Food allergies

The difference between a food allergy and an intolerance, what anaphylaxis looks like and what to do about it, why coeliac disease is neither, and why a gluten-free diet is not a general health measure.

Allergy, intolerance, and why the difference matters

These two get used interchangeably and they are not the same thing.

A food allergy is an immune system reaction. The body treats a harmless protein as a threat and mounts a response, which can be rapid and can be severe. Reactions typically begin within minutes of eating, and in a small number of cases they are life-threatening.

A food intolerance does not involve the immune system. It is usually a digestive problem — the body cannot break something down properly, as in lactose intolerance — and it produces bloating, wind, stomach pain or diarrhoea, often some hours after eating. It tends to be dose-related, so a small amount may cause nothing while a large amount causes trouble.

Intolerance can make life genuinely miserable. It is not dangerous. Allergy can be.

That is the whole reason to be careful with the words. Someone who describes a true allergy as an intolerance may not be taken seriously enough by a kitchen. Someone who describes an intolerance as an allergy contributes to the opposite problem, which is restaurant staff treating all such requests as preference rather than medicine.

What a reaction looks like

Allergic reactions to food commonly produce:

  • itchy skin, or a raised rash — hives
  • swelling of the lips, face and around the eyes
  • sneezing, or an itchy, runny or blocked nose
  • nausea, vomiting, stomach pain or diarrhoea
  • coughing, wheezing or breathlessness

Most reactions start soon after eating, though some appear later.

Several of those overlap with intolerance symptoms, which is part of why self-diagnosis is unreliable and why the next sections matter more than the list above.

Anaphylaxis

This is the part that must not be left out of anything written about food allergy, and it is what turns an inconvenience into an emergency.

Call 999 if any of the following happen:

  • the lips, mouth, throat or tongue suddenly swell
  • breathing becomes very fast or difficult
  • the throat feels tight, or swallowing becomes difficult
  • the skin, tongue or lips turn blue, grey or pale
  • someone becomes suddenly confused, drowsy or dizzy, or faints and cannot be woken
  • a child goes limp, floppy or stops responding normally

If an adrenaline auto-injector is available, use it immediately, following the instructions printed on the device itself. Then call 999 — and call even if the person appears to recover, because reactions can return after an apparent improvement.

Adrenaline is the treatment. Antihistamines are not a substitute for it in a severe reaction, and waiting to see whether an antihistamine helps is the delay that causes harm.

Anyone who has been prescribed an auto-injector should carry it, and the people around them should know where it is and how it works. A device left at home is of no use to anybody.

What triggers it, and what the label has to say

The foods most often responsible are cows' milk, eggs, peanuts, tree nuts, fish, shellfish, wheat and soya. Less commonly: celery, mustard, sesame and lupin.

UK law requires fourteen allergens to be declared in food: celery, cereals containing gluten, crustaceans, eggs, fish, lupin, milk, molluscs, mustard, peanuts, sesame, soybeans, sulphur dioxide and sulphites, and tree nuts.

Since October 2021, food that is packed on the premises where it is sold — the sandwiches and salads in a café, deli or bakery chiller — must carry a full ingredients list with the allergens emphasised within it. The requirement is generally known as Natasha's Law, after Natasha Ednan-Laperouse, who died in 2016 after eating a baguette whose sesame was not declared on the label.

That change matters because it closed the gap people were most likely to fall into: food that looks packaged and therefore looks labelled, sold by a business too small to have been covered by the previous rules.

Getting a proper answer

Suspected food allergy goes through a GP, who may refer on to an allergy specialist. The established tests are skin-prick testing and blood tests, alongside a food and symptom diary and, where appropriate, a supervised exclusion diet.

The important instruction is the one people skip: do not cut foods out of the diet without speaking to a GP first. There are two reasons, and both are practical rather than bureaucratic.

The first is nutritional. Removing a major food group, particularly from a child's diet, has consequences that need managing.

The second is that it interferes with getting a diagnosis. Testing works best while the suspected food is still being eaten, and someone who has avoided something for months may have made their own answer impossible to confirm.

Exclusion diets

An exclusion diet is a diagnostic tool, not a treatment. The suspected food is removed for a defined period, symptoms are observed, and the food is then reintroduced to see whether they come back. The reintroduction is the half that gets skipped, and without it the exercise proves very little.

It works best supervised, ideally with a dietitian where a significant food group is involved, and it is a test with a beginning and an end rather than a way of life.

Older advice tends to present these diets as straightforwardly effective. They are useful when done properly and misleading when not, and the difference lies almost entirely in whether the reintroduction happens and whether anybody is keeping track.

The practical side of running one — planning around work, family meals and shopping — is covered in more detail at World Food and Wine.

Coeliac disease is not an allergy

It is frequently filed alongside food allergies because the response to it is a strict diet, and it belongs in a different category altogether.

Coeliac disease is an autoimmune condition. Gluten triggers the immune system to attack the lining of the small intestine, damaging the villi that absorb nutrients. It is not an allergy and not an intolerance, and it does not cause the kind of rapid reaction described above.

It runs in families. Where a parent, sibling or child has been diagnosed, the chance of having it rises to roughly one in ten, which is why testing the immediate family after a diagnosis is recommended rather than optional.

Untreated, it does lasting damage, including to nutrient absorption and bone density, which is why it matters that it gets properly diagnosed rather than self-managed.

One critical practical point: anyone being tested for coeliac disease needs to still be eating gluten. Going gluten-free first can produce a negative result in someone who has the condition. Diet changes belong after the test, not before it, and a GP should be the one advising on the sequence.

The only treatment is a strict gluten-free diet for life. The practical business of cooking and baking without gluten is covered at World Food and Wine.

Gluten-free is not a general health measure

Two claims circulate about gluten-free eating that are not supported, and both deserve correcting plainly.

Gluten-free diets are not an established treatment for autism. The idea comes from a theory that incompletely digested gluten and casein form peptides with opioid-like effects. Systematic reviews have found insufficient good-quality evidence that these diets improve the core features of autism. Some later analyses report small effects on particular measures, but with small numbers of participants, short study periods and a real risk of bias. It is not a recommended treatment.

That matters more than an ordinary inaccuracy, because the suggestion leads to restrictive diets being imposed on children, with the nutritional consequences that carries and no demonstrated benefit to set against them. Any dietary change for an autistic child belongs with a dietitian and a paediatrician, not with an article.

Gluten-free diets do not treat “candida”. Candida infections are real — oral and vaginal thrush among them — and they are treated with antifungal medicines. The broader notion of a systemic candida overgrowth causing diffuse symptoms in otherwise healthy people is poorly supported, and there are no clinical trials showing that an anti-candida diet treats any recognised condition. Where people do feel better on one, the likeliest explanation is the processed food and sugar that went out alongside the gluten.

For anyone without coeliac disease, a wheat allergy or diagnosed non-coeliac gluten sensitivity, going gluten-free has no established health benefit. It is not a neutral choice either: gluten-free substitute products are often lower in fibre and higher in fat and sugar than what they replace.

Babies, and the advice that reversed

This is where guidance has changed most completely, and anyone working from what they were told fifteen years ago has it backwards.

Parents used to be advised to delay or avoid giving peanut and egg to babies, particularly those thought to be at risk. That advice has been reversed. Current UK guidance is that excluding egg and peanut from an infant's diet could increase their risk of food allergy.

In outline: exclusive breastfeeding for around the first six months; other foods introduced from about six months and not before four; and if egg and peanut are part of the family's diet, introduced before the age of one.

Infants at higher risk — those with eczema, particularly moderate to severe eczema or eczema starting in the first three months, or an existing food allergy — may benefit from earlier introduction, with age-appropriate egg and then peanut from around four months once they are already taking puréed fruit and vegetables. That group should speak to a health professional before starting, because the timing is worth getting right.

The evidence comes from trials that introduced allergenic foods early in both higher-risk and general-population infants, and found early introduction protective rather than harmful.

The risk of a severe reaction during introduction is low — in the region of one to two in a thousand, even among infants who already have a food allergy or severe eczema. If a reaction is suspected, the food is stopped and advice sought.

Having a food-allergic family member does not by itself place a baby in the higher-risk group, though it is a sensible reason to plan how introduction is done.

Changed