EAACI guideline: Preventing the development of food allergy in infants and young children (2020 update)

This guideline from the European Academy of Allergy and Clinical Immunology (EAACI) recommends approaches to prevent the development of immediate‐onset / IgE‐mediated food allergy in infants and young children. It is an update of a 2014 EAACI guideline.


| INTRODUC TI ON
Allergic reactions to foods such as hen's egg, cow's milk and peanut can impair an individual's health and quality of life and have substantial healthcare costs. 1,2 The prevalence is high, for example, in highincome countries, and up to one in ten people live with a food allergy, with the highest prevalence amongst infants and young children. 1 In 2014, the European Academy of Allergy and Clinical Immunology (EAACI) released a guideline to help countries, clinicians and families prevent food allergy. 3 The guideline has now been updated to include the latest research.
This guideline provides evidence-based recommendations about approaches for preventing the development of IgE-mediated / immediate-onset food allergy (hereafter 'food allergy'). This is defined as a reproducible adverse reaction to food mediated by an immunologic mechanism. The guideline does not focus on preventing conditions that may be associated with food allergy such as food sensitization, eczema or non-IgE-mediated conditions. Some populations are at greater risk of developing food allergy than others, including those with atopic heredity, eczema or IgE sensitization.
The guideline examines interventions for those at increased risk and those at general risk of food allergy. Table 1  Young children with symptoms suggestive of food allergy should be referred to a specialist centre for assessment and further management.

| Stakeholder involvement
The guideline was developed by an EAACI Task Force with representatives from 11 countries. Participants included a variety of disciplinary and clinical backgrounds, including allergists (specialist and subspecialists), paediatricians, primary care, patient representatives, immunologists, dieticians, statisticians and researchers. Methodologists led a systematic review of evidence, and clinical academics formulated recommendations for clinical care. Product manufacturers and other stakeholders had an opportunity to comment as part of a public consultation process at the final stage.

| Guideline approach
The Task  EAACI intends to update this guideline in 2026 unless there are significant advances before then.

| Collating evidence of effectiveness
Clinical questions were generated and prioritized by the Task Force.

The key clinical question was 'what is the effectiveness and safety of interventions to prevent the development of IgE-mediated / immediateonset food allergy in infants, children and adults?'
The Task Force worked with independent researchers to undertake a systematic review of research evidence (PROSPERO registration CRD42019127457). The methodology has been published 6 so far is only briefly described here. The reviewers searched 11 bibliographic databases, the reference lists of identified studies and 35 systematic reviews, and contacted experts in the field for trials published between 1946 and 31 October 2019.
Our systematic review 4 included 46 studies: 41 randomized controlled trials (hereafter 'trials') and, in the case of breastfeeding only, five prospective cohort studies with at least 1000 participants at general risk of food allergy or at least 200 participants at increased risk of food allergy. Studies involving infants, children and adults were eligible, but only studies about preventing food allergy during infancy and early childhood were identified.
As per the GRADE approach, 7 study findings were extracted and compiled using evidence profiles and summary of findings tables as published in the systematic review online materials. 4

| Formulating and reviewing recommendations
The Task Force met regularly in person and virtually over an 18month period. For each intervention, the Task Force considered the effect on food allergy outcomes across all studies, not solely individual studies. The populations and interventions were too heterogeneous to allow meta-analysis. Only intention-to-treat analyses were considered.
In addition, the applicability and generalizability of results, consistency of study findings and risk of bias were considered for each topic area. From this information, the Task Force made a judgement about the overall certainty of evidence regarding an intervention. The Task Force also drew on patient organizations and expert opinion to consider the balance of benefits versus harms, preferences and values, and resource implications and feasibility. 8 These factors were drawn together to formulate evidence-based recommendations for clinical care.
In line with the GRADE approach, the Task Force used specific wording to denote whether a recommendation was 'for' or 'against' an intervention (direction) and whether a recommendation was strong or conditional (strength). Table 2 sets out the wording conventions used. This guideline uses the wording 'The EAACI Task Force suggests' to denote a conditional recommendation. A conditional recommendation is still a recommendation for or against a particular intervention, it simply means that there may not be sufficient evidence about effectiveness or harms to conclude that this is the best approach in every case or that it should be universally implemented in all policy and practice.

Certainty of evidence
The extent to which the evidence can be relied upon, rated as very low, low, moderate and high. The GRADE approach was used to decide on the certainty of evidence, including risk of bias, directness, consistency and precision of the estimates.
Complementary feeding WHO defines complementary feeding as the process starting when breastmilk alone is no longer sufficient to meet the nutritional requirements of infants. 9 In recognition that not all infants may be fed breastmilk and in line with ESPGHAN 75 and EFSA, 10 this guideline defined complementary feeding as the process of introducing foods and liquids alongside with breastfeeding (or infant formula if applicable) when breastmilk (or infant formula) no longer meets the nutritional requirements of infants. WHO advises that complementary feeding should start from 6 mo of age, 9 but some choose to start from four to 6 mo of age, often with increasing amounts of foods in a developmentally appropriate consistency, which is also in line with a recent statement from EFSA. 10 Cow's milk protein Cow's milk protein, which may include formats such as yoghurt and cheese Early childhood Up to 5 y old Food allergy Reproducible adverse reaction to food mediated by an immunologic mechanism, involving specific IgE (IgE-mediated), cellmediated (non-IgE-mediated) or both IgE and cell-mediated mechanisms. For the purposes of this guideline, the term 'food allergy' is used as shorthand to mean IgE-mediated/ immediate-onset food allergy.
IgE-mediated / immediateonset food allergy Food allergy shown or suspected to be IgE-mediated, often with onset within hours after exposure. Solely non-IgE-mediated food reaction (eg eosinophilic oesophagitis / gastroenteritis) conditions are not included.
Increased risk Greater risk of food allergy due to having a condition associated with food allergy such as eczema or asthma or having immediate relatives with a history of any allergy, atopic dermatitis, asthma or hay fever.

Prebiotic
Non-digestible substances that provide a beneficial physiologic effect for the host by selectively stimulating the favourable growth or activity of a limited number of indigenous bacteria.

Probiotic
Live microorganisms, which, when administered in adequate amounts, may confer a health benefit on the host.

Regardless of the risk of food allergy
Regardless of whether the target group is at increased risk of food allergy or general / undifferentiated risk Sensitization Detectable specific IgE antibodies, either by means of skin prick test or determination of specific IgE antibody levels in a serum sample Significant Statistically significant, P < .05 Synbiotic A combination of prebiotic(s) and probiotic(s) that may beneficially affect the host by improving the survival and activity of beneficial microorganisms in the gut

Trial
Randomized controlled trial (RCT) The Task Force used specific wording to describe the strength of evidence and effect size supporting the recommendations (see Table 3).
All recommendations were agreed by consensus, except for the timing of the introduction of peanuts where 72% of eligible members who chose to vote were in favour of a conditional recommendation and 28% favoured a strong recommendation.
A draft of this guideline was externally peer-reviewed by invited experts from a range of organizations, countries and professional backgrounds. The draft guideline was also publicly available on the EAACI website for a 3-week consultation period in July 2020 to allow a broader array of stakeholders to comment. The chairs read all the comments and proposed changes as a result. The entire Task Force was then given the opportunity to review comments and approved the proposed edits.

| Editorial independence and managing conflicts of interest
The guideline development process was funded by EAACI. The Evidence about effectiveness was compiled independently by methodologists who had no conflict of interests. The recommendation can be adopted as a policy in many situations. The specific context may mean that it is not relevant, as refined by stakeholders Conditional recommendation against an intervention 'The EAACI Task Force suggests against …' Many people in this situation should not be offered the intervention.

| RECOMMENDATIONS
However for others, different choices will be appropriate. Clinicians could help each patient make decisions consistent with the patient's preferences The recommendation can be adopted as a policy in many situations. The specific context may mean that it is not relevant, as refined by stakeholders Strong recommendation against an intervention 'The EAACI Task Force recommends against …' Most people in this situation should not use this intervention The recommendation can be adopted as a policy in most situations No recommendation 'There is no recommendation for or against using …' Different choices will be appropriate for different people. Clinicians could help each patient make decisions consistent with the patient's preferences Policies may differ depending on context and should be developed with the involvement of a wide range of stakeholders recommendation. It does also recognize that in developed countries, (i) sometimes there is a need for a breastmilk substitute, and in these cases, an infant formula is recommended; (ii) some families choose to start complementary feeding earlier, between four and 6 months, which is also in agreement with a recent statement from European Food Safety Authority (EFSA) 10 ; and (iii) early complementary feeding need not have a negative impact on breastfeeding. 11,12

| Maternal dietary avoidance
The EAACI Task Force suggests against restricting the consumption of potential food allergens during pregnancy or breastfeeding in order to prevent food allergy in infants and young children (see online supplement Table S2).

Reason for recommendation
Our systematic review identified five trials about this topic in women at increased risk, two of which focused on dietary avoidance alone 13,14 and three combined with another intervention. [15][16][17] The review concluded that avoiding potential food allergens during pregnancy, when breastfeeding or in infancy, alone or combined with other interventions, may have little to no effect on food allergy in early childhood, but the evidence is very uncertain.
The majority of trials found no reduction in the prevalence of food allergy when women avoided dietary allergens such as egg and milk. The harm associated with avoiding foods during pregnancy and breastfeeding may be greater than any potential reduction in food allergy. Food allergens do not exist in isolation so removing food groups may also reduce the intake of vital nutrients and fibre, adversely affecting the health of women and their infants. 18

Strength of recommendation
This guideline is against maternal avoidance of dietary food allergens, but this is not the strongest recommendation possible because the certainty of evidence is very low. There were only a small number of studies, they contained varying interventions, and there was very low certainty about their effect on food allergy.

Practical implications
Professionals should encourage women to not restrict consumption of specific allergenic foods. Rather, they should follow local guidelines, eating a healthy, balanced diet when pregnant and breastfeeding. This applies regardless of the infant's risk of food allergy.

| Introducing hen's egg into the infant diet
The EAACI Task Force suggests introducing well-cooked hen's egg, Very low X may have little to no effect on food allergy, but the evidence is very uncertain Note: Small, medium and large effect sizes were all required to be deemed clinically important in order to be considered as a substantive effect.
part of complementary feeding to prevent egg allergy in infants (see online supplement Table S3).

Reason for recommendation
Our systematic review included two trials about cooked egg 11,12,19 and three about raw or pasteurized egg in general-and increasedrisk infants. 11,12,20,21 An additional subgroup analysis from one of the cooked egg studies has since been published. 22 The evidence suggests that introducing small amounts of cooked, but not raw egg or uncooked pasteurized, hen's egg into the infant diet as part of complementary feeding probably reduces the risk of egg allergy in infancy.
The benefits of introducing well-cooked egg probably outweigh potential harms.
The Task Force does not support early introduction of raw egg or uncooked pasteurized egg because the potential harms may outweigh the benefits. Studies found adverse reactions, including anaphylactic reactions. 11,12,20,21

Strength of recommendation
This guideline supports the introduction of well-cooked egg into the infant diet, but this is not the strongest recommendation possible because the certainty of evidence is moderate. There were only a small number of studies about cooked egg, their results were inconsistent, and there was only moderate to low certainty about the effect on egg allergy.
Evidence about raw egg or uncooked pasteurized egg was of low certainty. The available trials had inconsistent findings.

Practical implications
Healthcare professionals in countries where egg allergy is an issue could encourage families with infants at general and increased risk to start introducing about half of a well-cooked, small egg twice a week as part of complementary feeding from four to 6 months of age. This is in agreement with the recent European Food Safety Authority statement. 10 This amount of egg is based on a trial that found that eating at least 2 grams of egg white protein per week prevented egg allergy. 11,12 One other trial successfully prevented egg allergy with TA B L E 4 Recommendations for preventing the development of food allergy

Recommendations supporting interventions
The EAACI Task Force suggests introducing well-cooked hen's egg, but not raw egg or uncooked pasteurized egg, into the infant diet as part of complementary feeding to prevent egg allergy in infants.

Moderate
In populations where there is a high prevalence of peanut allergy, the EAACI Task Force suggests introducing peanuts into the infant diet in an age-appropriate form as part of complementary feeding in order to prevent peanut allergy in infants and young children.

Moderate
The EAACI Task Force suggests avoiding supplementing with cow's milk formula in breastfed infants in the first week of life to prevent cow's milk allergy in infants and young children Low

Recommendations against interventions
The EAACI Task Force suggests against restricting consumption of potential food allergens during pregnancy or breastfeeding in order to prevent food allergy in infants and young children.

Very low
The EAACI Task Force suggests against introducing soy protein-based formula in the first 6 mo of life to prevent cow's milk allergy in infants and young children.

Very low
The EAACI Task Force suggests against using bacillus Calmette-Guérin (BCG) vaccination to prevent food allergy in infants and young children.

Low No recommendation made
There is no recommendation for or against using breastfeeding to prevent food allergy in infants and young children, but breastfeeding has many benefits for infants and mothers and should be encouraged wherever possible.

Very low
For infants who need a breastmilk substitute, there is no recommendation for or against the use of regular cow's milk-based infant formula after the first week of life to prevent food allergy.

Low
There is no recommendation for or against using partially or extensively hydrolysed formula to prevent food allergy in infants and young children. When exclusive breastfeeding is not possible, many substitutes are available for families to choose from, including hydrolysed formulas.

Low
There is no recommendation for or against vitamin supplementation or fish oil supplementation in healthy pregnant and/or breastfeeding women and/or infants to prevent food allergy in infants and young children.

Very low
There is no recommendation for or against prebiotics, probiotics or synbiotics for pregnant and/or breastfeeding women and/ or infants alone or in combination with other approaches to prevent food allergy in infants and young children.

Low
There is no recommendation for or against using emollients as skin barriers to prevent food allergy in infants and young children.

Low
There is no recommendation for or against using preventive oral immunotherapy to prevent food allergy in infants and young children.
Low smaller amounts [PETIT study]. The trials utilized hard-boiled egg (10-15 minutes), but we would consider that equivalent amounts of egg in well-baked foods would also be appropriate. 23

| Introducing peanuts into the infant diet
In populations with a high prevalence of peanut allergy, the EAACI Task Force suggests introducing peanuts in an age-appropriate form as part of complementary feeding in order to prevent peanut allergy in infants and young children (see online supplement Table S4).

Reason for recommendation
Our systematic review included three trials about this, one in general-risk and two in increased-risk infants. 11,12,24 The review There was some inconsistency in the results. One study introduced peanut along with five other foods. Two studies focused on infants at very high risk and compared with complete abstinence from peanut for 5 years rather than more usual exposure. All of the studies took place in the UK. Therefore, the generalizability of the findings is uncertain and this led to a conditional recommendation.

Practical implications
In counties where peanut allergy is prevalent, healthcare professionals could encourage families to introduce peanuts as part of complementary feeding. Professionals should advocate introducing peanut in an age-appropriate form alongside continued breastfeed- ing. It appears that the most effective age to introduce is from four to 6 months of life. The evidence of benefit relates mainly to those at very increased risk, but this could be encouraged in those at general risk as well because many cases of peanut allergy are seen in this lower risk group. 28 Peanut should be introduced in an age-appropriate form to avoid any risk of choking or inhalation. For example, infants could be given one heaped teaspoons of diluted peanut butter (2g peanut protein) each week. 11,12 We suggest that peanut should not be the first solid to be introduced into the infant diet. The EAACI Task Force makes no recommendation for countries with a low prevalence of peanut allergy. In these countries, peanuts should be included in the diet according to normal eating habits and local recommendations.

| Breastfeeding
There is no recommendation for or against using breastfeeding to prevent food allergy, but breastfeeding has many benefits for infants and mothers and should be encouraged wherever possible (see online supplement Table S5).

Reason for recommendation
Our systematic review included seven studies about breastfeeding. [29][30][31][32][33] The review concluded that breastfeeding has many benefits for infants and mothers, but it may not reduce the risk of food allergy or cow's milk allergy.
The evidence was of low certainty because it is based on observational studies as it is difficult ethically to undertake randomized trials of breastfeeding.
Breastfeeding meets all of the nutritional needs of infants up to 6 months of age and is recommended by WHO. 34 Breastfeeding may also reduce societal costs associated with ill health. 35 Therefore, the balance of benefits and harms is in favour of breastfeeding, even though there is insufficient evidence about benefits related to preventing food allergy.

Practical implications
While breastfeeding may not prevent food allergy, professionals should support breastfeeding given its other positive benefits. 36 Professionals need to also sensitively support families that do not breastfeed their infants.

| Supplementation with cow's milk formula in the first week of life
The EAACI Task Force suggests avoiding supplementing with cow's milk formula in breastfed infants in the first week of life to prevent cow's milk allergy in infants and young children (see online supplement Table S6).

Reason for recommendation
Our systematic review included one trial about this. 37 The review found that avoiding supplementation with regular cow's milk formula in breastfed infants during the first three days of life may result in a large decrease in the risk of cow's milk allergy in early childhood.
The World Health Organization (WHO) also warns that any supplementation may be associated with a reduction in breastfeeding 34 and most healthy, mature infants do not need any supplementation to breastfeeding.

Strength of recommendation
This guideline supports avoiding supplementation with cow's milk formula in the first week of life amongst breastfed infants, but this is not the strongest recommendation possible because the evidence is of low certainty. There was only one trial available, and it contained multiple interventions, making it difficult to apply the findings to practice. However, the trend is supported by other studies not eligible for the review, which also found increased incidence of cow's milk allergy when cow's milk formula was used a temporary feed in the first week of life. 29,30 Practical implications Healthcare professionals and families could avoid supplementation with cow's milk formula in breastfed infants in the first week of life. It is important to support breastfeeding, and breastfeeding is usually sufficient with no need for supplementation in healthy, term-born infants.
If needed, the family should seek advice from healthcare professionals.
Other possible temporary supplementary options might include, for example, donor breastmilk, hydrolysed formula, amino acid formula or water, depending on clinical, cultural and economic factors.

| Regular consumption of cow's milk formula
For infants who need a breastmilk substitute, there is no recommendation for or against the use of regular cow's milk-based infant formula after the first week of life to prevent food allergy (see online supplement Table S7).

Reason for recommendation
Our systematic review included seven trials about this. [38][39][40][41][42][43][44] The review concluded that introducing conventional cow's milk-based formula after the first week of life did not have a consistent impact on the development of cow's milk allergy in infancy or early childhood.
There do not appear to be significant harms associated with regular consumption of cow's milk-based formula for either general-risk or increasedrisk infants after 3 months of age although WHO has warned that any supplementation may be associated with a reduction in breastfeeding. 34

Strength of recommendation
No recommendation could be made as the evidence is of low to very low certainty. The studies investigated different interventions, duration and comparators.

Practical implications
Breastfeeding is natural and beneficial and should be the preferred approach where possible. Where a breastmilk substitute is required, cow's milk-based formulas are preferred to standard cow's milk during the first year of life, due to nutritional value and ease of digestion.

| Hydrolysed infant formula
There is no recommendation for or against using partially or extensively hydrolysed formula to prevent cow's milk allergy in infants.
When exclusive breastfeeding is not possible, many substitutes are available for families to choose from, including hydrolysed formulas (see online supplement Table S8).

Reason for recommendation
Our systematic review included nine trials about this. 16,17,29,30,[38][39][40]45,46 The review found that partially or extensively hydrolysed whey or casein formula may not reduce the risk of cow's milk allergy compared with conventional cow's milk formula.
There is no consistent evidence that hydrolysed formula reduces the risk of food allergy nor is there consistent evidence that hydrolysed formula causes harm. There was little to no evidence that one type of hydrolysed formula was more effective than others.
The evidence here is of low certainty. Trials used different formulas, introduced them at different times, often had small samples and often did not use robust diagnostic criteria for food allergy.

Practical implications
Breastfeeding of all infants is preferable, but when a breastmilk substitute is needed, professionals could help families consider the best possible alternative for a family's individual circumstances. The options discussed could include a hydrolysed infant formula.

| Soy protein formula
The EAACI Task Force suggests against introducing soy protein formula in the first 6 months of life to prevent cow's milk allergy in infants and young children (see online supplement Table S9).

Reason for recommendation
Our systematic review included one trial about this. 40 The review concluded that soy-based formula may have little to no effect on cow's milk allergy in early childhood, but the evidence is very uncertain.
There may be more potential harms than benefits from using soy

Strength of recommendation
This guideline is against the use of soy protein formula in the first 6 months of life, but this not the strongest recommendation possible because the evidence is of very low certainty. There was one pertinent trial, which did not use a robust definition of food allergy.

Practical implications
Soy protein formula is unlikely to protect against cow's milk allergy. It may be considered for infants who cannot have dairy-based products because of cultural, medical, religious or family reasons such as a vegan lifestyle, persistent lactose intolerance or galactosaemia. 48 Professionals should discuss the benefits and potential harms fully with families.

| Vitamin and fish oil supplements
There is no recommendation for or against vitamin supplementation or fish oil supplementation in healthy pregnant and/or breastfeeding women and/or infants to prevent food allergy in infants and young children (see online supplement Table S10).

Reason for no recommendation
Our systematic review included eight trials about vitamin or fish oil supplements in general-or increased-risk populations. [49][50][51][52][53][54][55][56] The review found that vitamin supplements for pregnant and/or breastfeeding women or infants may have little to no effect on food allergy in early childhood, but the evidence is very uncertain.
The review found that fish oil supplements during pregnancy, when breastfeeding or in infancy, may not reduce food allergy in infancy or early childhood. However, when taken during pregnancy and continued during breastfeeding, fish oil may reduce food allergy slightly in young children at increased risk.
There was no consistent evidence that these supplements cause harm in healthy women and infants.
The evidence about vitamin supplements was of very low certainty, and the evidence about fish oil was of low certainty. Studies used different supplements, doses, timelines, target groups and combinations of interventions.

Practical implications
Women who are not getting the recommended daily allowances of vitamins, minerals and omega-3 through their diet may benefit from supplementation for health reasons, according to national guidance, but not for the purposes of preventing food allergy in infants.

| Prebiotics, probiotics and synbiotics
There is no recommendation for or against prebiotics, probiotics or synbiotics for pregnant and/or breastfeeding women and/or infants alone or in combination with other approaches to prevent food allergy in infants and young children (see online supplement Table S11).

Reason for no recommendation
Our systematic review included nine trials about this. [57][58][59][60][61][62][63][64][65] The review found that prebiotics, probiotics and synbiotics for mothers and infants may have little to no effect on food allergy in infancy and early childhood, but the evidence is very uncertain. There is no evidence that they cause harm in healthy women and infants.
The evidence here was of very low certainty. The studies differed in size, duration of supplementation, type of supplementation, timing of supplementation, diagnostic criteria and duration of follow-up. The clinical effects and safety of any single probiotic or combination of probiotics, prebiotics or synbiotics cannot be extrapolated to other probiotics as they are immunologically distinct. This makes it difficult to provide a clear recommendation.

Practical implications
Professionals should help families consider the pros and cons of different prebiotics, probiotics or synbiotics, being clear that they may have little to no effect on the development of food allergy. Where professionals decide to use these in premature infants, caution is advised. Professionals should consider advising against their use where immunosuppression may be possible.

| BCG vaccination
The EAACI Task Force suggests against using bacillus Calmette-Guérin (BCG) vaccination to prevent food allergy in infants and young children. Our systematic review included two studies about this in general-risk infants. 56,66 The review concluded BCG vaccination may have little to no effect on food allergy in infancy and early childhood, but the evidence is very uncertain (see online supplement Table S12). This recommendation is based on low certainty evidence, with some harms noted for immunodeficient infants. 67 BCG is part of the immunization schedule in many countries where tuberculosis prevalence is high. Families should be encouraged to follow the immunization programmes for their country. Our recommendation is against using the vaccination for preventing food allergy.

| Emollients
There is no recommendation for or against using emollients as skin barriers to prevent food allergy in infants. Our systematic review included one trial about this. 65 It concluded that emollients may have little to no effect on food allergy in infancy and early childhood, but the evidence is very uncertain (see online supplement Table S13). A further trial, published after our systematic review, found that emollients did not reduce food allergy at 2 years in high-risk infants. 68 Different emollients may have different effects. 69

| Preventive house dust mite oral immunotherapy
There is no recommendation for or against using oral immunotherapy to prevent food allergy in infants. Our systematic review

Facilitators to implementation
Audit criteria

Resource implications
Introducing cooked egg into the infant diet Infants may not enjoy the texture or taste until later in infancy.
Knowledge about the preventive impact amongst parents and healthcare professionals.
Proportion of infants who are eating 2 grams or more of egg white protein by 6 mo of life.
Minimal, egg is readily available in most countries.
Introducing peanut into the infant diet Families with known peanut allergy may want to refrain from introducing peanut in the home due to risk of allergic reactions in other family members. Parents may not wish to feed their infants quantities of monosaturated fat. It may be difficult to feed infants peanut butter or peanut snacks.
Knowledge about the preventive impact amongst parents and healthcare professionals.
Proportion of infants who are eating 2 grams or more of peanut protein a week by 6 mo of life in countries with a high prevalence of peanut allergy.
Minimal, peanuts are a lowcost source of protein.
Avoiding supplementation with cow's milk formula in the first week of life Standard cow's milk formula has historically been used in some countries as a supplementation in the first week of life, often without reason and without parent knowledge.
Breastfeeding is usually sufficient in healthy Minimal as alternatives such as water, glucose water, breastmilk, hydrolysed formula or amino acid formula are relatively inexpensive in the quantities required.
Avoiding introducing raw (pasteurized) egg into the infant diet None, except for certain lifestyles, where this is a normal part of infant feeding.
Knowledge about potential harms amongst parents and healthcare professionals.
Proportion of healthy infants who are not given raw egg or uncooked pasteurized egg the first 6 mo.
No additional resources are needed.

Pregnant and breastfeeding women
following a usual diet, not avoiding potential food allergens None, except for those following a vegan lifestyle.
Knowledge about potential harms amongst parents and healthcare professionals.
Proportion of healthy pregnant / breastfeeding women on a normal healthy diet.
No additional resources are needed.
Avoiding the use of soy protein formula for food allergy prevention Families with lactose intolerance or following a vegan lifestyle may find it difficult to find alternatives.
Knowledge about potential harms and alternatives amongst parents and healthcare professionals.
Proportion of healthy infants who are not given soy protein formula.
None as alternatives are similar in cost.
Avoiding BCG vaccination for food allergy prevention BCG vaccination programmes are important for TB prevention and there are established programmes in some countries. The vaccination has benefits, but not for the prevention of food allergy.
Knowledge about potential harms amongst parents and healthcare professionals.

None.
No additional resources are needed.
included one trial of preventive house dust mite oral immunotherapy in increased-risk infants. 70,71 The review concluded that oral immunotherapy may have little to no effect on the development of food allergy in infancy and early childhood, but the evidence is very uncertain (see online supplement Table S14).

| Implications
This guideline will help healthcare professionals consider evidence demonstrated to be safe. [10][11][12]26 Where a breastmilk substitute is used, the best alternative should be recommended, such as an infant formula adapted to the nutritional needs of infants. Some families may prefer a hydrolysed formula, though there is no clear evidence that this prevents food allergy. We suggest avoiding supplementation of cow's milk-based formula in breastfed infants during the first week of life as this may be associated with increased food allergy, but there is no evidence that avoiding regular use of cow's milk-based formula after the first week prevents food allergy.
Recent research has suggested that introducing cooked hen's egg and peanut in an age-appropriate form as part of complementary feeding may prevent the development of allergy to egg and peanut, the latter in countries with a high prevalence of peanut allergy.
This is similar to recommendations in other current guidelines. 72,73 There may be a number of facilitators and barriers that may affect the implementation of these recommendations, including historical and cultural habits and socioeconomic systems ( Table 5).
Education of professionals and families is paramount.

| Strengths and limitations
A strength of this guideline is that it is informed by a balance of evidence and expert opinion. A comprehensive systematic review was undertaken evaluating the evidence according to well-established GRADE methods. We focused on randomized controlled trials to provide the highest quality available evidence. This differs from previous systematic reviews, which have synthesized the results of different studies. 4 The review was led by independent methodologists with no conflicts of interest.
It is a strength that the recommendations were not only based on the best available evidence, but also expert clinical and patient

| Research gaps
There is much left to learn about preventing food allergy. Table 6 sets out key priorities. Some of these require new high-quality stud-

| Conclusion
This guideline supports breastfeeding and provides simple recom- Thomas Werfel and Margitta Worm for their support with the guideline; Ana Antunes and Anna Gandaglia for their assistance; and EAACI for funding the generation of the guideline.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1111/pai.13496.