A roadmap from test to table for celiac disease, which most people who have it don’t know

Key takeaways

• A 2018 global review found celiac antibodies in 1.4% of people tested, and its authors, citing earlier studies, wrote that 83% to 95% of cases in developed countries go undiagnosed².
• The disease can show up as an itchy, blistering rash, iron deficiency, thinning bones, abnormal liver tests or slow growth in children, sometimes with no stomach trouble at all⁴⁵.
• If you’ve already cut gluten, celiac groups advise eating 3 to 10 grams of it a day for at least six weeks before testing, though one group notes there’s limited agreement on the dose⁸⁹.
• An FDA “gluten-free” label means less than 20 parts per million of gluten, and in January 2026 the agency began gathering evidence on barley and rye that labels don’t name, such as in malt¹¹⁰.
• In kitchen tests, a shared toaster kept gluten-free bread under 20 ppm, while gluten-free pasta boiled in wheat-pasta water reached up to about 115 ppm until it was rinsed¹³¹⁵.

In January 2026, the Food and Drug Administration asked the public for help with a gap in food labels¹.

Wheat is a major food allergen, so a package must name it. Barley and rye aren’t on that list¹. An ingredient called malt can come from barley without the label saying so. Some spices, flavorings and colors don’t have to name their sources either¹.

For most shoppers, that’s trivia. For people with celiac disease, it’s a daily hazard. Their immune system attacks the small intestine when they eat gluten, a protein in wheat, rye, barley and their crossbreeds⁴¹⁰.

About 1% of people worldwide have it, and the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) estimates about 2 million Americans do³. Many of them don’t know it³. For them, a strict gluten-free diet is the only treatment there is⁵.

Celiac disease is common, and most cases go unrecognized

In 2018, researchers pooled 96 studies from around the world. Among 275,818 people screened with blood tests, 1.4% had celiac antibodies. In studies that confirmed cases with a tissue sample, the rate was 0.7%².

Women and girls had higher confirmed rates than men and boys, and children had higher rates than adults². Citing earlier studies, the authors wrote that 83% to 95% of people with celiac disease in developed countries remain undiagnosed².

U.S. counts vary. An FDA labeling page says up to 3 million Americans, above NIDDK’s 2 million³¹¹.

It runs in families. A 2015 pooled analysis of 10,252 parents, siblings and children of people with celiac disease found that 7.5% had it, a 2023 report in the Medical Journal of Australia noted¹². In that report’s own study, Brisbane researchers screened 202 such relatives and found nine new cases. Seven were children¹².

Genes set the stage, but they don’t decide. Celiac disease almost always occurs in people with gene variants called DQ2 or DQ8. About 30% of people carry one, yet only about 3% of carriers develop the disease, NIDDK says⁴. Celiac Canada puts that figure at about 4%⁸.

It’s more common with Down, Turner or Williams syndrome, and it can occur alongside type 1 diabetes or thyroid disease³.

The symptoms often show up far from the gut

Children are more likely than adults to have digestive symptoms, NIDDK says⁴. These include bloating, gas, chronic diarrhea or constipation, greasy, foul-smelling stools, vomiting and belly pain⁴.

A child’s growth can tell the story, too. Watch for slowed growth, short height, delayed puberty, weight loss, irritability and damaged enamel on permanent teeth⁴.

Adults may have no gut symptoms at all. NIDDK lists fatigue, joint or bone pain, depression or anxiety, headaches, numbness from nerve damage, canker sores and fertility problems in both women and men⁴.

The skin can be the only clue. Dermatitis herpetiformis is an itchy, blistering rash that tends to appear on the elbows, knees, buttocks, back or scalp⁴. It affects about 10% to 20% of adults and 2% to 3% of children with untreated celiac disease. Some people have the rash and nothing else⁴.

The American College of Gastroenterology (ACG) updated its celiac guideline in 2023. According to a summary in American Family Physician, it lists clues that make testing reasonable⁵:

• iron deficiency or weight loss with no clear cause;
• osteopenia, meaning thinning bones;
• raised liver enzymes;
• nerve damage in the hands or feet;
• mouth ulcers;
• poor growth in children;
• thyroid disease or repeated bouts of pancreatitis.

Left untreated, celiac disease can lead to osteoporosis, anemia, malnutrition and, rarely, a small-intestine cancer or lymphoma³. That’s true even for people without symptoms, NIDDK notes⁴.

Diagnosis follows a set order, and gluten has to stay on the menu

The first test is a blood test for an antibody called tissue transglutaminase IgA, or tTG-IgA. In people who are eating gluten, it catches 63% to 93% of cases. It rarely flags people who don’t have the disease: its specificity is 96% to 100%⁵.

Your clinician should also check total IgA. People who make too little of it can get a falsely normal result, so doctors switch to IgG-based tests⁵.

For most adults, a positive result leads to an upper endoscopy, a thin scope passed through the mouth into the small intestine. The ACG advises taking at least four biopsy samples, because the damage can be patchy⁵. When suspicion is high, the scope may be worth it even if blood work is negative⁵.

Some children can skip the scope. Under 2020 European pediatric guidelines, doctors can diagnose a child without a biopsy when three things line up⁷:

• tTG-IgA is at least 10 times the upper limit of normal;
• a second blood sample is positive for endomysial antibodies (EMA);
• the family agrees.

Neither symptoms nor gene tests are required⁷. The ACG also suggests a no-biopsy route for selected children⁶. For adults who can’t or won’t have an endoscopy, it calls the same approach reasonable, but treats the result as a possible diagnosis rather than a confirmed one⁶.

If you’ve already stopped eating gluten

Blood tests and biopsies lose accuracy on a gluten-free diet. A gene test can rule celiac disease out, as our article on medically appropriate elimination diets explains. If you carry DQ2 or DQ8, your clinician may suggest a gluten challenge: eating gluten again, on purpose, before testing⁸.

Advice varies by group:

Group	Daily gluten	Rough bread equivalent	How long
Celiac Canada, standard challenge (2026)	6 to 10 grams	2 to 3 slices of wheat bread	6 to 8 weeks
Celiac Canada, modified challenge	3 grams	About 1.5 slices	At least 6 weeks
Coeliac Australia (2021)	10 grams for adults, about 5 for children	About 4 slices	At least 6 weeks

Coeliac Australia notes that a slice of bread holds roughly 2 to 4 grams of gluten, and that experts haven’t fully agreed on the dose or length⁹. Both groups advise against a challenge in children under 5 or during the growth spurt of puberty⁸⁹. Celiac Canada suggests building up over the first week and, if you have symptoms, retesting about four weeks in⁸.

Do this with a clinician, because the tests need careful timing.

Your celiac roadmap, from first hunch to follow-up

Use our summary of the guidance to plan talks with your clinician.

Stage	What happens	What to ask
1. Suspect	You notice gut symptoms, a rash, low iron, thin bones or slow growth, or a relative has celiac disease45	“Could this be celiac disease? Should I be tested?”
2. Test while eating gluten	tTG-IgA plus total IgA5	“Am I eating enough gluten for an accurate result?”
3. Confirm	Endoscopy with at least four biopsies for most adults; a no-biopsy route for some children57	“Do I meet the no-biopsy criteria, or do I need a scope?”
4. Start the diet	Strict, lifelong gluten-free eating, with a dietitian’s help5	“Can you refer me to a dietitian who knows celiac disease?”
5. Tell your family	Testing can be considered for parents, siblings and children5	“Which of my relatives should be tested?”
6. Follow up	Antibody checks, bone and nutrient checks, a pneumococcal vaccine and possibly a repeat biopsy56	“When should we recheck my blood work and my bones?”

What a gluten-free label promises, and the gap the FDA is studying

Since the FDA’s 2013 rule, a food labeled gluten-free must contain less than 20 parts per million (ppm) of gluten¹⁰¹¹. The agency picked that line because it was the lowest level tests could reliably measure across many kinds of food¹⁰. Some research suggests most people with celiac disease can tolerate trace amounts¹¹.

A 2024 review notes that about 10 milligrams of gluten a day may harm some people with celiac disease. At 20 ppm, that takes about 500 grams of food, a little over a pound¹⁵.

What the label covers. The claim is voluntary and applies to FDA-regulated packaged foods, including supplements¹¹. It doesn’t govern restaurant menus, though the FDA urges restaurants to follow it. Most meat and poultry labels fall under the USDA, and most alcoholic drinks under the Alcohol and Tobacco Tax and Trade Bureau¹¹.

Fermented foods got their own rule in 2020. Fermenting or breaking down proteins chops gluten into fragments that standard lab tests can’t count¹⁰. That affects yogurt, cheese, sauerkraut, pickles, green olives, vinegar, beer and the hydrolyzed plant proteins in soups and seasonings¹⁰.

So since August 13, 2021, makers of these foods who claim gluten-free must keep records showing their ingredients met the limit before fermentation, plus records on how they prevent cross-contact¹⁰. Cross-contact means gluten getting into a food by accident from shared equipment or surfaces. For distilled foods, the FDA checks for the absence of protein, because gluten doesn’t carry over when liquid is distilled¹⁰.

Oats are a special case. They aren’t a gluten-containing grain under the rule, but the FDA calls cross-contact with wheat, barley or rye a major concern¹. Oats may carry the gluten-free label if they test under 20 ppm¹¹. The ACG suggests choosing gluten-free oats, and notes that some people with celiac disease react to oat protein itself⁶.

The 2026 question. The FDA’s request for information asked where and how often rye and barley go unnamed, how much gluten oats pick up and how they’re labeled, and whether shoppers skip flavors or spices that might hide gluten¹. The comment period, extended once, ran into spring 2026¹. As of the notices we reviewed, the agency hadn’t proposed a new rule. Until it does, read “malt” as a possible barley warning¹.

Cross-contact: what kitchen experiments found

Celiac families are often coached to be extra cautious with shared toasters, knives and pots¹³. A team at Children’s National Hospital measured how much gluten those habits move and published the results in the journal Gastroenterology in 2019¹³:

• Toaster: gluten-free bread toasted in a shared toaster stayed under 20 ppm in repeated tests¹³.
• Knife: gluten-free cupcakes cut with a knife used on regular cupcakes stayed under 20 ppm in most cases; 2 of 28 went over¹³¹⁵.
• Pasta water: gluten-free pasta boiled in water used for wheat pasta reached about 34 to 116 ppm. Rinsing it under running tap water brought it under 20 ppm, and using fresh water in a rinsed pot left gluten undetectable¹³¹⁵.

The authors called the work small and preliminary, and told families to keep following their care team’s advice¹³. Tricia Thompson of the group Gluten Free Watchdog argued that testing one small portion per sample could miss crumbs. She still advises cleaning shared toasters or using toaster bags¹⁴.

A 2024 review in the journal Nutrients gathered several such studies¹⁵. When one knife spread condiments on both regular and gluten-free bread, 18% of mayonnaise samples and 10% of peanut butter samples topped 20 ppm¹⁵. Fryers gave mixed results: 5 of 20 restaurant fry orders from shared fryers topped 20 ppm in one study, versus 1 of 74 fast-food samples in another¹⁵. The reviewers concluded that shared equipment adds little risk when it’s cleaned well, while shared cooking water is a clear risk¹⁵.

Follow-up care: the part people tend to skip

Going gluten-free starts treatment; it doesn’t end care. The ACG points newly diagnosed patients to a dietitian⁵.

Antibody levels often return to normal within 6 to 12 months on the diet⁶. The gut lining is slower, taking a median of three years to heal, so the ACG suggests considering a repeat biopsy around two years in⁵. Routine repeat biopsies aren’t recommended for children without symptoms⁶.

Ask your clinician about three more checks:

• Bones. The ACG suggests considering a DXA scan, an X-ray test of bone density⁶.
• Nutrients. Gut damage can reduce how well you absorb iron, folate, vitamin B12 and the fat-soluble vitamins A, D, E and K⁶.
• Vaccines. The guideline recommends a pneumococcal vaccine after diagnosis⁶.

Limited evidence suggests probiotics don’t improve symptoms⁵. If symptoms linger, ask your care team to look for hidden gluten. Rarely, in refractory celiac disease, the gut doesn’t heal even on a strict diet³.

No medicine replaces the diet yet. The ACG guideline says there’s no effective drug treatment⁵. Our article on digestive enzymes covers gluten-digesting pills and drugs still in development.

Gluten-free is medicine here, but for others, it’s a different question

Natural-health readers often say food is medicine. Celiac disease is the clearest proof: no pill or supplement does what a strict gluten-free diet does for it⁵.

The diet also rewards habits many readers already value, like cooking from scratch, reading labels and knowing where food comes from. Because gluten comes from wheat, rye, barley and their crossbreeds, plain fruits, vegetables, beans, eggs, fish, meat, rice and corn start out free of it¹⁰.

But gluten-free eating is also a trend, and the trend can work against a diagnosis. Drop gluten before testing, and the tests lose the signal they need⁵.

If gluten seems to bother you but celiac tests are negative, something else may be going on. Our article on food intolerances covers non-celiac gluten sensitivity.

How strong is the evidence?

Approach	Rating	Why
Strict, lifelong gluten-free diet after diagnosis	Strong evidence	The only effective treatment in the 2023 ACG guideline5
tTG-IgA blood test while eating gluten	Strong evidence	Specificity of 96% to 100% in people eating gluten5
No-biopsy diagnosis in children who meet strict criteria	Strong evidence	Backed by the 2020 European pediatric guideline, built on a formal evidence review, and by the ACG67
No-biopsy diagnosis in adults	Promising but early	ACG calls it reasonable but treats it as a possible diagnosis6
Shared toasters carry little risk	Promising but early	One small study, with sampling questioned1314
Shared pasta water spreads gluten	Promising but early	Seen in more than one small study; rinsing helped15
Handheld gluten sensors	Not enough evidence	No proof they improve outcomes; false readings56
Probiotics for celiac symptoms	Not enough evidence	Limited evidence, no clear benefit5

Questions readers ask

The bottom line

Celiac disease is common, often silent and easy to miss when it shows up as low iron, weak bones or a rash. The path to an answer runs in order: test while you still eat gluten, confirm with a biopsy or a no-biopsy route that fits, then start the diet with expert help.

Living well with it means reading labels closely, guarding against the few kitchen habits that spread gluten and keeping up with follow-up care. For this disease, gluten-free eating isn’t a trend. It’s the treatment.

Related reading

• Elimination diets are real medicine for some conditions and a risk for others
• Your body makes digestive enzymes to spare, and pills help when it can’t
• Leaky gut can be measured, but most “heal your gut” protocols haven’t been tested
• Half of adults overrated a 12-grain bread, so learn to spot real whole grains
• Most food intolerances come down to dose, while gluten sensitivity remains a puzzle

Sources

1. U.S. Food and Drug Administration. “Labeling and Preventing Cross-Contact of Gluten for Packaged Foods; Request for Information.” Federal Register 91 FR 2781, January 22, 2026. federalregister.gov
2. Singh P, Arora A, Strand TA, et al. “Global Prevalence of Celiac Disease: Systematic Review and Meta-analysis.” Clinical Gastroenterology and Hepatology, 2018;16:823–836. celiac.org
3. National Institute of Diabetes and Digestive and Kidney Diseases. “Definition & Facts for Celiac Disease.” NIH, last reviewed October 2020. niddk.nih.gov
4. National Institute of Diabetes and Digestive and Kidney Diseases. “Symptoms & Causes of Celiac Disease.” NIH, last reviewed October 2020. niddk.nih.gov
5. American Family Physician. “Diagnosis and Management of Celiac Disease: Guidelines From the American College of Gastroenterology” (summary of Rubio-Tapia A, et al., American Journal of Gastroenterology, 2023). January 2024. aafp.org
6. Mejía Pérez LK. “Celiac Disease ACG Guidelines Update” (review of the 2023 ACG guideline). HCPLive, December 6, 2023. hcplive.com
7. Husby S, Koletzko S, Korponay-Szabó I, et al. “European Society Paediatric Gastroenterology, Hepatology and Nutrition Guidelines for Diagnosing Coeliac Disease 2020.” Journal of Pediatric Gastroenterology and Nutrition, 2020;70(1):141–157 (Leiden University repository). scholarlypublications.universiteitleiden.nl
8. Celiac Canada (Canadian Celiac Association). “Gluten Challenge” (for health care professionals). Last updated June 16, 2026. celiac.ca
9. Coeliac Australia. “Gluten Challenge Fact Sheet.” June 2021. coeliac.org.au
10. U.S. Food and Drug Administration. “Food Labeling; Gluten-Free Labeling of Fermented or Hydrolyzed Foods” (final rule). Federal Register, August 13, 2020. govinfo.gov
11. U.S. Food and Drug Administration. “Questions and Answers on the Gluten-Free Food Labeling Final Rule.” Content current as of June 30, 2022. fda.gov
12. Muir R, Sehgal A, Tye-Din JA, Daveson AJM. “Undiagnosed Coeliac Disease Identified With Active Case Finding in First-Degree Relatives.” Medical Journal of Australia, 2023;219(8). mja.com.au
13. Children’s National Hospital. “Do Celiac Families Really Need 2 Toasters?” (news release on Weisbrod VM, Silvester JA, et al., Gastroenterology, 2019). ScienceDaily, September 30, 2019. sciencedaily.com
14. Thompson T. “Gluten Free Watchdog Comment on the Recent Gluten Cross Contact Study.” Gluten Free Watchdog, October 8, 2019. glutenfreewatchdog.org
15. Damasceno RPB, Zandonadi RP, Mendes M, et al. “Risk of Gluten Cross-Contamination Due to Food Handling Practices: A Mini-Review.” Nutrients, 2024;16(8):1198. pubmed.ncbi.nlm.nih.gov