Should I try a gluten-free diet to see if it helps?

Not before being tested for coeliac disease. This is the single most important practical point on the topic, because both the blood test and the biopsy for coeliac disease become unreliable once gluten has been removed from the diet — antibody levels fall and the gut lining recovers. The extremely common pattern of feeling unwell, going gluten-free, feeling somewhat better and then asking for a test later forecloses the one diagnosis that genuinely matters, and reversing it requires eating gluten again for at least six weeks, which many people understandably refuse. Coeliac disease affects around one per cent of people and is mostly undiagnosed, so the test is worth doing. Ask for it first, then experiment.

Is non-coeliac gluten sensitivity real?

The symptoms are real. The attribution to gluten specifically usually is not. When people who identify as gluten sensitive are given gluten or an identical-looking placebo without knowing which, only a minority — typically reported between about 16 and 30 per cent — react consistently to gluten itself. Many report worsening on placebo. What does reliably reproduce symptoms in blinded challenge is fructans, a fermentable carbohydrate abundant in wheat, which a gluten-free diet also removes without anyone noticing. So the more accurate description for most people is a reaction to something in wheat, most likely fructans, with a substantial contribution from expectancy. That is a meaningfully different conclusion from "gluten sensitivity is not real", and it points towards a different and more effective treatment.

Why would fructans cause the same symptoms?

Because they behave like other FODMAPs. Fructans are short chains of fructose that humans cannot digest, so they pass into the colon where they draw water in osmotically and are rapidly fermented by gut bacteria, producing gas and distension. In people with visceral hypersensitivity — a heightened perception of normal gut sensations, which is the central mechanism in irritable bowel syndrome — that distension causes disproportionate pain and bloating. Wheat is one of the largest sources of fructans in the Western diet, so cutting out bread, pasta and cereals removes a large fructan load at the same time as removing gluten. The two exposures are impossible to separate without a properly controlled challenge, which is exactly why so many people conclude that gluten is the culprit.

Is a gluten-free diet harmful if I do not need it?

It is not dangerous, but it is not free either. Commercial gluten-free breads, pastas and cereals are generally lower in fibre, iron, folate and B vitamins and often higher in fat, sugar and salt than the products they replace, because removing gluten damages the structure and manufacturers compensate with refined starches. Whole grain intake tends to fall, which has cardiovascular implications of its own. The products typically cost between one and a half and five times as much, which is a genuine burden for people with coeliac disease who have no alternative and an unnecessary one for those who do. There is also a social cost, and in a minority the restriction becomes a route into disordered eating, particularly when exclusions gradually expand.

What should I do if I feel better without gluten but tested negative for coeliac disease?

The reasonable next step is a properly structured low FODMAP approach with a dietitian rather than indefinite gluten avoidance, because the evidence suggests fructans are the more likely culprit and the low FODMAP process identifies your individual triggers rather than excluding a whole food group permanently. That process has three stages: a short restriction phase of four to six weeks, systematic reintroduction of each FODMAP group to find which ones actually cause symptoms and at what quantity, and then a personalised long-term diet that is as unrestricted as possible. Many people who thought they were gluten sensitive discover they tolerate sourdough bread, small quantities of wheat, or wheat at some times and not others — which is far more workable than lifelong exclusion.

How does GMJ News cover gluten and wheat sensitivity?

GMJ News covers the imperative of testing before gluten withdrawal, coeliac disease diagnosis including the IgA deficiency pitfall and biopsy-free paediatric criteria, wheat allergy as a distinct entity, blinded rechallenge methodology and its consistent findings, fructans and the FODMAP explanation, amylase-trypsin inhibitors and the case for the term non-coeliac wheat sensitivity, nocebo effects in unblinded exclusion diets, the nutritional and financial costs of gluten-free products, the three-phase low FODMAP protocol as the more appropriate intervention, and the risk of restrictive eating patterns. Part of PHIG's contested science and gastroenterology programmes.

Home › Topics › Gluten and Non-Coeliac Sensitivity

Gluten and Non-Coeliac Sensitivity

GMJ News knowledge hub · last reviewed September 2026 · Georgian Medical Journal

Summary Statistics Glossary GMJ news FAQ Documents Organizations Research

Coeliac disease is a well-defined autoimmune enteropathy affecting around 1% of the population, and wheat allergy is a distinct IgE-mediated condition — both are diagnosable and both require gluten exclusion. Non-coeliac gluten sensitivity is more complicated, and the honest position is that it exists as a symptom experience while its attribution to gluten specifically is not supported in most people who report it: double-blind placebo-controlled rechallenge studies consistently find that only a minority react to gluten itself, while fructans and other FODMAPs in wheat, together with nocebo expectancy, explain a substantial proportion of symptoms (WHO). The clinically important consequence is procedural rather than philosophical: coeliac serology becomes unreliable once gluten has been withdrawn, so the very common pattern of self-diagnosis followed by dietary exclusion forecloses the one diagnosis that genuinely matters, and anyone considering a gluten-free diet should be tested before starting rather than after.

Key messages

THE PROCEDURAL POINT THAT MATTERS MOST: test before excluding gluten, never after

Coeliac serology and duodenal biopsy both become unreliable once gluten has been withdrawn, because the antibodies fall and the villous architecture recovers. The extremely common sequence — symptoms, self-diagnosis, gluten-free diet, later request for testing — forecloses the one diagnosis that genuinely changes management and requires a formal gluten challenge of several weeks to reverse, which many people are unwilling to undertake. Anyone considering a gluten-free diet should be tested first. This single piece of advice is more clinically valuable than any argument about whether non-coeliac gluten sensitivity exists.

SETTLED: coeliac disease and wheat allergy are real, distinct and diagnosable

Coeliac disease is an autoimmune enteropathy affecting around 1% of the population, driven by gluten in genetically susceptible individuals carrying HLA-DQ2 or DQ8, diagnosed by tissue transglutaminase IgA with total IgA, confirmed by duodenal biopsy in most adults, and requiring strict lifelong gluten exclusion to prevent malabsorption, osteoporosis, infertility and a modest increase in small bowel lymphoma risk. Wheat allergy is a separate IgE-mediated condition. Both are objectively testable, and both are under-diagnosed — a substantial majority of coeliac disease remains undetected in most countries.

CONTESTED ATTRIBUTION: symptoms are real; gluten as the cause usually is not

Non-coeliac gluten sensitivity describes people with genuine symptoms — bloating, abdominal pain, altered bowel habit, fatigue, headache, brain fog — that improve on a gluten-free diet in the absence of coeliac disease or wheat allergy. The symptoms are not in doubt. What double-blind placebo-controlled rechallenge studies consistently show is that only a minority reliably react to gluten specifically when they do not know whether they are receiving it, with reported rates typically in the range of 16-30% of those who identify as gluten sensitive. This is a statement about the causal agent, not about the reality of the illness.

THE BETTER EXPLANATION: fructans and other FODMAPs in wheat

Wheat contains fructans, a fermentable oligosaccharide and a major dietary FODMAP source, and a gluten-free diet incidentally removes a large proportion of dietary fructans. Blinded crossover studies, most notably work from Oslo and Monash, have shown that fructan challenge reproduces symptoms in self-reported gluten-sensitive individuals while gluten challenge often does not. This offers a coherent mechanism grounded in osmotic load, colonic fermentation and visceral hypersensitivity, and it explains why a low FODMAP approach frequently succeeds where gluten exclusion has partially helped — and why many such patients are better characterised as having irritable bowel syndrome.

ALSO REAL: nocebo and expectancy effects in an unblinded intervention

Gluten-free eating is an intensely visible, effortful and identity-linked intervention, which maximises expectancy effects in both directions. Rechallenge studies routinely find that a substantial proportion of participants report symptom worsening on placebo, and that symptom scores rise when people believe they have consumed gluten regardless of what they actually consumed. Acknowledging this is not an accusation of imagination — nocebo effects produce measurable physiological changes — but it does mean that unblinded personal experience cannot establish causation, which is precisely why formal rechallenge protocols exist.

THE HARMS OF UNNECESSARY EXCLUSION ARE UNDERSTATED

A gluten-free diet is not nutritionally neutral. Commercial gluten-free replacement products are frequently lower in fibre, iron, folate and B vitamins and higher in fat, sugar and salt than their conventional equivalents, and are substantially more expensive — a genuine equity issue. Reduced whole grain intake has its own cardiovascular implications. There is also the social and psychological cost of restriction, and in a minority the diet becomes a vector for disordered eating. None of this argues against exclusion where it is indicated; it argues for establishing whether it is indicated.

Key statistics

~1%

population prevalence of coeliac disease, with the majority undiagnosed in most countries

BSG/ESsCD

Test first

coeliac serology and biopsy become unreliable once gluten is withdrawn — the key practical rule

BSG/NICE

16-30%

of self-identified gluten-sensitive people react specifically to gluten on blinded rechallenge

Gastroenterology/Gut

Fructans

reproduce symptoms in blinded challenge where gluten frequently does not

Gastroenterology 2018

HLA-DQ2/DQ8

present in almost all coeliac disease — a negative result has high negative predictive value

ESsCD

Nutritionally poorer

gluten-free replacement products typically lower in fibre and micronutrients and more costly

Nutrients/BDA

Gluten and wheat sensitivity — where the disagreement actually lies

Source: Bars show strength of supporting evidence. The symptoms are real; the attribution to gluten specifically is what the evidence does not support.

Glossary of key terms

Coeliac disease diagnosis
GastroenterologySerological testing begins with IgA tissue transglutaminase antibody together with total IgA, since selective IgA deficiency is around ten times more common in coeliac disease and produces false negatives — in which case IgG-based testing is used. Endomysial antibody is more specific and used for confirmation. Duodenal biopsy showing villous atrophy, crypt hyperplasia and intraepithelial lymphocytosis remains standard in adults, with at least four duodenal and one bulb specimen because the disease is patchy. Paediatric guidelines permit biopsy-free diagnosis when tTG exceeds ten times the upper limit of normal with positive endomysial antibody. All of this requires the patient to be consuming gluten — typically the equivalent of two slices of bread daily for at least six weeks before testing.

FODMAPs and fructans in wheat
Nutrition scienceFODMAPs are fermentable oligosaccharides, disaccharides, monosaccharides and polyols — short-chain carbohydrates that are poorly absorbed, osmotically active and rapidly fermented by colonic bacteria, producing gas, luminal distension and symptoms in people with visceral hypersensitivity. Wheat is a major source of fructans, a FODMAP oligosaccharide, in Western diets. Because a gluten-free diet removes wheat, rye and barley, it incidentally removes a large proportion of dietary fructans, which means the intervention conflates two entirely different exposures. This is the single most important insight in the field: an uncontrolled gluten-free diet cannot distinguish a gluten effect from a fructan effect, and the blinded studies that separate them generally implicate fructans.

Blinded rechallenge methodology
Clinical researchThe design that resolves attribution questions. Participants who report gluten sensitivity are placed on a background diet controlled for FODMAPs, then randomised in crossover fashion to receive gluten, placebo, or in some designs fructans, in indistinguishable form — typically muesli bars or capsules — with adequate washout between periods and blinding of both participant and assessor. Results have been consistent across several groups: symptom scores rise on placebo in a substantial proportion, only a minority respond specifically and reproducibly to gluten, and fructan challenge often outperforms gluten in reproducing symptoms. Criticisms of these studies are legitimate — small samples, short challenge duration, possible nocebo saturation from repeated challenges — but no comparably rigorous design supports the opposite conclusion.

Amylase-trypsin inhibitors and other wheat components
Emerging scienceWheat contains several bioactive components besides gluten and fructans that have been proposed as symptom drivers. Amylase-trypsin inhibitors are plant defence proteins that activate toll-like receptor 4 on intestinal myeloid cells and drive innate immune activation in animal models, and have been proposed as a mechanism for wheat-related symptoms independent of both gluten and FODMAPs. Wheat germ agglutinin and other lectins are also periodically implicated, generally on weaker evidence. This work is mechanistically interesting and preliminary in humans, and the honest position is that "non-coeliac wheat sensitivity" may be a more accurate label than "gluten sensitivity" because it makes no claim about which component is responsible.

Gluten-free products and nutritional quality
DieteticsComparative analyses of gluten-free versus conventional equivalents consistently find that the replacements are lower in fibre, protein, iron, folate, niacin and B vitamins, and frequently higher in fat, sugar and salt, because removing gluten damages dough structure and manufacturers compensate with refined starches, fats and sugars. Fortification of gluten-free flour is mandatory in fewer jurisdictions than for wheat flour. Cost premiums of 150-500% are routinely documented, which makes the diet a real financial burden for people with coeliac disease who have no choice, and an unnecessary one for those who do. People following the diet for coeliac disease should have dietetic input and periodic assessment of iron, folate, B12, vitamin D and bone density.

Gluten-free diets and disordered eating
Psychology/NutritionAny restrictive diet adopted without medical necessity carries a risk of becoming a vehicle for disordered eating, and food exclusion for perceived health reasons is a recognised presentation route for avoidant restrictive food intake disorder and for orthorexic patterns. Features that should raise concern include progressive expansion of exclusions beyond the original one, distress or anxiety when eating outside the rules, social withdrawal around food, weight loss and preoccupation disproportionate to symptom benefit. Clinicians frequently overlook this because the exclusion is framed in health terms and may be partially effective. Conversely, people with genuine coeliac disease also experience elevated rates of disordered eating, and the necessary vigilance about food can amplify it — so the concern applies in both directions and requires care rather than assumption.