Although 20-30% of people consider themselves to be “allergic” to food, the prevalence of true food allergy is 6-7% in children and only 1-2% in adults. Allergen skin testing and serum specific IgE (sIgE) testing are commonly used to screen for food allergies, but the food provocation test is the gold standard for diagnosing food allergies. It includes open, single-blind and double-blind tests. 1. Open test The physician gives the patient the normal form of the provoked food, and both parties are informed. When the type of suspected food is widespread, the patient can use this method at home to screen for foods with low suspicion and mild expected symptoms, and then verify the positive food using a blind method. 2. Single-blind test The physician fully disguises the stimulating food in terms of color, texture, odor and taste, and sets up a placebo. The test is completed before the patient is unblinded. For patients with suspected psychogenic food allergy, placebo can be given first, and if the result is positive, at least two tests are required. 3. Double-blind test Mainly used for scientific research. Before the test, the patient should strictly avoid the intake of the suspected food, and if asymptomatic and without symptomatic drugs, the dose should be gradually increased from a safe dose on an empty stomach at 20-30 minute intervals until the total cumulative dose is greater than or equal to the daily intake. Test only one food at a time, in a medical facility by an experienced physician, and if a food has caused a lethal reaction, try to avoid it or perform the test in a hospital or even under intensive care unit supervision. All signs, symptoms, and relevant tests should be accurately recorded before each dose is administered, and the patient should continue to be observed for 1 to 2 hours after the final dose is given. The excitation test should simulate natural exposure as much as possible, while paying attention to the presence of concomitant factors, such as exercise, menstruation, or other co-morbidities.