Article 1646563. https://doi.org/10.3389/fspor.2025.1646563

Zong, G., Lebwohl, B., Hu, F. B., Sampson, L., Dougherty, L. W., Willett, W. C., Chan, A. T., & Sun, Q. (2018). Gluten intake and risk of type 2 diabetes in three large prospective cohort studies of US men and women. Diabetologia, 61(10), 2164-2173. https://doi.org/10.1007/s00125-018-4697-9

Evidence rating and review record

Each row identifies the method, source, outcome or recommendation, and comparison being rated. Ratings from different tools are not interchangeable. The source is either a named published review or guideline or this page's own judgment. The rating in the banner is for this diet's main recommendation, the treatment of diagnosed celiac disease, and the banner names the tool that produced it. A high rating means the research is solid; it does not mean the diet is right for everyone, and it does not mean a coach should recommend it.

Every row on this page uses GRADE, the same 4-level scale as the clinical guidelines you will meet in practice. High means further research is very unlikely to change the conclusion. Moderate means further research may change it. Low means further research is likely to change it. Very low means the conclusion is uncertain. GRADE rates each outcome separately, so a diet can be high for one outcome and very low for another.

Four things about this table matter before reading it. The first 3 rows and the irritable bowel syndrome row are ratings that guideline panels attached to recommendations, not ratings of a single pooled result; the European and British panels give the same treatment recommendation different labels, and both are shown. The rows on celiac disease, on gluten sensitivity, and on everyone else answer different questions and are not interchangeable. The heart disease and death rows rate associations with how much gluten people ate, not the effect of choosing a gluten-free diet. And 1 row records an outcome this page did not rate, with the reason.

Ratings marked as this page's own judgment were assigned from the evidence given in each row's reasons; ratings from a published guideline or review are reported exactly as published.

On a narrow screen, scroll the table sideways to see every column.

Evidence ratings, their sources and reasons

Outcome	Certainty	Who rated it, and for which comparison	Why it is rated this way
Treatment of diagnosed celiac disease: the European panel's recommendation of a strict, lifelong gluten-free diet	Moderate	GRADE. Al-Toma et al., 2026, the European Society for the Study of Coeliac Disease guideline panel. The certainty of evidence the panel attached to its recommendation that adults with celiac disease follow a strict, lifelong gluten-free diet, compared with a gluten-containing diet	As published, with a strong recommendation and 100 percent panel agreement. The supporting tables give different ratings for individual outcomes; this is the recommendation's rating. Symptom resolution is rated Moderate, from observational studies with serious risk of bias, and healing, long-term complications and quality of life are each labeled Low. For each of those 3 outcomes the circle symbols printed beside the label disagree with it, 2 showing the symbol for Moderate and 1 the symbol for Very low; this page does not choose between label and symbol, and carries only the recommendation's own rating.
Treatment of diagnosed celiac disease: the British panel's recommendation of a lifelong gluten-free diet	Low	GRADE. Penny et al., 2026, the British Society of Gastroenterology guideline panel. The certainty of evidence attached to its recommendation that adults diagnosed with celiac disease maintain a lifelong gluten-free diet, compared with a gluten-containing diet	As published, with a strong recommendation and 94 percent agreement. The supporting table rates non-randomized studies of each outcome: quality of life, fractures, metabolic and liver outcomes, mucosal healing and nutrition each Low, lowered for serious risk of bias and raised for a strong association, and cancers of the lymphatic system Very low, lowered for serious risk of bias and serious imprecision. The guideline notes that no randomized placebo-controlled trial has evaluated the diet's long-term health impact. The European and British labels for the same treatment differ and are not averaged.
Reliability of celiac testing after gluten has been reduced or removed: the European panel's gluten challenge recommendation	Low	GRADE. Al-Toma et al., 2025, the same European panel. The certainty of evidence attached to its recommendation that a gluten challenge is required when a person suspected of celiac disease has reduced or removed gluten before testing	As published, with a strong recommendation and 100 percent agreement. The supporting table rates the outcome it used, damage to the intestinal lining after a gluten challenge in adults already avoiding gluten, Low, lowered for serious inconsistency, because doses ran from milligrams to grams and durations from 3 days to 3 months, and for serious imprecision. This rates a recommendation about the challenge; it is not an accuracy estimate for every celiac test. The British guideline's separate recommendation on the challenge dose, 3 to 6 grams of gluten a day for at least 6 weeks, is rated Moderate and answers a different question (Penny et al., 2026).
Healing of the small intestine, measured by the ratio of villous height to crypt depth	Moderate	GRADE. This page's own judgment. Gluten-free diet against continued gluten for 1 year, in 40 adults with celiac antibodies found by screening relatives (Kurppa et al., 2014)	Randomized trials start high. Not lowered for risk of bias: allocation used random number tables with permuted blocks, the nurse was masked to patient characteristics, analysis was by intention to treat, and the biopsies were coded and not read until the trial ended; one participant in the gluten group started the diet after randomization. Lowered 1 level for imprecision: a single trial of 40 people, although the difference was statistically clear. Inconsistency and publication bias cannot be judged from 1 trial. Not lowered for indirectness, because the row is defined for screen-detected adults; the authors caution against extending it to people with other antibody patterns.
Digestive symptoms, same trial and comparison	Low	GRADE. This page's own judgment. Same trial (Kurppa et al., 2014)	Randomized trials start high. Lowered 1 level for risk of bias, because participants knew their diet and symptoms were self-reported, and the authors note a placebo effect may have contributed. Lowered 1 level for imprecision: a single trial of 40 people; the difference in the total symptom score was 0.4 points on a 7-point scale, from 0.1 to 0.7. Inconsistency and publication bias cannot be judged from 1 trial, and indirectness is not lowered, as in the row above. Two levels down from high gives low.
Global symptoms of irritable bowel syndrome, gluten-free diet against a control diet	Very low	GRADE. Vasant et al., 2021, the British Society of Gastroenterology guideline. The quality of evidence attached to its weak recommendation against a gluten-free diet in irritable bowel syndrome, based on a meta-analysis of 2 randomized trials with 111 participants (Dionne et al., 2018)	As published. The guideline reports that the 2 trials showed a reduction in global symptoms that was not statistically significant, a relative risk of symptoms remaining of 0.42, from 0.11 to 1.55. The meta-analysis's own GRADE table rates the same comparison very low: serious risk of bias, serious inconsistency, no serious indirectness, very serious imprecision, and publication bias strongly suspected (Dionne et al., 2018). The guideline's recommendation and the meta-analysis's estimate are separate items that carry the same label.
Gluten-specific symptoms under blinded challenge: the share of adults with suspected gluten sensitivity whose symptoms appeared with gluten but not with placebo	Low, for the finding that fewer than half met their trial's definition of gluten-specific symptoms	GRADE. This page's own judgment. The 4 blinded, placebo-controlled challenge trials that reported individual responses, 231 adults, as tabulated in a review (Molina-Infante & Carroccio, 2017); challenges of about 1 to 2 weeks	Randomized blinded trials start high. The threshold is whether fewer than half of participants react to gluten itself. The review classified 38 of 231 as gluten-specific; the 4 trials ranged from 8 to 34 percent, each below half, so inconsistency is not lowered. Imprecision is not lowered: a simple 95 percent interval around the combined share, about 12 to 22 percent, stays below half. That interval treats the 4 different trials as one and is not a pooled estimate, and one trial's own interval reaches about half, so the finding rests on the trials together, not on each trial alone. Whether the trials' differing challenge methods reflect people outside them is judged under indirectness. Lowered 1 level for risk of bias: only 4 of the review's 10 trials reported individual responses, one used a placebo containing fermentable carbohydrate, and washouts as short as 1 week risk carry-over in crossover designs. Lowered 1 level for indirectness: the trials defined a response differently and used gluten doses and durations, some below usual intake, that may miss some true responders, as the review notes. Publication bias could not be assessed. This row does not rate the share who reacted to placebo, and placebo symptoms do not show that expectation caused any one person's symptoms.
Coronary heart disease, highest against lowest fifth of estimated gluten intake, adults without celiac disease	Low, for no clear association	GRADE. This page's own judgment. 2 prospective cohorts analyzed together, 110,017 adults, 26 years (Lebwohl et al., 2017)	Cohort studies start low. The hazard ratio, 0.95 from 0.88 to 1.02, is precise, with 6,529 cases, and was checked for consistency between the 2 cohorts. Not lowered further for risk of bias; residual confounding is reflected in the low starting point. Not raised: no large effect or dose-response pattern. This rates an association with gluten intake in people not avoiding it, not the effect of choosing a gluten-free diet. The Cochrane ratings in the next 2 rows answer different questions and are not used to support this one.
Death from cardiovascular disease, lower against higher gluten intake, adults without known celiac disease	Low	GRADE. Schmucker et al., 2022, a Cochrane review. Lower against higher estimated gluten intake in 2 cohort studies covering 3 cohorts from the United States and United Kingdom, followed 11 to 26 years	As published: hazard ratio 1.00, from 0.95 to 1.06. Lowered 1 level for risk of bias and 1 for indirectness, because the participants were health professionals in the United States or volunteers in the United Kingdom. This rates gluten intake, not adopting a gluten-free diet. The review's abstract and its detailed table describe these cohorts differently; this page follows the table.
Non-fatal heart attack, lower against higher gluten intake, adults without known celiac disease	Low	GRADE. Schmucker et al., 2022. Lower against higher estimated gluten intake in 2 United States cohorts of health professionals, 26 years	As published: hazard ratio 0.99, from 0.89 to 1.10. Lowered 1 level for risk of bias and 1 for indirectness, for the same reasons as the row above. An intake association, not the effect of a gluten-free diet.
Systolic blood pressure, healthy adults, gluten-free against usual diet, 6 months	Very low	GRADE. Schmucker et al., 2022. 1 randomized trial, 60 healthy adults in Italy	As published: mean difference 6.9 mmHg lower, from 17.1 lower to 3.3 higher. Lowered 1 level for risk of bias and 2 for imprecision: a very small sample and an interval that crosses no effect and includes a difference of 2 mmHg, which the review treats as minimally important. A different comparison from the 2 cohort rows above.
Motivation and pleasure component of negative symptoms (CAINS-MAP), 5 weeks, inpatients with schizophrenia or a related disorder and raised antigliadin antibodies, gluten-free against gluten-containing diet	Low	GRADE. This page's own judgment. 1 inpatient trial, 43 randomized and 39 analyzed (Kelly et al., 2026)	Randomized trials start high. Lowered 1 level for risk of bias: 4 randomized people were excluded under early-treatment criteria, later withdrawals differed between arms, 4 of the 5 in the gluten group, and 2 people in the gluten-free group contributed 4-week rather than 5-week data; blinding with identical flour does not remove these concerns. Lowered 1 level for imprecision: a single small, short trial with a modest effect, an effect size of 0.39, and the share improving by 20 percent did not differ clearly, 10 of 21 against 5 of 18. Inconsistency and publication bias cannot be assessed from 1 trial. Not lowered for indirectness, because the row is defined for this narrow inpatient, antibody-positive population. Broader negative-symptom and other symptom scales did not show clear differences, and this row does not rate them.
Stereotyped behavior and cognition, children with autism, a mixed set of gluten-free, casein-free and gluten-free-only diets against usual diets	Very low	GRADE. This page's own judgment. A meta-analysis of 8 small randomized trials, 297 children, which did not rate certainty: 5 trials removed gluten and casein and 3 removed gluten alone; the stereotyped behavior estimate came from 5 trials (3 and 2 of each kind) and the cognition estimate from 3 (1 and 2) (Quan et al., 2022)	Randomized trials start high. Lowered 2 levels for risk of bias: only 1 of 8 trials stated that the children were blinded, and the authors name performance bias as the main source of bias. Lowered at least 1 level for imprecision: small trials, and the cognition interval reaches nearly no effect; the cognition result also lost statistical significance when the one trial that removed both proteins was taken out. Very low is the floor. The rating applies to this mixed set of diets and is not a rating for a gluten-free diet alone or for a gluten-free, casein-free diet alone.
Thyroid-stimulating hormone, adult women with Hashimoto's thyroiditis without celiac disease, gluten-free against control diets, 1 to 3 months	Very low	GRADE. Araújo et al., 2025. 3 randomized trials, 110 participants, all women	As published: mean difference 0.63 uIU/mL lower, from 1.63 lower to 0.36 higher. Lowered for serious risk of bias, 2 of the 3 trials at high overall risk, and for extremely serious imprecision. The review rated every outcome very low, including the 2 thyroid antibodies, which moved in opposite directions.
Time-trial performance, competitive cyclists without celiac disease, gluten-free diet against the same diet with added gluten, 7 days	Low	GRADE. This page's own judgment. 1 randomized, double-blind crossover trial, 13 cyclists (Lis, Stellingwerff, Kitic, et al., 2015)	Randomized trials start high. Not lowered for risk of bias: an independent observer randomized the order from a computer-generated list, cyclists and researchers were blinded, and all 13 completed. Lowered 2 levels for imprecision: a single trial of 13 people, sized to detect a 10 percent difference, which is larger than differences that can decide a competition, so smaller effects cannot be ruled out. Inconsistency and publication bias cannot be assessed from 1 trial. Not lowered for indirectness, because the row is limited to this population and time frame. For athletic performance more broadly, this page's judgment is very low: 1 short cycling trial and 1 nonrandomized 6-week pilot study in 15 men, which does not raise certainty (Zdzieblik et al., 2025).
Body weight and body mass index, adults without celiac disease	Not rated	This page's own judgment	The trials identified differ in population (healthy adults; adults with metabolic syndrome, fatty liver disease, or obesity and type 2 diabetes), in diet (gluten-free, gluten-reduced or low-gluten) and in comparison, and one compared each group only with its own starting values. They are reported in section 4 as measurements, and this page makes no claim that the diet changes body weight, so no single rating would describe them (Ehteshami et al., 2018; Hansen et al., 2018; Jazayeri et al., 2026; Naeem et al., 2026; Schmucker et al., 2022).

How this archive talks about weight. This page reports what studies measured, including body weight, and does not treat weight change as a goal or as a measure of a diet's success. Weight a person loses without meaning to can be a sign of a health problem, and where it is, the page names it as a reason to see a clinician. People are described the way the study described them.

Status: reviewed and approved by James Peth, MS, MPH, PhD, on October 4, 2026, after a combined final review of 5 archive pages, a focused recheck and a final confirmation. Written October 3, 2026, revised the same day after the combined final review, and repaired after a focused recheck of those repairs. Ratings marked as this page's own were assigned from the evidence; published ratings are reported as published. What changed: After the focused recheck, the fatty liver trial is described from its full report, with its self-reported adherence and inconsistent tables noted; the metabolic syndrome trial's waist measurements are reported as conflicting; the celiac center weight study gives its exact categories, counts and limits; the gluten sensitivity row explains its precision judgment; and the British guideline's outcome ratings were checked against its supporting table.

Next scheduled review: August 2027. Pages built on annual guidance are rechecked each year. The celiac disease guidelines cited on this page, from the European Society for the Study of Coeliac Disease, the British Society of Gastroenterology and the American College of Gastroenterology, should be checked first for updates.

---

Diet Reference Archive | One James

Prepared by James Peth, MS, MPH, PhD

© 2026–2027 One James. All rights reserved.

This page is for education and coaching practice. It is not medical or dietary advice.