Food sensitivity test: what the IgG results mean and what to test instead

A typical at-home food sensitivity kit takes a few drops of blood from a finger-prick and scores your IgG antibodies (IgG, or immunoglobulin G, is the most common antibody in your blood) against 100 to 200 foods. The scores are genuine lab measurements of exposure. The immune system makes IgG to the food proteins it meets every day, so the foods you eat most often tend to score highest. This explains why eggs and wheat flag on so many reports, and why allergy specialists do not use them.

Do food sensitivity tests work?

Food sensitivity tests work as measurements and fail as diagnoses. An IgG panel is an accurate measurement giving the wrong answer: it measures exposure, not reaction ¹. The European Academy of Allergy and Clinical Immunology (EAACI) concluded that food-specific IgG4 (the IgG subclass most panels report) reflects repeated exposure to a food and, if anything, tolerance of it ¹. The Canadian Society of Allergy and Clinical Immunology (CSACI) found no body of research supporting IgG testing for adverse food reactions and warned that acting on it can lead to unnecessary dietary restriction ². The US expert panel convened by the National Institute of Allergy and Infectious Diseases (NIAID) recommends against using food-specific IgG4 to diagnose food allergy ³.

Why your results flag the foods you eat most

People are often surprised that the report lists foods they eat daily without obvious trouble. Exposure explains it. A person who has eggs for breakfast and bread at lunch will usually show higher IgG to egg and wheat than to lamb or buckwheat, because the panel is ranking how much contact the immune system has had with each food. Change your diet for a few months and the ranking shifts with it.

There is a second problem with reading these scores. No agreed cut-off separates a normal IgG level from a meaningful one, labs use different scales, and the NIAID panel lists allergen-specific IgG4 among tests that are not standardized ³.

Already have food sensitivity test results? How to read them

If your results are already in your inbox, read the red and amber bands as a rough list of what you have been eating. Do not cut every flagged food. Two parts are worth acting on: a food you already suspected before the test, which can go into a single supervised elimination trial (below), and any warning sign listed further down, which needs a doctor rather than a diet. A standard blood test often says more. <a href="/conditions/anemia">Anemia</a>, <a href="/biomarkers/ferritin">low ferritin</a> or <a href="/learn/inflammation-markers-explained">a raised inflammation marker</a> points somewhere specific in a way a food panel cannot.

Food allergy, intolerance or sensitivity: three different things

A food allergy is an immune reaction driven by IgE, the antibody behind hay fever and anaphylaxis; it usually starts within minutes to two hours and can be dangerous. A food intolerance involves no immune reaction: lactose intolerance, for example, is a shortage of the enzyme lactase, and symptoms depend on how much you eat. Celiac disease is a third category, an autoimmune condition where gluten triggers damage to the small intestine, and it needs its own blood test. "Food sensitivity" is not a defined medical diagnosis; the term is used loosely, often in the marketing of IgG panels. In the UK and Europe the same IgG kits are often sold as a "food intolerance test".

Many people who think a food is the problem turn out not to react to it on proper testing. In a US survey of more than 40,000 adults, about 19% reported a food allergy, but only about 10.8% described symptoms convincing for a true IgE allergy ⁹. In a UK population study, about 20% believed they had a food intolerance, yet blinded food challenges confirmed it in only about 1.4% to 1.8% ¹⁰.

Food allergy is tested with food-specific IgE, not with the total IgE number; what total IgE can and cannot tell you is covered on the <a href="/biomarkers/ige-total">total IgE page</a>.

What an IgG food test measures

Making IgG to things you swallow is a normal part of the gut immune system's housekeeping. IgG4 rises with repeated exposure and is linked to tolerance: it climbs during successful allergy immunotherapy, as the body learns to stop reacting ¹. <a href="/biomarkers/igg">Total IgG</a> on a standard immunoglobulin panel is a different number altogether; it reflects your overall antibody production and says nothing about any single food.

What allergy and gut experts say

The major allergy bodies that have examined IgG food testing have reached the same position.

The fair counterweight is a 2004 randomized trial in Gut. Among 150 people with irritable bowel syndrome (IBS), symptom scores fell about 10% more on a diet that cut the foods their IgG test flagged than on a sham diet that cut other foods, and about 26% more among those who stuck to it fully ⁵. That result is real and worth knowing. It has not been confirmed by large independent trials, it does not show the flagged foods caused the symptoms (people cutting many common foods change their diet in several ways at once), and a test that helps some people with IBS choose a diet trial is still not a test that diagnoses anything.

What it costs to act on a positive result

A red list of 15 or 20 foods invites a big cut, and that cut has costs ². Dropping dairy, wheat and eggs together removes major sources of calcium, iodine, B vitamins and fiber. Children on restrictive diets can fall behind on growth, and people with a history of disordered eating can be pushed back into it. The most serious cost is a missed diagnosis. The celiac blood test only works while you are still eating gluten ⁶, so stopping gluten because a panel flagged wheat can make a later test falsely normal (our <a href="/conditions/celiac-disease">celiac disease guide</a> covers the details).

If food seems to make you feel worse, test this instead

The useful starting point is the symptom, matched to the test that can confirm or rule out a cause. These are the routes clinicians use, and each links to the page that covers the detail.

How to run an elimination diet properly

An elimination diet is the closest thing to a home test that works, because it watches what your body does rather than what your antibodies do. It works best on one suspect at a time and, ideally, with a registered dietitian. Children, pregnant people and anyone with a history of disordered eating should do this only with a doctor or dietitian.

1. 1Pick one suspectChoose one food or food group you already suspect from experience, such as dairy. A panel's red list is a poor guide to which one. If the suspect is wheat or gluten, get celiac testing while still eating it before you start.
2. 2Remove it for 2 to 4 weeksCut it fully, read labels, and keep a daily diary of symptoms, sleep and stress so you can see change against a baseline.
3. 3Bring it back on purposeReintroduce the food in a normal portion over a few days. A real trigger brings symptoms back in a pattern you can repeat; no change means the food can return to your diet. Skip this step for any food that has ever caused hives, swelling of the lips, tongue or throat, wheeze, trouble breathing or vomiting within about 2 hours: reintroducing those foods is done only as an oral food challenge with an allergist.
4. 4Use the low-FODMAP plan for IBSFor IBS, a limited low-FODMAP trial, ideally guided by a dietitian, is the evidence-backed route ¹²: restriction for a few weeks, structured reintroduction, then a personal long-term diet that keeps as many foods as possible.

For what a standard panel can and cannot pick up, see <a href="/learn/what-can-a-blood-test-detect">what a blood test can detect</a>; for gut testing of a different kind, see <a href="/learn/gut-microbiome-test-explained">gut microbiome tests explained</a>; and for eating patterns with better evidence behind them, see the <a href="/learn/anti-inflammatory-diet">anti-inflammatory diet</a>.

Free, no accountDrop in your lab report and see every marker explained in plain words, placed on the body, against your lab's range. It's read on your device: nothing is uploaded and nothing is kept.

Read my results Download the app