Everything You’ve Always Wanted to Know About Soy Allergy (Except How to Cure It)

Image by Polina Tankilevitch on Pexels

Soybean is one of the world’s oldest crops and has been feeding the people in East Asia for thousands of years. In the West, it’s mainly used to provide nutrients, texture and form to a large range of processed foods, as well as vegetarian and vegan options and formula for milk-allergic infants. It causes both immediate and delayed allergic reactions, but it’s the least problematic of all the major food allergens and, on the whole, relatively unlikely to provoke severe symptoms. That said, it’s the only priority allergen that has a history of causing epidemics, thanks to the allergenic proteins contained in its seed hull. The soy allergic should be aware that soy protein can be found in cosmetic products, and that soy is also used to make eco-friendly items like candles and cleaning products, as well as gluten-free playdough and soft toys for children, although the only products on record as causing problems are soy-containing pillows.

And now for the details, which include:

What is an allergy to soy?

Soy (Glycine max)—aka soybean, soya bean, soja bean—is an edible legume native to East Asia. The historical record suggests that the Chinese were the first to use soybeans as a food about 3000 years ago. Archaeological records suggest that soybean was domesticated up to 6000 years earlier, first by the Chinese and then the Japanese, before spreading to Southeast Asia.

Soy has several advantages over other crops, including lower cost of agricultural inputs and lower water consumption. This has helped to make it one of the most important legume crops for the feed and food industries, especially in Asia and the USA, although the consumption of soy in Europe has greatly increased over the past few years.

Currently, the biggest producers are Brazil, the US and Argentina who produce over 80% of the world’s supply.

Europe is the world’s third largest producer of non-GMO (non-genetically modified) soy, after China and India. Italy, France and Serbia are the biggest producers within the European Union.

Soybean has the highest protein content among the legumes, making up about 36–46% of the bean, depending on the cultivated variety. More importantly, according to the protein digestibility corrected amino acid score (PDCAAS), the scale currently used to measure protein nutritional quality, the quality of soybean protein is higher than that of other plant proteins and similar to that of animal protein.

Soybeans also have a high fat content—about 20%, the second highest content among all food legumes, after peanut—and are a good source of omega-6 and omega-3 essential fatty acids.

The seeds also contain other beneficial compounds such as B vitamins, fibre, calcium, potassium, iron and zinc, and biologically active compounds including isoflavones, which have been linked to a wide range of health benefits.

Research has found that, when eaten as a whole food, soy can lower ‘bad’ (LDL) cholesterol levels and blood pressure. It may also keep your blood vessels flexible, reduce your risk of having a heart attack, protect against strokes, protect against gestational diabetes mellitus and plain old diabetes, reduce menopausal symptoms (namely, hot flashes), keep your bones strong (if you are a postmenopausal woman), protect against breast cancer (if you start eating it during childhood), reduce the risk of prostate cancer, reduce the risk of endometrial cancer, reduce inflammation, protect against gout, put less pressure on your kidneys than animal proteins, help with weight loss, benefit gut health and cognitive function, reduce the symptoms of depression and iron out wrinkles.

Concerns about the negative effects of soy isoflavones on maturation, fertility and thyroid function, and possible male feminisation have all been refuted (although if a man insists on eating excessive amounts of soy—say, about 9 times the average amount a man might normally eat—he might suffer a slight degree of feminisation).

As such, soybeans appear in our diets in many forms, ranging from sprouts and immature beans (edamame) to foods made from the mature beans, such as soy milk, tofu, tempeh, miso and soy sauce.

Soybean also serves as a primary alternative protein source in vegetarian diets, forming the basis of hundreds of dairy and meat alternatives. It’s also an important source of nutrition for infants with milk allergy. Soy-based formulas were introduced over than 100 years ago and are currently used for the treatment of cow’s milk allergy, lactose and galactose intolerance, and galactosemia (a rare genetic metabolic disorder which makes babies unable to convert milk sugars into glucose).

However, most of the soy eaten in the Western world comes in the form of processed foods. Soy products—soy protein isolate (SPI), soy protein concentrate (SPC) and soy flour, which are made up of ≥90%, 65–90% and 50–65% protein, respectively—are widely used in the food industry as water- and fat-binders, texturisers, gelatinisers, emulsifiers, foaming and whipping material, and protein fillers in foods like pastries and baked goods, processed meats, infant foods, chocolates, energy bars and breakfast cereals.

Soy is also an ingredient in hundreds of household goods and chemical products. But, perhaps surprisingly, it’s biggest use is in the production of animal feed; over three-quarters (77%) of the soy grown globally is fed to livestock and much of the rest is used to make biofuels and industry or vegetable oils. Only 7% is actually eaten by people as a whole food product.

Unfortunately, soy can cause allergic reactions in a small percentage of those people. This happens because their body’s immune system mistakes one or more harmless soy proteins for toxic invaders and creates IgE antibodies against them. The next time they eat soy, the antibodies recognise certain soy proteins and prompt a response from their immune system cells. These, in turn, release a variety of chemicals into the bloodstream, including histamine, the chemical that is primarily responsible for the symptoms of allergy.

The first case of soybean allergy was reported in 1934 by American doctor W. W. Duke who described the cases of 5 people who worked in a soybean mill and developed asthma as a result of breathing in soy allergens. It is also the first (and, possibly, only) reported case of successful immunotherapy, as one patient whom he treated with ‘ascending small doses of soy bean extract ‘ was cured of his symptoms and able to resume work at the mill as long as he wore a mask.

Soy causes immediate, immunoglobin E (IgE) allergic reactions and delayed, non-IgE-mediated or mixed reactions including food protein-induced allergic proctocolitis (FPIAP), food protein-induced enteropathy (FPE), food protein-induced enterocolitis syndrome (FPIES) and eosinophilic oesophagitis (EoE).

Atopic dermatitis (AD), aka allergic eczema, which I shall now just call ‘eczema’ (although, strictly-speaking, AD is the most common subtype of eczema), can also be exacerbated by eating soy. Eczema can either be an IgE-mediated form of allergy and produce symptoms soon after eating the offending food, or it can be a non-IgE-mediated form and show up several hours or days later.

However, although soy is counted as a major allergen in many Western countries, requiring mandatory labelling, and is one of the most common triggers of delayed allergies in infants, it’s the least problematic food in the list of major allergens. In Japan, where it’s commonly consumed, it’s not even on their list of major food allergens and, in 2021, a panel of expert from the United Nations Food and Agriculture Organisation (FAO) and the World Health Organisation (WHO) recommended that soy not be listed as a global priority allergen.

That said, countries with a higher consumption of soy tend to have a higher reported prevalence of soybean allergy and there is concern that, because of the rise in popularity of vegan diets in Western countries, the incidence of soybean allergy may rise due to the increasing availability of soy-containing food products.

Identified allergens

The proteins (and occasionally carbohydrates) in a food that are capable of provoking allergic reactions are called allergens. Allergens are named using the first three letters of the genus—Glycine—the first letter of the species—max—and a number reflecting the order in which they were identified.

As of March 2026, 8 soybean allergens have been added to the WHO/IUIS allergen database (the official, peer-reviewed database of allergens maintained by the World Health Organisation and International Union of Immunological Societies):

Image by DS stories on Pexels

How common is soy allergy?

A definitive diagnosis of allergy can currently only be established with a food challenge, but challenges are costly in terms of both time and resources, as well as potentially risky. So many prevalence studies measure sensitisation—using skin or blood test data—but sensitisation is not allergy. Being sensitised to something simply means that your immune system recognises it, but you may not actually react to it; in fact, many people don’t react to whatever it is they are sensitised to. Studies that use sensitisation data therefore tend to produce allergy prevalence numbers that are larger than they should be.

In a similar vein, studies that estimate allergy prevalence using questionnaires usually produce somewhat inflated numbers as people can self-report allergies that they do not actually have, although robust studies will use certain criteria to evaluate respondents’ answers and determine whether their symptoms suggest an allergy or not.

IgE-mediated allergy

A review of 40 papers on soy allergy in children published between 1909 and 2013 reported a worldwide prevalence of soy allergy of 0.27% in the general population although, because it was surprisingly close to the level of self-reported allergy to soy of 0.2% (a number which is often much higher than proven allergy), the researchers suspected that the actual prevalence might be somewhat lower.

In 2023, a team of researchers investigating the prevalence of the 8 biggest food allergens in Europe reviewed 36 soy-related studies carried out in the whole region and calculated that around 0.33% of Europeans had been diagnosed with an allergy to soy during their lifetimes.

A 2019 study–part of the EuroPrevall project following groups of children in different age groups over time—that looked at adult food allergy in 6 European cities found a prevalence ranging from 0% in Reykjavik (Iceland), Lodz (Poland), (Madrid) Spain and (Athens) Greece, to 0.03% in Utrecht (the Netherlands) and 0.08% in Zurich (Switzerland).

A 2016 EuroPrevall study of British children living in Hampshire determined that the prevalence of soy allergy among 2-year-old children was 0.4%. Among infants, the prevalence of soy allergy was around 0.1% with soy being the 3^rd most bothersome food allergy among infants with a non-IgE-mediated food allergy (with a prevalence of 0.2%).

A 2011 study carried out in young Danish adults found soy to be the third most common food allergen, affecting 0.1% of the 22-year-olds.

In the US, a 2010 review used study data to estimate a prevalence of soy allergy of 0.4% among infants and children and 0.3% among adults. Self-reported soy allergy around that time was put at 0.25% in children and 0.35% in adults, surprisingly close to the proven number. A later survey carried out between October 2015 and September 2016 reported an increased prevalence of 0.5% among children, double the previous number.

A survey carried out between October 2015 and September 2016 that focused on adults found that soy allergy reportedly affected around 1.5 million (0.6%—also double the previous number) adults in the US, that almost half (45.4%) of adult with soy allergy had had a severe reaction, that almost half (48.3%) had had to go to the hospital after having an allergic reaction and that over a third (37.3%) had a prescription for an adrenaline auto-injector.

In Canada, a survey of 3,613 households carried out between May 2008 and March 2009 put the prevalence of self-reported soy allergy at 0.32% in children and 0.16% in adults.

Soy allergy is considered relatively uncommon in Australia and there are no prevalence studies on record that use any kind of testing, but a 2025 study that looked at parent-reported allergies in children aged between 6 to 8 years old residing in a remote region of Tasmania revealed that, according to the 1052 admissible questionnaires returned, 89 children were reported to have experienced reactions to food and 4 to soy specifically (giving a prevalence of 0.03% in the general population and 4.5% among the food-allergic), none of whom were reported to have suffered a severe reaction.

In Japan, a 2020 study following 651 children from the age of 5 determined that, as they got older, many of the children developed a sensitisation to birch pollen (from 2.2% at age 5 to 13.9% at age 9) and that about a third of them (around 4.7%) of them also developed a sensitisation to soy.

According to a 2012 review, soy often makes the top 6 of most common allergens in so-called developing and emerging economies, coming in at number 2 in Malaysia, number 3 in Thailand and Zimbabwe, number 4 in China, Taiwan and South Africa, number 5 in Hong Kong and number 6 in the Philippines and Brazil. Two studies put soy in either 4^th or 6^th position in Mexico, and either at number 6 or not in the top 6 in Singapore.

A 2020 study carried out among 1795 children with suspected food allergy in Mexico City reported that, whereas milk was the most important allergen in children younger than 5 years old, soy was among the top 5 allergens affecting children aged between 6 and 17 years old.

Non-IgE-mediated and mixed allergies

Specific prevalence information for non-IgE-mediated conditions is more hard to come by.

About 2.6% of the global population is estimated to be affected by eczema, which is just over 204 million people. It’s a condition that’s more likely to affect young children and females, and food is thought to be a trigger in 20% to 30% of the cases, with the most common allergens being milk, egg, soy, wheat, peanut and fish.

Although food-triggered eczemaaffects children more than adults, quite a few adults still have the condition. The prevalence of food allergy in children with eczema is estimated to be somewhere in the range of 15% to 30% and the prevalence of food allergy in adults with eczema is thought by most experts to be between 1% and 3%, with between 9% and up to 24.5% of that number estimated to be new, adult-onset cases.

In the US, a 1999 review of 17 years’ worth of oral food challenges in more than 600 children found that allergic reactions to egg, milk, wheat, and soy accounted for around three quarters of eczematous reactions.

10 years later, a study evaluating food allergy in American children and young adults with eczema concluded that the foods responsible for eczematous reactions hadn’t really changed much over the years, stating that ‘Allergies to egg, milk, wheat, soy, peanut, tree nuts, and seafood continue to account for approximately 90% of food-allergic reactions’.

Japanese research around the same time also put soy among the top 5 foods allergies experienced by people with eczema.

Not everyone with eczema and soy allergy suffers from a worsening of their skin condition after eating soy; in general, about 1 in 2 people with eczema and food allergy may experience a worsening of their skin condition after eating their food trigger, often after experiencing immediate-type symptoms. Only around 1 in 10 will experience worsening eczema as their only symptom.

A 2012 German review of 1843 oral food challenges given to children with suspected food allergies revealed that soy was the trigger most likely to cause a worsening of the skin condition, with wheat worsening the eczema of just over two thirds (69%) of the children with soy allergy.

A 2013 study from the Czech Republic reported that, although about a third of the 175 adolescents and adults with eczema were sensitised to soy, under 1 in 10 actually reacted to it, and none experienced a worsening of their skin condition—their symptoms were exclusively of the immediate kind (oral allergy syndrome and/or hives). A separate study reported that about 8 of the 228 (3.5%) people with eczema who were tested were sensitised to soy, 4 of whom reacted with immediate symptoms (oral allergy syndrome and/or hives) and 4 with a worsening of their skin condition.

The prevalence of food protein-induced allergic proctocolitis (FPIAP) is not actually known, although it is one of the most frequent causes of rectal bleeding in children. Food protein-induced allergic proctocolitis (FPIAP) is normally caused by exposure to cow’s milk (and other foods) via breast milk, although it can result from exposure to proteins in infant formula, including soy-based formula. Research from Israel and Brazil has estimated that about 3% to 6.8% of cases are triggered by soy.

Experts do not know the exact prevalence of food protein-induced enterocolitis syndrome (FPIES), but it’s estimated to occur in the general population at a prevalence ranging from 0.015% in Australia to 0.7% in Spain and reports of cases have been on the increase in recent years, either because of an increase in new cases or because of an increased awareness of the condition among doctors.

When it comes to FPIES, soy is frequently reported as a trigger for the chronic version seen in infants, especially in countries in which infant soy formula is commonly used.

Around the world, soy is a problem food for just over a third of infants (37.2%) with FPIES, especially American infants—notably those who are introduced to formula at a relatively young age—affecting between 4 in 10 (41%) and 5 in 10 (47%) when they are about 7 or 8 months old. That said, that’s not true for every state in the union; in Texas, rice is the No.1 problem-causing allergen, surpassing both milk and soy.

Soy seems to be less of a problem for infants in Europe, where it is reported to be the fourth most problematic allergen for children with FPIES in the UK, and the fifth in Italy. In Australia, around a third of infants with FPIES are allergic to soy and, in Israel, a study looking into FPIES caused by milk reported that none of the infants they were investigating were also allergic to soy.

Developing FPIES to soy in later childhood is relatively common in Japan, where tofu is often used in infant foods.

Cases of eosinophilic oesophagitis (EoE) have been reported to be on the increase since the turn of the century, probably because the condition is better recognised. EoE is now thought to affect 1 or 2 people in 2000 but, in people who have food allergies, the number is more like 1 in 20. EoE to a food often develops in someone who already has a standard, IgE-mediated allergy to that food.

Eosinophilic oesophagitis is more common in males and can occur at any age, but it becomes more common as people get older, peaking in adults aged between 30 and 50.

Soy causes about 1 in 10 cases of eosinophilic oesophagitis (EoE), although a recent American study examining the effectiveness of elimination diets for children with EoE reported that soy affected about 1 in 5.

Image by Jaqor Q.I. on Pexels

Will it go away?

IgE-mediated allergy

There are two main types of soy allergy; one is a primary allergy to soy itself that develops in infancy or early childhood and the other is a secondary allergy to soy that is caused by a primary allergy to birch pollen that cross-reacts with soy (an example of Pollen Food Syndrome) and tends to develop in school-age children and adults.

Infants and young children who become allergic to soy tend to outgrow the allergy during childhood. Early studies of children with soy allergy found that infants outgrew their allergy by the age of 3.

Later studies carried out in Germany came up with slightly different data; one study that followed 216 children from birth to the age of 6 found that soy allergy tended to peak at the age of 5 and then decrease during the sixth year of life. Another that followed 1 314 children from birth to the age of 13 found that the prevalence of soy sensitisation progressively increased from 2 % at the age of 2 to 7 % at the age of 10.

More recently, a study carried out in Israel involving 234 children with suspected food allergies reported that soy was the 4^th most common food allergy (affecting 23 children) and that it resolved in 83% (19) of them.

Around the same time, an American study that examined the data of 133 patients seen at a referral centre used reported that the average age of onset of soy allergy symptoms was 7 months and that, by the age of 4, 25% of the children had outgrown their allergy, by the age of 6, 45% had outgrown their allergy, and by the age of 10, 69% had outgrown their allergy.

Children who were also sensitised to lentils generally took longer to outgrow their allergy to soy and those who did not outgrow their allergy tended to have higher soy IgE levels measured in their blood in the first 2 years of their lives, levels which gradually increased and then peaked around the age of 8. Those who outgrew their allergy had lower soy IgE levels that peaked around the age of 3 before decreasing.

Adults who are allergic to soy are less likely to develop tolerance, possibly because they are allergic to birch pollen and this is causing cross-reactions with soy, or because they have persistent peanut allergy which is causing either cross-reactions or a co-allergy to soy.

Non-IgE-mediated and mixed allergies

Research carried out in the 1980s involving children with eczema suggest that allergy to the five major allergens (egg, milk, soy, wheat, and peanut) is less likely to be outgrown than allergies to other foods, finding that soy allergy is outgrown by half of the children diagnosed one year after diagnosis and by two thirds two years after diagnosis, while they are still of preschool age.

FPE and FPIAP rarely persist beyond the age of 2 and over half of young children with FPIES acquire tolerance by the age of 3. In general, FPIES tends to resolve in most children before the age of 5, although this varies per food type (allergy to liquid food is often outgrown before solid) and location (and eating culture).

Research suggests that FPIES to soy may take longer to outgrow than FPIES to milk; one American study, for example, followed 14 patients for around 2 years and reported that milk allergy was outgrown in 6 of 10 (60%) infants and soy allergy in 2 of 8 (25%). Another reported that milk allergy was outgrown by 55% of infants by the time they were 32 months old, and that soy allergy was outgrown by 28% of infants when they were 34.5 months old. Another found that the average age for developing tolerance to oat was 4 years, to rice, 4.7 years, to milk (in those without measurable IgE levels), 5.1 years and to soy, 6.7 years.

However, a Korean study on 23 infants with FPIES reported that tolerance rates to milk and soy were 27.3% and 75.0% at 6 months of age, 41.7% and 90.9% at 8 months and 63.6% and 91.7% at 10 months, respectively. Infants with an allergy to soy outgrew it by the time they were 14 months old.

Image by Kamran Aydinov on Freepik

Cross-reactions to soy

Technically-speaking, a person can be allergic to soyand another food (or foods, or aeroallergen(s)) either by cross-reactivity—the immune system mistakes the proteinin one allergen for aprotein with a similar structure inthe other—or by an independent sensitisation to each food and/or aeroallergen(a co-sensitisation or co-allergy), in which case the immune system has developed specific IgE antibodies against each allergen. It can be difficult to determine whether reactions are caused by cross-reactions or co-allergies,but the end result is the same; problems, problems.

The most likely suspects for cross-reactions with soy are other legumes. Over the years, researchers looking at the structure of legume proteins and investigating their effect on the antibodies in the blood of allergic people have reported the potential for cross-reactivity between soy and lupin, peanut, pea, chickpea, mung bean, alfalfa, broad bean, azuki bean, and lentil.

In an effort to determine what this actually all meant, a team of American researchers gave 69 children who had had a positive skin test to a legume (peanut, soybean, green bean, pea, or Lima bean) food challenges with all of the legumes. 41 of them had a positive food challenge, but only 2 (5%) had a positive challenge to 2 legumes.

In total, 31 of the positive reactions were to peanut, 10 were to soybean and 2 were to pea. Both of the children who had a positive challenge to 2 legumes had a history of anaphylaxis to peanut and both had a positive food challenge to soybean. Thus, 20% of the soy-allergic children reacted to peanut. Their symptoms during their soybean challenge were relatively mild and both of them outgrew their allergy within 3 years of the challenge while on a peanut- and soybean-restriction diet.

In a follow-up study, the researchers tested the blood of the same group of children with allergens from peanuts, soybean, Lima bean, pea, chickpeas and green beans and found what looked like extensive potential from cross-reactions that they knew did not occur. They concluded that the results of test tube trials did not correspond with reality and that people who were allergic to one legume should not remove all legumes from their diets.

A study carried out about 15 years later in Spain, where legumes are a popular part of the diet and legume allergy is the fifth most prevalent type of food allergy among children, reported that although the average legume-allergic child showed symptoms to around 3 legumes (often including lentil and chick pea), they generally tolerated soy.

Finally, a recent study carried out in the Netherlands found that, although sensitisation to multiple legumes was common, an allergy to more than one legume was quite rare among the soy-allergic; the highest percentage reached was 16.7% in people also allergic to lupin and 16.7% to pea. There was one leguminous exception; the peanut. Almost two-thirds (63.3%) of the soy-allergic were also allergic to peanut.

Peanut is the only legume that should potentially worry the soy-allergic; in fact, it’s thought to be the most common coexisting allergy in people who are allergic to soy. One American study of 122 children allergic to soy found that 107 (88%) were allergic to peanuts, too. Another noted that 38/41 (93%) of the children referred for an oral food test because of a sensitisation to soy were allergic to peanut. Ultimately, 18 were found to be allergic to soy, and 16 (89%) of those children were allergic to peanut.

People who are sensitised or allergic to both soy and peanut are generally allergic to the storage proteins in both—Gly m 5, 6 and 8 in soy, and their peanut homologues, Ara h 1, 3 and 2, respectively.

People who are sensitised to these storage proteins are more likely to have serious reactions than people sensitised to other types of allergens, and severe reactions to soy have been linked to severe reactions to peanut.

There seem to be 2 main patterns of sensitisation in soy allergy; there are the people who develop co-allergies to soy and peanuts because they are allergic to storage proteins, as described above, and there are those who develop allergies to soy because of their allergy to birch pollen and its proteins, notably the Bet v 1 PR-10 protein which is homologous with the Gly m 4 protein in soy.

A cross-sensitisation to birch pollen is more likely to occur in countries in which birch is endemic, such as northern Europe (Switzerland, Denmark, the Netherlands), central Europe (Germany, Austria, Czech Republic), North America (Canada, northern states of the USA), northern parts of China and Japan, where alder (also in the birch family) is widespread.

People who are sensitised or allergic to soy and birch people are less likely to be sensitised to peanut. If they are, it’s to the peanut’s PR-10 protein, Ara h 8, which is often associated with milder symptoms. They are also more likely to be sensitised to other fruit and nuts usually associated with birch-related Pollen Food Syndrome.

For example, an analysis of 1057 food-allergic patients in Vienna which was investigating co-sensitisation patterns identified a ‘soybean-nuts-fruits’ cross-reactive cluster that was linked to an ‘alder-birch’ cluster in 130 patients, mainly due to cross-reactivity between bet v 1 (birch), Gly m 4 (soybean), Ara h 8 (peanut), Cor a 1 (hazelnut), Mal d 1 (apple) and Pru p 1 (peach).

As ever, it pays to remember that you can’t fit everyone into neat little boxes, such as this patient, whose allergy problems started with reactions to only chick peas but who then, over the years, developed allergies to lentils, white beans, lupin, soybeans and peas. But not peanuts. Or birch pollen.

Soy allergy is often associated with milk allergy. Most infants who are allergic to soy are allergic to milk; 77% according to one American study and 91% according to an Israeli one.

This may be because infants who are allergic to cow’s milk can be offered soy formula as an alternative; thus, those who are offered soy formula but cannot drink it were often those who could not drink cow’s milk formula in the first place. The numbers are not so scarily high the other way around; around 2–3% of infants are thought to be allergic to milk, and only around 10–14% of them are thought to have a concomitant soy allergy.

Most infants become allergic to soy when they are less than 6 months old and infants with immediate IgE-mediated allergy to cow’s milk are just as just as likely to become allergic to soy as those with delayed, non-IgE-mediated allergies.

That said, infants with non-IgE-mediated milk-induced gastrointestinal allergies have a pretty big chance of being allergic to soy. Among infants with FPIES, for example, American studies have reported numbers like 44% and 60%.

Because of the reportedly high frequency of allergy to soy in infants with diseases like cow‘s milk-induced enteropathy (FPE) or enterocolitis (FPIES), soy protein-based formulas are generally not recommended for these infants who are advised to have milk hydrolysed protein formulas instead.

Surprisingly, several cross-reactive IgE epitopes (the part of the allergen to which the IgE antibody attaches itself) have been identified between soy allergens (Gly m 5, Gly m 6, Gly m Bd 30K) and cow’s milk allergens (Bos d 8 and Bos d 9). That said, whether infants who are allergic to soy are suffering from a separate, independent allergy or are experiencing symptoms because of a cross reaction between milk and soy is difficult to tell.

Image by Joshua Chekov on Unsplash

Symptoms of soy allergy

Soy allergy can be IgE-mediated, non-IgE-mediated or mixed; a combination of both. These variations generally present different types of symptoms.

Immediate reactions to soy

Immediate allergic reactions are caused by IgE antibodies. These antibodies bind to certain immune system cells—mast cells and basophils—and trigger the release of histamine and other inflammatory chemicals that cause the characteristic symptoms of allergy.

Immediate reactions are the most common type of allergic reaction to soy and they range from rashes to life-threatening anaphylaxis. Reactions are different for different people, and they can also be different for the same person, varying in severity from episode to episode.

There are broadly 4 main categories of IgE-mediated symptoms to soy.

The first category is the most well-known one; immediate symptoms caused by a primary allergy to soy, most often seen in infants and children, but also in adults.

The symptoms involved include:

• Skin symptoms: eczema (atopic dermatitis), hives (urticaria), facial swelling (angio-oedema) and/or tongue and/or throat and/or hands, redness (erythema), itchy skin (pruritus), oral allergy syndrome (OAS, commonly manifesting as ‘itchy’ or ‘burning’ symptoms often limited to the mouth, sometimes also swollen lips and cheeks)
• Breathing symptoms: blocked nose (nasal congestion), runny nose (allergic rhinitis), wheezing, difficulty breathing/shortness of breath (dyspnoea), persistent cough, hoarse voice
• Digestive (GI) symptoms: nausea, diarrhoea, vomiting, stomach pain
• Cardiovascular symptoms: low blood pressure (hypotension), rapid heart rate (tachycardia), loss of consciousness (syncope)
• Neurological symptoms: headaches, dizziness, blurred vision, anxiety, confusion, seizures, fatigue and malaise (aka ‘a feeling of impending doom’, which can occur during anaphylactic reactions (2))

Symptoms generally appear within 10 to 60 minutes.

The allergens that provoke these symptoms tend to be stable—that is, resistant to cooking and digestion—and eating small amounts of soy protein often causes systemic reactions (generalised reactions that spread from one organ to others and affect the whole body rather staying in one area).

Luckily, symptoms (in infants and young children) are often quite mild. An American study which looked at the data of children with eczema who had had food challenges to several major allergens including soy over a 13-year period reported 53 challenges to soy, 22% of which had resulted in ‘minimal’ reactions, 55% in ‘mild’ reactions and 23% in ‘moderate’ reactions. No severe reactions had been registered in all that time.

Another American study of 133 soy-allergic children (aged between 2 months and 17.5 years) seen at one allergic clinic reported that the most common symptoms the children had experienced before going to the clinic were were gastrointestinal (vomiting, diarrhoea, bloody stools and stomach pain) affecting 41%, skin symptoms (hives, swelling, rash and eczema) affecting 39% and respiratory (wheezing, coughing, difficulty breathing, runny and blocked noses) affecting 9%.

A Japanese review of 125 children who had undergone food challenges at their clinic between 2004 and 2010 reported that the majority (82%) had suffered from skin symptoms, around half (51%) had suffered respiratory symptoms, 12.7% had suffered from gastrointestinal symptoms and the same surprisingly large number had suffered from anaphylactic symptoms. Their numbers were similar to those reported in another Japanese study, but are different from the types of symptoms reported in American and Israeli studies, a difference researchers thought may be due to the different types of soy food involved.

The second category of symptoms is those that are caused by a secondary allergy to soy provoked by a cross-reaction with birch pollen.

This type of allergy is mainly associated with the milder symptoms of oral allergy syndrome (OAS), which include:

• Itching/burning lips / ears / tongue / palate / throat
• swelling (oedema) of the skin inside of the mouth / cheeks / lips / tongue / around the eyes
• Hay fever-like symptoms: teary, itchy eyes, runny nose
• Hives on the face and neck
• Eczema on the face and neck
• Sneezing

Systems generally appear within half an hour of eating the offending food and, because they are provoked by a related pollen allergy which often appears later in life, they tend to affect older children and adults more often than younger children, although pollen-related food allergies are increasingly affecting younger children.

In about 10% of the cases, people who are suffering a pollen-related reaction to food will go on to have more serious allergic symptoms.

A European study of 30 soy-allergic adults and children reported that 21 had a pollen-related soy allergy. 1 of those people had a case history of anaphylaxis and one suffered from symptoms of anaphylaxis during the food challenge. Most of the other soy- and birch pollen-allergic subjects had histories of OAS, many with accompanying skin and/or gastrointestinal symptoms, and most of those people suffered from all 3 types (skin, gastrointestinal and respiratory symptoms) during the food challenges.

Several things can worsen allergic reactions to soy in people also allergic to birch, including:

• having large amounts, i.e. drinking soy-based drinks, taking soy-based supplements or eating a lot of minimally-processed soy foods like tofu or sprouts all at once
• fasting
• taking antacids like proton pump inhibitors (PPIs)
• birch pollen season

Birch pollen season can also worsen the reactions of soy-allergic people who are primarily allergic to soy allergens not related to pollen. This was revealed in a Dutch study which described the case a 26-year-old Dutch woman with a primary allergy to soy but whose symptoms got worse during the birch pollen season. Lab tests revealed that, during the pollen season, her IgE levels to 3 soy allergens increased, but the ones to the pollen-related allergen increased 3.3.-fold. The researchers theorised that her worsening symptoms were probably due to a seasonally over-reactive immune system.

Soy can also provoke anaphylaxis, whether you’re allergic to the soy itself (primary food allergy) or to birch pollen which is cross-reacting with soy (secondary allergy).

In 1997, a study carried out in Spain stated that ‘severe allergic reactions’ has been caused soy ‘hidden’ in processed foods including typical Spanish sausage products, boiled ham, soup stock cubes and doughnuts.

In 2007, an Italian report described the case of a 32-year-old woman who had suffered multiple episodes of hives, tongue and facial swelling, bronchospasm and anaphylaxis that had required her to have emergency treatment on 4 occasions. The latest episode was due to soy flour in a pizza base.

In 2016, an Italian man was reported to have suffered from severe stomach pain, vomiting, diarrhoea and hives after eating a risotto. The symptoms subsided after he vomited the food out. But 20 years previously, he had suffered an anaphylactic attack after eating a cracker containing soybean, and another one a few years later after eating a pizza. It was an unusual case in that he was 69 years old when he had his last reported attack, and had obviously become allergic to soy during his adulthood. Additionally, although it was determined that he was allergic to soy, he was not allergic to any other legume, including the more well-known anaphylaxis-causing legume, peanut.

In 2018, a case of anaphylaxis to a chocolate drink was reported in 38-year-old Spanish man. That case was unusual in that it involved an allergen that had not previously been identified in a case study (Gly m Bd 30K).

Generally when a soy-based drink or dessert causes anaphylaxis, it does so in someone who is also allergic to birch and its equivalent soy protein (Gly m 4). So far this has been reported in 4 cases in Sweden, 7 cases in Germany and 6 cases in the Netherlands. What these countries all have in common is an abundance of birch or alder trees (which are in the birch family).

In 2013, a German review of food-induced anaphylaxis in adults seen at one allergy centre in Berlin between 2007–2011 identified soy as the third most common cause (responsible for 21 cases).

A 2023 review of data from the European Anaphylaxis Registry (covering 100 tertiary allergy centres in 10 European countries and Brazil) reported that soy was a common trigger for anaphylaxis, notably in Germany, France, and Switzerland, and the fourth most common cause of food anaphylaxis in adults. Because nearly all of the affected people also suffered from hay fever, the researchers concluded that they were suffering from secondary soy allergy provoked by birch-pollen.

Don’t panic: To be clear, the official definition of anaphylaxis is probably not what you think it is.

This means that many of the cases of anaphylaxis reported in medical studies are not actually life-threatening—when dealing with an emergency, however, since it’s impossible to predict which reactions will become life-threatening, every case of anaphylaxis should be treated as if it is potentially deadly.

In the 2023 analysis mentioned previously, researchers found that anaphylaxis to soy is generally milder than anaphylaxis triggered by any other major food allergen (except for celery), with two thirds (65.7%) of the cases they looked at having been registered as a grade 2 reaction (out of 4 possible grades).

However, there are reported cases of fatal reactions to soy. One case was reported in 2002 in France. Other reports are even older.

A 1991 analysis of deaths to anaphylaxis in America describes one death due to sausage pizza fortified with soy protein.

Three years later, a Swedish report describes a fatal anaphylactic reaction due to a hamburger with added soy protein.

A Swedish team also looked through a national register of severe reactions to food and found 61 cases reported between 1993 and 1996, 4 of which were fatal anaphylactic reactions in soy-allergic to adolescents caused by meatballs, hamburgers and kebabs containing soy. The symptoms were initially mild and were then followed by a symptom-free period for 30–90 min which rapidly degenerated into fatal cases of asthma. Notably, all four of the youngsters were severely allergic to peanuts and did not known that they were allergic to soy.

There has been some discussion as to whether the foods that these children ate were contaminated with peanut protein, or whether the methods used to identify the type of protein they had eaten was flawed. The debate exists because people simply do not expect soy to be a cause of severe or fatal allergic reactions. A 2021 study of specific foods causing anaphylaxis that spanned 41 countries around the world reported that ‘soya was not a major cause of food anaphylaxis in any region’. Including Sweden. But sometimes, it is..

A third category of soy-induced immediate symptoms are respiratory ones due to aeroallergens. They include hay fever (rhinoconjunctivitis), sneezing, coughing, wheezing, severe shortness of breath, or difficulty in breathing (dyspnoea), and immediate or late onset asthma

Soy aeroallergens generally tend to affect people who work with soy; freight and harbour workers who deal with large shipments of soy, crop harvesters and mill workers, as well as workers at soy-processing facilities, food processing plants and food supplements manufacturers. And quite a few bakers.

Soy proteins have also been responsible for a rare case of hypersensitivity pneumonitis, aka bird fancier’s lung, farmer’s lung, hot tub lung or humidifier lung. In this case, the allergic inflammation was in the lung of a female veterinary diet researcher who had spent 20 years specialising in the research of animal diets that happened to contain soy.

In rare cases, sensitisation to soy via an inhalant allergen can also lead to a food allergy, as was the case for a man who worked as a factory producing bakery improvers (mixtures to improve the quality of baking dough). The 33-year-old worked for 8 years at the factory, unloading raw materials and handling soy products before starting to suffer from hay fever-like symptoms and asthma while at work. His symptoms gradually became worse until he also started experiencing oral allergy syndrome, stomach pain and vomiting when he ate soy products (biscuits, milk or yoghurt) which he had previously eaten without problems.

A fourth and final category of soy-induced immediate symptoms is protein contact dermatitis, which is when a person develops a skin rash or localised hives 15-30 minutes after coming into contact with soy. It has been reported as a consequence of touching food, like the case of this woman whose hand eczema rapidly worsened after contact with tofu or, more commonly, when someone uses a cosmetic product containing the offending food.

Delayed reactions to soy

Delayed allergic reactions can occur hours or even days after exposure to an allergen, unlike IgE-mediated reactions that often happen within minutes. These reactions either involve diseases that rely on cell-mediated mechanisms (immune responses that do not rely on the production of IgE antibodies but instead involve the activation of T cells and macrophages which leads to inflammation and tissue damage) or by ‘mixed’ diseases that rely on both IgE- and cell-mediated mechanisms.

The most common types of delayed reactions to soy are eczema (atopic dermatitis), food protein-induced allergic proctocolitis, food protein-induced enteropathy, food protein induced enterocolitis syndrome and eosinophilic oesophagitis.

Eczemais a chronically relapsing inflammatory allergic condition that specifically affects the skin and looks like this. It’s classified as a ‘mixed’ form of allergy that can produce either immediate or delayed reactions that can occur up to 48 hours after eating a trigger food.

In 2012, German researchers analysed the results of 1843 consecutive oral food challenges and revealed that soy was the food most likely to provoke a worsening of eczematous symptoms.

Food Protein-Induced Allergic Proctocolitis (FPIAP) is quite common and considered relatively benign. It mostly affects infants and symptoms generally appear within the first 6 months (often within the first month) of life. The most common symptom is blood-streaked stools (hematochezia) which is sometimes accompanied by:

• mucous in the stool
• mild diarrhoea
• stomach ache
• gas (flatulence)
• anal chaffing/pain on defecation
• refusal to eat and irritability

Otherwise, the child is perfectly healthy-looking and growth is unaffected.

Food Protein-induced Enteropathy (FPE) affects the small intestine, resultingin digestive symptoms including:

• intermittent vomiting
• chronic diarrhoea
• malabsorption (steatorrhea) which can be accompanied by a ‘failure to thrive’ (FTT), that is, a failure to show proper growth
• rarely: bloody stools

Food protein-induced enteropathy (FPE) is a subset of protein-losing enteropathy (PLE);; PLE is an umbrella term that encompasses a variety of diseases that lead to excessive protein loss in the gastrointestinal tract, while FPE is a specific type of PLE triggered by reactions to food proteins.

One team of scientists has proposed the catchy term ‘food-protein induced protein-losing enteropathy’ (FPIPLE) to refer to the condition, with the following signs to identify it:

• below-average weight gain (their current weight or rate of weight gain is significantly below what is expected for their age, sex and ethnicity)
• and/or swelling (oedema) brought on by abnormally low levels of protein in the body (hypoproteinemia)
• anaemia

Food Protein Induced Enterocolitis Syndrome (FPIES) is a delayed allergic reaction to food that affects the gastrointestinal (GI) tract. There are two main types of FPIES, chronic and acute.

Chronic FPIES is quite rare and occurs mostly in infants who eat the trigger food on a daily basis. It can be recognised by intermittent vomiting and diarrhoea and, occasionally, failure to thrive (which means that a child is not getting in enough calories to reach a similar weight and size to other children of the same age and sex). Happily, once a diagnosis is made, eliminating soy from their diet allows them to make a quick recovery. Cases of chronic FPIES in adults are vanishingly rare, but not unheard of.

Acute FPIES is by far the most common form. In children, symptoms often occur within 2 to 4 hours after eating the offending food and can include:

• vomiting
• pallor
• lethargy
• dehydration
• diarrhoea
• shock or hypotension (i.e. low blood pressure) which can manifest as dizziness, fainting or blurred vision (as well as pallor and lethargy)

Sometime children with chronic FPIES end up developing acute FPIES and this form of FPIES can, in turn, develop into an IgE-mediated form of allergy.

In adults, the typical symptoms of acute FPIES are not the same as those seen in children. They can also appear faster. They include:

• stomach pain and cramps
• diarrhoea
• vomiting
• dehydration
• transient weakness and shivering (possibly hypothermia)
• lethargy
• weight loss

People who experience severe symptoms of acute FPIES may have a longer-lasting form of the disease.

In Australia, a rare case of soy allergy in an infant who was exclusively breast-fed has been reported. In this case, the infant first showed acute symptoms—profuse vomiting, pallor and diarrhoea requiring the administration of intravenous fluids—after being fed his first bottle of formula when he was 5 months old. He reacted with severe symptoms again a month or so later after being breast fed by his mother who had had a large helping of soy-based ice cream 12 hours earlier. What was particularly remarkable was that his mother had been eating processed foods containing soy the whole time but the infant had never shown any sign of the (milder) symptoms of chronic FPIES. The allergists suspected that, after the ice cream, his mother had just happened to feed him at a time when the soy excretion in her breast milk was at peak concentration.

These types of allergies do not all fit into neat little boxes. You can get cases involving a mixture of conditions, like the case of a 5-month-old girl who had both FPIES and PLE.

Eosinophilic oesophagitis (EoE) is an inflammation of the oesophagus caused by a food allergy, environmental allergens or acid reflux. It is characterised by symptoms including:

• food impaction; this is when food becoming stuck in the oesophagus which can lead to a sensation of squeezing in the chest, and can be accompanied by excessive salivation (unlike choking, a person can still breathe and talk, but they cannot eat or drink any more)
• difficulty swallowing (dysphagia)
• abdominal pain
• reflux (the flow of liquid back from the stomach into the oesophagus)
• vomiting
• heartburn (pyrosis)
• stomach pain
• food refusal

Symptoms are variable and often age-dependent. In infants, EoE tends to provoke general symptoms of oesophageal difficulties such as gagging, vomiting, feeding difficulties and weight loss, or so-called ‘failure to thrive’. Young and school-age children have symptoms that are indistinguishable from those associated with gastroesophageal reflux, such as abdominal pain, vomiting and an unpleasant taste in the back of the mouth that comes from regurgitating sour or bitter liquid. Older children and adults are more likely to have trouble swallowing and to get food lodged in their oesophagus (food impaction) and, less commonly, to suffer from heartburn.

Because symptoms can be severe, if you do suspect that you’re allergic to soy, it’s important that you see your GP/family doctor and get a referral to an allergy clinic for further testing.

Threshold for reactions

VITAL®, the Australian initiative for voluntary incidental trace allergen labelling, put out recalculated threshold doses for the ‘Big 14’ allergenic foods in 2020. Using a database containing datasets from studies carried out worldwide that used double-blind, placebo-controlled food challenges (DBPCFC), they calculated that the lowest threshold dose of protein that was needed to produce a reaction in 1% of the population allergic to soy is 0.5 mg. (Note: in this case, the ‘population allergic to soy’ is 87 people who were given a DBPCFC)

41.9 mg was the dose needed to produce a response in 10% of the test subjects, and 1779 mg was the dose needed to provoke a reaction in half of the test subjects.

Note: we’re talking about milligrams of soy protein.

Whole-fat soy flour (used to make bread) contains about 40% protein and a loaf of bread generally contains 0.5 to 1% soy flour. A typical slice of bread weighs 38 grams so 2.5 slices of bread weigh about 100g. Assuming they come from a loaf containing 1% soy flour (40% of which is protein), those 2.5 slices contain 0.4g of soy protein (100 grams weight x 0.01 percent soy flour x 0.4 percent soy protein in soy flour) and 1 slice contains 0.16g (160mg) of soy protein. So half of the soy-allergic population could theoretically eat about 11 slices of bread without reacting, and double that if the loaf contained 0.5% soy protein.

A doughnut, which is often yeast-raised weighs about 38g. But it’s made of defatted soy flour which contains 50–54% protein, and each doughnut contains about 2% soy flour. Assuming a doughnut has been made from dough containing 2% soy flour (50% of which is protein), 2.5 doughnuts would weigh about 100g and contain 1g (100 grams weight x 0.02 percent soy flour x 0.5 percent soy protein in soy flour) of soy protein, so one doughnut contains 0.4g (400mg) of soy protein. So half of the soy-allergic population could theoretically eat about 4.5 doughnuts without problems (not even a sugar rush).

As difficult as that was to get through, it’s not the whole picture either, because it doesn’t take into account which allergen a person is allergic to, which will also affect what they can eat. (Someone allergic to birch pollen, for example, could probably as many doughnuts as they wanted and only suffer the consequences of eating way too much junk food. YOLO)

The dose of soy required to provoke a reaction in a very sensitive person is actually relatively high—of all the foods defined as major allergens, only fish, shrimp, lupin and wheat require a higher initial dose according to the VITAL project and, by the time you reach the threshold amount needed to provoke a reaction in about 80% of the soy-allergic, only shrimp comes in at a higher dose.

According to one study, compared to peanut, hazelnut, milk and egg, the amount of soy needed to provoke a reaction is over a 100 times more.

That said, the amount of soy that is needed to provoke a reaction says nothing about how severe the reaction will be.

Ultimately, the threshold dose needed to provoke symptoms varies widely between people. It also varies per person, depending on the circumstances around the meal.

Your threshold can be lowered and your allergic reactions worsened by things called ‘cofactors’. Cofactors include things like how much you eat and whether those ingredients have been cooked or processed, as well as exercise, anti-inflammatory drugs, alcohol, infection and stress.

Soy has been specifically implicated in several cases of food-dependent exercise-induced anaphylaxis (FDEIA).

Cofactors are thought to play a role in about 14% to 30% of all anaphylactic reactions.

Please note: the amount of soy needed to provoke a reaction says nothing about how severe the reaction will be. And, even if your reactions have been mild in the past, it does not mean that they will continue to be mild.

Image by Ivan Samkov on Pexels

Diagnosing soy allergy

A diagnosis of soy allergy will primarily be based on your clinical history—a record of consistent symptoms following the consumption of soy or soy-containing foods. This will require you to provide your allergist with answers to questions about your general medical background (including any other allergies you may have and relevant illnesses in your family) and your dietary history (what you ate to provoke your symptoms, what those symptoms were, how long they lasted, whether you had exercised or taken painkillers and many other details).

Your medical history determines what comes next; on the basis of your answers, the allergist will try to determine what type of allergy you have—a primary, immediate-type allergy, a cross-reactive allergy or a delayed-type allergy—or whether it could be something else, and this will determine the tests they ask for to come up with a diagnosis.

Diagnosing IgE-mediated reactions to soy

Skin tests

An IgE-mediated sensitisation to soy is typically confirmed by a skin prick test, which involves someone placing a small sample of soy extract onto your skin (generally the forearm of an adult/older child or the upper back of a young child) and pushing it through the top layer of skin by pricking it with a lancet. It takes about 15 minutes to see a reaction (or not).

This test is often carried out first because it is quick and simple to perform and gives rapid results, but it is generally used to rule out an allergy rather than to confirm one, because it has excellent negative predictive value—if the skin weal is under a certain size, you are highly unlikely to have an allergy—but poor positive predictive value—the skin weal has to be very large before an allergist can say with any kind of confidence that you probably have an allergy.

The accuracy of the skin prick test can be limited by the fact that the processing of commercially-made allergen extracts can destroy the heat-sensitive allergens, notably those that people allergic to birch pollen react to. In such cases, someone could get a false negative result and be wrongly told that they are not allergic to soy.

For example, in a study of Belgian birch-allergic children with a potential soy allergy, none of the children who turned out to be allergic to soy had a positive test with a commercial skin testing extract, and in a study of Swiss adults with a known soy allergy, only 45% of the subjects had a positive response to the commercial skin testing extract.

One solution to this problem is the prick to prick test. This test is very similar to the skin prick test, except first the lancet is used to puncture fresh food—such as mildly processed soy drinks or soy flour—and then it is used to prick your skin. When the food is in liquid form, the technique is actually the same as the one used for the skin prick test and, when the food is solid, it’s often ground down and put in saline solution.

The prick to prick test often produces superior results to commercial extracts because the fresh food used should contain all of the allergens that a person can react to. The lab used by the clinic can also prepare the extract in specific ways which may add to its efficacy.

Blood tests

Sometimes, the doctor may decide to order a blood test, aka an immunoassay. Perhaps the skin prick test was inconclusive, or the suspected allergen is not available for skin prick testing, or you’re unable to undergo the test for some reason.

A blood test involves having a small sample of blood drawn so that it can be sent to a lab where technicians will use allergen extracts to check whether there are IgE antibodies in your blood that react to them. It can take 1 or 2 weeks to get the results.

Blood tests can be less sensitive or specific than skin tests, but they have other advantages: they are perfect for people who cannot stop taking certain medications or have extensive skin disease or tattoos, and they can safely be used on infants, squirming toddlers and people who are at risk of suffering an anaphylactic reaction.

Blood test panels also typically include a whole range of potential allergen extracts including other foods or aeroallergens that the allergist may want to check your reaction to.

However, whereas IgE levels to major food allergens like milk, egg, peanut and fish can be used to predict whether or not a person will have symptoms when they eat a food—i.e., are allergic to it—those for soy have been found to have poor predictive value.

Although some researchers claim to have found soy-specific IgE levels (cut-off points) that can predict whether or not someone will pass a soy food challenge and whether or not someone has outgrown their allergy to soy several more have failed in their efforts to find particularly meaningful cut-offs.

For more specific information, a component blood test—aka Component Resolved Diagnosis (CRD)—can be carried out. Instead of using extracts of whole foods containing only (heat-stable, plentiful) allergens, the CRD tests the reaction of IgE antibodies in your blood to isolated, individual proteins. This improves the diagnostic sensitivity of the test as allergens that would otherwise be missing from the whole food extract or exist only in tiny amounts are present in concentrated form in the CRD test.

This type of test enables the doctor to see exactly which allergen(s) you react to, which allows them to determine whether you are sensitised to cross-reactive allergens that are unlikely to produce symptoms, and whether you are sensitised to certain allergens that could affect your management plan.

For example, if you’re sensitised to the Gly m 4 allergen which is associated with birch pollen allergy, you’re more likely to suffer from mild symptoms when you eat or drink mildly processed soy products, but you probably don’t have to avoid cooked or extensively-processed soy foods (although you should avoid having lots of soy protein at once, like you’d find in a glass of soy milk).

However, if you’re sensitised to the more stable and heat-resistant Gly m 5 and/or Gly m 6 allergens, you’re more likely to show symptoms and to have severe reactions and will have to avoid all types of soy-containing food, even when it’s cooked and highly-processed.

A sensitisation to Gly m 8 has also been associated with a greater chance of showing symptoms, although it is still far from foolproof.

Component blood tests are also made up of very large panels of allergens which include many other foods and aerollergens that the allergist may want to check your reaction to and can help to determine whether or not a sensitisation to a cross-reactive allergen will be symptomatic or not.

Unfortunately, although CRD could potentially reduce the need for oral food challenges and contribute to tailored management plans, it’s not yet considered a routine diagnostic method and it’s not comprehensive; the most widely used tests neither contain all of the identified allergens (which are also not all of the possible allergens), nor are they universally available. Testing for certain specific soy allergens would require special preparation and is therefore only likely to be done for research purposes.

Additionally, as it is with skin and standard blood tests, CRD is better at confirming an allergy than at eliminating the possibility of one. And, because sensitisation patterns differ according to geography and populations, with different allergens being more important in different regions and in people of different ages, allergists need to understand their patient populations so that they interpret the results of the tests correctly.

In some geographical areas, for example, a sensitisation to Gly m 4 is more likely to indicate the potential for serious symptoms, and a sensitisation to Gly m 5 or Gly m 6, milder symptoms. That’s why you need an expert to look at your test results.

A positive skin or blood test does not mean that you are allergic to something. While skin prick tests and blood tests help with diagnosis, positive results only show sensitisation to specific allergens. Being sensitised to a food doesn’t mean that you’re allergic to it and that you will develop any symptoms.

For example, when Italian researchers recruited 131 children with positive skin prick tests to soy and gave them all oral food challenges to find out how many would actually react, only 8 of them did.

And when Japanese researchers went through their medical records and identified children who had tested positive for a sensitisation to soybeans, they found that only 307 of 1710 (just under 1 in 5) actually showed symptoms after eating soybean. Similarly, Czech researchers reported that only 5 of the 52 adults and children with eczema who had tested positive to soy actually had symptoms after eating it. And Dutch researchers investigating the use of blood tests in the diagnosis of soy-allergic adults concluded that ‘a challenge test it still needed to confirm a clinically relevant soy allergy.’

A positive test result simply means that your immune system is specifically aware of an allergen or allergens in that food. Why some people later develop an allergy to that food, and some do not, is not yet known.

Neither can the results of your blood or skin test predict how severe your reaction to eating some soy might be; a large skin weal or high level of IgE in your blood do not mean that you will have a serious allergic reaction if you happen to eat a soy-containing sausage roll.

Similarly, Japanese researchers identified 1710 children of all ages who had tested positive for a sensitisation to soybeans and found that only 307 (18%) of them showed symptoms after eating soybean, and a Czech study reported positive soy test results in 52 adults and children with atopic dermatitis, but only 5 (9.6%) had symptoms after eating it. Dutch researchers investigating the use of blood tests in the diagnosis of soy-allergic adults concluded that ‘a challenge test it still needed to confirm a clinically relevant soy allergy.’

Neither can the results of your blood or skin test predict how severe your reaction to eating some soy might be; a large skin weal or high level of IgE in your blood do not mean that you will have a serious allergic reaction if you happen to eat a soy-containing sausage roll.

Food challenge

The only way to get a definitive diagnosis of soy allergy, and to have some idea of how severe your reactions may be and how much soy is needed to provoke them, is to undergo an oral food challenge. This generally involves eating a very small amount of soy (often in the form of soy powder, tofu or milk), waiting for a reaction, and then doing it again, gradually increasing the dose until an objective—visible—reaction occurs or a maximum dosage is reached. It can take around 4 hours, depending on the type of challenge undertaken and the length of observation time needed.

You can read more about oral food challenges here.

Oral food challenges are generally undertaken either when someone’s history and their test results disagree (i.e. they have negative tests results but their history strongly suggests an allergy, or vice versa) or to check whether someone has outgrown their allergy to ensure that they don’t unnecessarily restrict their diet or worry about hidden allergens in processed foods.

Because of the risk of severe reactions, oral food challenges should only be done by an experienced consultant in a medical setting.

In general, however, challenges to soy are less likely to produce severe reactions than challenges to the other major food allergens. They tend to require much larger doses and most people experience skin and gastrointestinal symptoms.

Although food challenges help to diagnose food allergies and identify a suspect food, there are other reasons to undergo food challenges, namely:

• to identify culprit foods in cases of allergies to multiple unknown foods
• to determine a patient’s threshold—how much soy they can eat without reacting—so that dietary advice based on the outcome of the challenge can be given
• to confirm the development of tolerance to soy

Practically speaking, most people do not undergo this kind of test since it requires a lot of time and resources. And oral challenges are rarely, if ever, offered to people whose history includes severe reactions to a suspected food. Whenever possible, allergy diagnoses are based on a combination of medical history and lab tests.

Diagnosing Non-IgE-mediated and mixed reactions to soy

Non IgE-mediated diseases are difficult to diagnose for several reasons, not least of which is the fact that IgE testing is often of no use. This makes the clinical history especially important for the diagnosis of these types of conditions. Even then, the symptoms are not easy to connect to the actual meals because of the time delay, and the symptoms associated with digestive allergies lack the skin and respiratory signs that doctors usually associate with allergy.

Skin tests

Eczema is diagnosed based on personal and family history of allergy and a skin examination. While there are no standard diagnostic criteria, there are certain features that a doctor can look for to diagnose it.

That said, these criteria are based on the characteristics of paediatric eczema, which is not the same as the manifestation of eczema is adolescents or adults, making diagnosis of eczema in older age groups more challenging. Sometimes people with eczema in these age groups will have to undergo additional tests to rule out other diseases first and a skin biopsy may be needed before a diagnosis of eczema is made. However, these differences is now being taken into account and guidelines are being updated.

Once the diagnosis of eczema is made, efforts will first be made to try and get the skin condition under control using topical skin creams and drugs before any further testing is done. Generally, only if the skin is not getting any better will tests be carried out to see whether allergens, like food, could be aggravating the condition.

The identification of potential food allergens is generally done by looking for specific IgE antibodies to a food using skin prick tests or blood tests (the latter is often used if the skin condition is too bad for a skin test, or medications are being taken that will interfere with the results or if the tests involve a young infant).

In cases of delayed allergy, doctors may use the atopy patch test (APT). This test generally involves walking around with food (either fresh or in solution) contained in tiny aluminium capsules taped to your back for up to 3 days and having your skin checked for a reaction after 48 hours and 72 hours.

The APT has been used relatively often used to try and diagnose food allergy in people with eczema, with mixed results. Some researchers have found that it is, in combination with skin prick tests, quite reliable, others have found that it’s a lot of bother for little added diagnostic value (indeed, it can provide a lot of false positive results with irritated skin) and have concluded that a food challenge is the only way to be sure of an accurate diagnosis.

That said, the APT has been used successfully in some cases, enabling doctors to recommend elimination diets that have led to clear improvements for some people, but not for everyone (primarily because their skin conditions were also triggered by other things).

The atopy patch test has also been used to try and diagnose delayed digestive allergies, with mixed results; in the case of food protein–induced enterocolitis syndrome (FPIES), for example, it has proven itself to be both ‘a promising diagnostic tool for the diagnosis of FPIES’ and ‘not helpful in identifying the [trigger] foods’, while showing ‘poor utility in the follow-up prediction of outgrowing FPIES in children’. With food protein-induced allergic proctocolitis (FPIAP) and food protein-induced enteropathy (FPE), it has shown low sensitivity and with eosinophilic oesophagitis (EoE), it has shown that it can ‘identify potential causative foods’. For these diseases, it is not the diagnostic instrument of choice.

A skin application food test (SAFT) may be used instead for children under the age of 4. It’s basically the same thing, but the capsule of food is only applied to the skin for 10 to 30 minutes. It’s had mixed results, having been described as reliable and child-friendly and also as not being as good as the APT in diagnosing certain food allergies.

Elimination diets and food challenges

While skin tests may provide an indication of sensitisation, they cannot diagnose a food allergy; that has to be done with afood challenge during which the doctor can see whether or not, in addition to the immediate reactions, the suspected food also produces a worsening of the skin symptoms within the next 48 hours (often within a day). If it does, the food can then be eliminated from a person’s diet and their skin condition will be monitored for the next few months to see if there is a persistent improvement. When more than one food is suspected, the next challenge will be done a few weeks after the first one.

The diagnosis of delayed digestive allergies generally starts with exclusion; first other possible causes of the symptoms are eliminated and only then will the suspected food(s) be excluded from a patient’s diet—and, if they are breastfeeding, from their mother’s diet, too.

If the symptoms disappear, the first step towards a diagnosis involves reintroducing the foods one by one into the diet and seeing if the symptoms return. If the symptoms don’t disappear, it could be that the diet has not been restricted enough or that other foods should (also) be considered for elimination. Or it could be that something other than an allergy is responsible, in which case, the allergist’s job ends and another specialist’s begins.

When it comes to food protein-induced allergic proctocolitis (FPIAP), as it’s generally quite a benign condition, the allergists may want to wait for 2 to 4 weeks to see if the condition goes away by itself. If symptoms persist, a diet that eliminates the offending food should cause them to go away, often within 72–96 hours. The food can be (briefly) reintroduced into the infant’s diet 1 to 2 months later to (unofficially) confirm the diagnosis once symptoms reappear. Theoretically, an oral food challenge is necessary to officially confirm the diagnosis but, in practice, this isn’t done if the infant looks otherwise healthy.

The diagnosis of food protein-induced enteropathy (FPE) relies on seeing how a patient responds to an elimination diet and then performing an endoscopy and biopsy to check on the state of the small intestine. If the right food is eliminated, the symptoms should disappear and the tissue samples should look normal within 1 to 4 weeks.

The diagnosis can be confirmed by an oral food challenge, which essentially involves reintroducing the food into the diet 1 or 2 months after it was eliminated (which can be done at home if previous reactions have not been severe). If the food produces vomiting and diarrhoea within 1.5 to 3 days after being eaten, the diagnosis is confirmed and it can be taken out of the diet. Most children outgrow the condition by the time they’re 2, sometimes 3 years old.

The diagnosis of food protein induced enterocolitis syndrome (FPIES) mainly relies on a person’s clinical history and symptoms appearing when the offending food is reintroduced after an elimination diet.

In the case of chronic FPIES, an elimination diet should result in the symptoms going away within 3 to 19 days. Reintroducing the trigger food should produce the symptoms of acute FPIES—i.e. projectile vomiting—which should be enough confirmation.

In the case of acute FPIES, eating the offending food should be followed by symptoms that should fit specific diagnostic criteria including copious vomiting within 4 hours. Although confirmation of the diagnosis officially requires an oral food challenge, because it often produces nasty symptoms that the patient quite rightly has no wish to suffer through, in practice, this is rarely done and challenges for the diagnosis of chronic FPIES are more common.

However, since FPIES symptoms tend to be different for adults with acute FPIES and there are no strict diagnosis guidelines for them, oral food challenges are often necessary.

There are other reasons to undergo food challenges in cases of FPIES, including:

• identifying a culprit food in cases of allergies to multiple foods
• confirming the development of tolerance to a trigger food, which is often done between 12 to 18 months after the most recent reaction although, in the case of soy, a challenge to see whether the allergy has been outgrown may be considered when an infant is older than 6 months

Many clinics will only carry out a food challenge in an infant to see whether they have outgrown their allergy.

Although the majority of people with FPIES will have negative skin or blood tests to their trigger food, in some cases people do have an IgE sensitisation too. This is called ‘atypical FPIES’ and it affects between 1 in 4 and 1 in 8 people with FPIES. According to American research, the foods most commonly associated with this type of FPIES are egg, milk and peanut, but this may just be because those are the foods most typically eaten by American children (who make up the bulk of these studies). A person can have atypical FPIES to several foods, and those foods can include anything, from shrimp to avocado.

Some children with atypical FPIES may take longer to outgrow their condition (if, indeed, this happens at all) or may develop a classic IgE-mediated food allergy with potentially more dangerous symptoms. As such, periodic testing for an IgE sensitisation is advised in children who also have an IgE-mediated food allergy to other foods or suspected food-induced eczema.

When diagnosing eosinophilic oesophagitis (EoE), other conditions that produce similar symptoms, like gastroesophageal reflux disease (GERD), are first eliminated as a possibility before any intrusive testing is done. Then, if eosinophilic oesophagitis is still suspected, an upper GI endoscopy (aka an oesophagogastroduodenoscopy) and biopsies are carried out to look for specific levels of eosinophils in the oesophageal tissue (15 or more eosinophils per high-powered field, to be precise).

Standard elimination diets for cases of EoE are often based on the most common causes of the disease, either ‘2 food diets’ (dairy and wheat), ‘4 food diets’ (dairy, wheat, egg, and legumes) or ‘6 food diets’ ((dairy, wheat, egg, legumes peanuts/tree nuts and fish/shellfish). These are called ‘empiric’ diets, i.e. diets that are based on observation and experience. The diet can be made less cumbersome by starting small, first with one food (i.e. milk) or two foods aand then eliminating more foods if the symptoms don’t disappear.

The empiric diet approach is not the only approach. Sometimes the foods to be eliminated are determined using lab tests—atopy patch test and SPT and/or blood test—first (a targetted approach). Both methods work equally well for both children and adults although the targetted approach has the advantage of often requiring the elimination of fewer foods. That said, a lot of people with Eo don’t have any measurable IgE antibodies to their trigger food, so the targetted approach can only help some.

The diet normally takes about 6 weeks. If the symptoms go away and the tissue samples look good, the trigger is assumed to be one or more of the foods that was eliminated. In order to pinpoint the trigger(s), each food is reintroduced back into the diet one by one. If a reintroduced food causes symptoms to return and/or biopsy specimens to look abnormal, then it is identified as a trigger food and must be eliminated from the diet indefinitely. (In the real world, children may balk at undergoing so many intrusive tests or there may not be the capacity to perform them, in which case, they will probably not be required for a diagnosis.)

Elimination diets are best performed under expert guidance, because there is a risk that excluding a food from your diet because you think that you may be allergic to it or because it causes mild or delayed symptoms can lead to you developing an IgE-mediated allergy to that food, often with severe—sometimes fatal—reactions.

The good news is that, when a delayed digestive allergy is diagnosed, excluding offending food(s) from the diet leads to the resolution of symptoms and the repair of the underlying tissue damage in most children and adults, and improves quality of life, even though the diet may be difficult to stick to. When it comes to young children, if staple foods are being eliminated, dietary guidance may be needed to ensure normal growth and development.

Image by 麗娜 顏 on Pexels

Image by Jakub Żerdzicki on Unsplash

Good to know

People (especially children) with soy allergy tend to be atopic—that is, have the tendency to develop allergies.

For example, a study of 30 soy-allergic people from Switzerland, Denmark and Italy found that 29 had at least 1 atopic disease; 10 had eczema, 20 had asthma and 21 had hay fever.

An American review of patient data identified 133 children with soy allergy and reported that almost two thirds (64%) had asthma, almost three quarters (71%) had allergic rhinitis and about four fifths (81%) had eczema. An earlier chart review of American patients focussed on food challenges and reported that almost all of 75 the children (95%) who failed a food challenge with soy had an allergy to another food. Just over two-thirds (68%) had eczema, just over half (52%) had asthma and over a third (41%) had hay fever symptoms.

Note that the reverse is not necessarily true: just because you’re atopic does not mean that you have a strong chance of being allergic to soy. An Italian study investigating the presence of soy allergy in atopic children gathered two groups together: in the first group were 505 children with a history suggestive of food allergy. In the second, 243 5-year-olds whose parents were atopic and who had been fed soy protein formula for their first 6 months of life in an effort to try and prevent the development of milk allergy (that probably didn’t work, see later). In the first group only 6 (1.2%) had a positive food challenge and in the second group, only 1 (0.4%) failed the food challenge.

You don’t have to work with soy to develop a respiratory allergy; being in the general vicinity of people working with soy can be enough.

In the early 1980s, mysterious outbreaks of asthma broke out in Cartagena and Barcelona, both port towns in Spain. Suddenly, pockets of people suffering with respiratory allergy started to materialise, their symptoms arriving suddenly and then disappearing again, only to pop up somewhere else in the area. Some people even died. The symptoms were not attributable to ‘normal’ pollution and there was no discernable pattern to the outbreaks.

An investigation took about 8 years to uncover the source of the phenomenon. The discovery that these outbreaks only seemed to occur on sunny days when there was a sea breeze pointed investigators in the right direction; towards the harbour. The times of the outbreaks were the final clue needed to determine that it was the unloading of soybeans from ships that was causing the problems. An allergic reaction to soybean dust was then confirmed as the cause of the asthma.

Subsequent studies demonstrated that soybean allergens could be detected in the air in Barcelona during ‘epidemic days’ and identified the culprit allergens in the soybean hulls. A subsequent investigation of the victims of the ‘soybean epidemic’ found that, although their symptoms had improved in the years after the outbreaks, just over half of them were still sensitised to soybean 8 years later.

A few years later, an American team looking into the bouts of epidemic asthma that had occurred in New Orleans in the 1950s and 1960s determined that they had likely also been caused by ships in the harbour carrying loads of soybean.

Later research found that soybean allergens were present in the air around Spanish soy unloading and/or processing plants, animal feed plants and pig stables, and that the food-allergic living in rural Argentina, a big soybean producer country, were often sensitised to soy because they repeatedly inhaled soybean dust in their environment, and that the air in southwestern Ontario, (3) Canada, an area of intensive soybean cultivation and processing, was filled with allergens during the soybean harvest period, and that the city of Maringá, (4) Brazil, another area of bumper soy production and processing, has soybean allergens in the air, all year round.

So, if you suddenly find yourself wheezing on a windy day, and you’re not sure why, find out whether there is any soybean production or processing going on in your area and, if there is, consider a trip to the allergist.

You can develop an allergy to soy after a organ transplant .

Although it can happen after different types of transplant, the liver is by far the organ most likely to provoke a new food allergy, and soy is the 4^th most common food allergy you’re likely to get.

Children are much more likely to develop an allergy after a liver transplant than adults; in fact, there are case reports describing children and adults getting a liver from the same donor and the child developing a food allergy while the adult does not. Why this is so is not known, although it may be something to do with the fact that the child needs a liver transplant in the first place, or it may be because their immune system is too immature to be able to suppress the expression of new acquired food allergies.

Or it could be something about the liver itself, as there are cases that also describe people getting different organs from the same donor and only the person getting the liver developing a new food allergy. The liver contains blood stem cells that can develop into all types of blood cells, including immune system cells that would be sensitised to the allergen now causing the problems.

There are several risk factors associated with acquiring a food allergy after a liver transplantation, including being under the age of 1 at the time of the transplant, having eczema at the time of the transplant, having a family history of atopy (including asthma or hay fever) and/or having the Epstein-Barr virus in your bloodstream.

Symptoms generally manifest themselves within 18 months of the operation (often, under a year, but a time period of 17.6 years has also been reported) and often affect the skin; facial swelling (angio-oedema) and hives (urticaria) occur in around 4 to 5 in 10 people. Gastrointestinal manifestations (diarrhoea, vomiting, stomach pain) also affect around half of people with LTFA, while respiratory symptoms affect about 1 in 10. Anaphylaxis affects around 16 in 100 children with LTFA, but no fatal cases have been reported so far. Although the majority (around 4 in 5) of the new food allergies are IgE-mediated, some people also get mixed or non-IgE-mediated allergies like Eosinophilic Gastrointestinal Disorders (EGIDs).

Unfortunately, once a child develops a food allergy after an organ transplant, they’re likely to keep it, although their symptoms may improve over time and, if they have developed a multiple food allergy, they may outgrow their allergy to at least some of the foods. Ultimately, around 1 in 4 may end up with an unrestricted diet.

This does not seem to be the case for adults, who seem more likely to lose the allergy after a few months. However, Due to the small number of reported cases of food allergy transfer after solid organ transplant in adults, it’s difficult to say whether the allergy is more likely to be temporary or permanent.

Although less common, there are also reports of people developing new allergies to soy after bone marrow transplants and after cord blood transplants (when blood stem cells are collected from umbilical cord blood and given to someone who needs a stem cell or bone marrow transplant but doesn’t have a matched donor). The latter type is more likely to resolve.

Having a cat as a pet may decrease the risk of your child developing soy allergy.

This is according to a Japanese study that analysed the data of 66,215 children and found that being exposed to cats during early infancy reduced a child’s risk of developing soy allergy, at least during the first 3 years of life.

It should be pointed out that the diagnosis of food allergies was based entirely on parent-reported doctors’ diagnoses rather than medical test results.

And their results do not agree with those of a British study which reported a link between having dogs and having a lower risk of getting food allergy but found no link between having a cat and having a lower risk of food allergy.

Or, worse yet, those of a Polish study that noticed that having a dog during pregnancy was associated with a lower risk of having a child who developed food allergies, but that having a cat, hamster, guinea pig, rabbit before and during pregnancy actually increased the risk of having a child who developed food allergies.

These differences were put down to differences in study size (the previous ones were much smaller) and method which may have masked the effect of having a cat. The fact is that research does tend to suggest a link between having pets—notably dogs—and having a lower risk of developing food allergies. It certainly doesn’t seem to make things worse so, if you are on the fence about getting a furry friend, just do it.

You can reintroduce your child to soy using a soy ladder.

Food ladders for allergenic foods are based on what the latest science has to say about how to make a food less likely to provoke a reaction—in the case of soy, the idea that fermentation can reduce allerginicity by up to 99%, and that baking can also make soy less allergenic.

The soy ladder exists to help parents to reintroduce soy to young children who are growing out of a delayed, non-IgE mediated allergy to soy, such as eczema, and should only be used after a healthcare professional has okayed its use.

You can find a simple explanation of the soy ladder here, a short guide to the soy ladder here and a slightly more wordy guide to introducing soy to your children at home including some baked soybean recipes here.

People who prefer to get their info in video form can watch one here.

Feeding your cow milk-allergic infant a soy-based formula is very unlikely to lead to them developing an allergy to soy.

Research that is now over two decades old indicated that up to 14% of milk-allergic infants might develop an allergy to soy when fed soy-based formula.

As a result, current guidelines from the World Allergy Organization Diagnosis and Rationale for Action against Cow’s Milk Allergy (DRACMA), the European Society for Paediatric Gastroenterology Hepatology and Nutrition (ESPGHAN), the European Academy of Allergy and Clinical Immunology (EAACI), the British Society for Allergy & Clinical Immunology (BSACI), and the Royal Australasian College of Physicians all recommend avoiding feeding soy-based formula to infants under 6 months old who are allergic to milk (and ESPGHAN recommends testing for soy allergy first).

The Canadians recommend finding another solution altogether.

All of the above organisations recommend breastfeeding for at least 6 months, if possible, and longer if desired, and all recommend using an extensively hydrolysed cow’s milk formula (in which milk proteins have been fragmented to make them less allergenic), or an amino acid-based formula (made from individual amino acids and devoid of food proteins) as a first choice for the treatment of infants with an allergy to cow’s milk.

But the advice to avoid soy is not universal.

The American Academy of Pediatrics (AAP) considers soy-based formula safe to use as soon as the baby is born, as long as the pregnancy was full term.

Soy-based formula is also recommended for infants who have galactosemia, lactose intolerance, or who are not able consume dairy-based products for cultural or religious reasons.

And the advice to avoid soy formula in general may be overcautious.

A 2014 review of 21 studies including children with milk allergy or eczema calculated a prevalence of sensitisation to soy of 8.7% (based on skin prick testing) or 8.8% (based on blood testing) in infants after using a soy-based formula, and a prevalence of challenge-proven allergy of just 2.5 %. An examination of 11 studies in which soy-sensitised children either ate soy on a regular basis or successfully passed a food challenge with soy determined that only 11.2 % of these soy-sensitized children actually had a clinical allergy (i.e. showed symptoms).

Finally, the researchers looked at 3 studies that included 1,430 infants under 6 months old that evaluated for the risk soy allergy in infants younger than 6 months and found that only two of them probably had soy allergy, leading them to state:

‘Not enough evidence exists to show a higher risk of allergy in infants younger than 6 months. The concern about soy allergy is no reason to postpone the use of SBFs in IgE-mediated cow’s milk allergy infants until the age of 6 months.’

Other research-based recommendations and highlights on soy-based formula include:

• Soy-based formulas are as nutritious as cow milk formulas. Whereas formulas used to be made using soy flour protein, over the last few decades they have been changed to improve digestibility and protein quality and are supplemented with essential amino acids like methionine, taurine, and carnitine, long chain polyunsaturated fatty acids and minerals like calcium, zinc and phosphorus. Multiple studies have documented the normal growth and development in infants fed using soy-based formula and reported normal physiological parameters such as average energy intake, blood albumin concentration, bone mineralisation and micronutrients status
• Soy-based formulas are not the same as soy beverages like soy ‘milk’ which are marketed to older children and adults and should not be used as a substitute for formula for children under 2 years old. Soy drinks are not nutritionally comparable with milk, even if they have the word ’milk’ on the packet (for this reason, European Council Regulation 1234/2007 prohibits the use of the word ‘milk’ for drinks that are not made from a ‘normal mammary secretion’). Feeding infants exclusively with plant-based drinks can lead to severe nutritional deficiencies
• Soy-based formula is not recommended for preterm infants. This is because soy formulas have a higher aluminium content than cow’s milk-based formula and breast milk, and aluminium competes with calcium absorption, which could increase the risk of osteopenia (reduced protein and mineral content of bone tissue). Modern soy-based formulas are now supplemented with phosphorus and calcium, and the aluminium content is maintained at a very low level, but the long-term effects of early aluminium exposure have not yet been fully clarified
• Soy-based formulas are not suitable for infants diagnosed with cow milk protein-induced enteropathy or enterocolitis, because of the high frequency of sensitivity to both milk and soy antigens in these infants
• There is little evidence that isoflavones present in soy-based formula have negative hormonal effects in children, although this is still under debate because it’s difficult to assess whether soy food consumption early in life (<24 months) is completely safe. To date, however, no long term effects on sexual development, thyroid disease, immune function, or neurodevelopment have been demonstrated in infants fed soy-based formula
• Using a soy formula instead of a cow’s milk formula does not reduce the risk of getting allergies
• Soy-based formula does not increase the risk of getting peanut allergy

Scientists are hard at work on a cure for soy allergy in the form of immunotherapy.

Up to now, however, the few efforts that have been reported—notably for people with pollen-related food allergy—have not produced promising results.

However, there was a more encouraging study published by Japanese researchers in 2021. They used subcutaneous immunotherapy (allergy shots) with birch pollen extract to treat a small group of children aged between 6 and 10 who were allergic to birch pollen and soy and whose allergies required them to stick to a restricted school lunch program.

The children underwent a rush treatment (which produced a few easily treatable systemic side-effects) and, when they were checked again one year after the start of their treatment, all of the 6 children could drink at least 100 ml of soy milk and eat some sprouts and were free of dietary restrictions school. Those who had been allergic to apple, too, were also able to eat at least some apple without symptoms.

Currently, immunotherapy for soy allergy is not available outside research labs.

Image by 360floralflaves on Unsplash