select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to research on giving peanut products to babies

Research published in The Journal of Allergy and Clinical Immunology looks at the window of opportunity to prevent peanut allergies.

This Roundup accompanied an SMC Briefing.


Prof Jean Golding, Emeritus Professor of Paediatric and Perinatal Epidemiology, University of Bristol, said:

“This is a very useful summary of a number of well-conducted randomised controlled trials which definitively showed that by introducing a young baby to peanut products at age 4-6 months, a substantial number of cases of peanut allergy would be prevented. They use the existing information to compare two strategies: (i) to just target the infants at high risk because of severe eczema, or (ii) to target all infants. Their calculations show that targeting all infants will prevent far more peanut allergies than just concentrating on the high risk group. This is an important public health message.”


Prof Mary Fewtrell, Professor of Paediatric Nutrition at the UCL GOS Institute of Child Health, said:

“This paper uses data from 2 large, high quality food allergy prevention trials conducted in the UK (LEAP and EAT) together with observational data from the LEAP screening cohort (Peanut Allergy Sensitisation (PAS)). A number of factors related to the infants enrolled in these trials (high risk of allergy in LEAP and general population in EAT), the timing of introduction of peanut (from 3-4 months in EAT, between 4 and 11 months in LEAP) and the fact that compliance with peanut introduction was lower in the EAT trial, meant that the primary trial analyses were not able to answer the specific question of whether introducing peanut before 6 months (from 3-4 months) was more effective at preventing peanut allergy than introducing it between 6-11 months.

“After reviewing the data from the LEAP/EAT trials together with other trials on this topic, some expert groups (including SACN in the UK and EFSA) concluded that peanuts (along with other food allergens) should be treated like other complementary foods, which are generally recommended from around 6 months of age, although they should be introduced in the first year and not delayed as in the past.  However, not all expert groups agreed with this approach and some concluded that there should be a more pro-active introduction of peanut from around 4 months in high-risk infants, although in practice the definition of high risk may be problematic. For example, the British Society for Allergy & Clinical Immunology (BSACI) have produced guidance suggesting that ‘at risk’ infants could receive peanut and egg from 4 months1.  Of course, it is important to highlight that all of this is only relevant in populations where peanut allergy is a public health concern.

“In the current manuscript, the authors have combined the EAT, LEAP and PAS data, and used this combined dataset to model the effect of introducing peanut at different ages from 4 months on the estimated risk of peanut allergy, taking into account whether the infants were at high risk or not based mainly on the presence of eczema.  They conclude that the preventive effect of peanut introduction decreases with age over the first year, and their model suggests that a 77% reduction in peanut allergy would be expected if peanut was introduced to all infants with eczema from 4 months and to those without from 6 months.  Targeting only high-risk infants in a population would be much less effective as a public health strategy to reduce the burden of peanut allergy.  

“It’s important to note that current NHS advice – ‘You can give your baby nuts and peanuts from around 6 months old’ – is based on the SACN 2018 review of the evidence as it was when they wrote their report on feeding in the first year of life, and many other groups agreed with this approach as they did not think there was sufficient evidence to suggest earlier introduction of peanut for at risk infants. The current paper adds useful information which should certainly be considered when the evidence on this topic is next reviewed.  It will be up to individual bodies to decide if the data presented in this paper should prompt an early review of the evidence on this topic. The main limitations of the results, which are covered in the manuscript, are that it uses a modelling approach which involves making a number of assumptions.  These are described and justified in detail in the supplementary materials. When these findings are reviewed by expert groups, I’m sure they will look carefully at the models and assumptions.

“The message from the paper is that, in settings where peanut allergy is a public health concern, it would be a good idea for peanuts to be introduced in an age-appropriate form between 4 and 6 months of life – the authors specifically suggest 4 months for infants with eczema and 6 months for those without.   This may well be a sensible approach, but infant feeding recommendations are not made considering only one outcome such as food allergy. The risks and benefits of any proposal for a range of outcomes has to be considered, and that’s what the expert groups will do when deciding whether to modify the existing advice.  For example, would introducing peanut before 6 months have any risks for other outcomes?   Would advising parents to introduce peanut from 4 months lead them to also start other foods early which might have negative effects on breastfeeding? The other thing that would need to be thought through is the practical implications of any change to the advice. For example, would recommendations on ‘early introduction’ be confined to peanuts, so parents would be advised to offer small amounts of peanut from 4 months in infants with eczema, but not to introduce other foods?  As the authors point out, education is going to be important here both for health professionals and parents so that any change in the advice is effective and safe.”

“Another important practical issue, which is mentioned in the paper, is that whole nuts should never be given to infants as they are a choking hazard – so peanuts (as with any complementary food) must be given in a safe and age-appropriate form, for example as peanut butter.



Prof Alastair Sutcliffe, Professor of General Paediatrics at UCL, said:

“This interesting paper describes a computer generated modelling study by which it is determined that if peanuts were introduced into the human diet of infants at  4 month, then 6 months then serially at later and later monthly ages, what is the risk of the development of peanut allergy alleged to affect 1-2% of European infants. The studies which inform the model were world class studies which have re-orientated thinking regarding this tiresome condition of children.

“So I for one would be highly supportive of the idea that early introduction of peanuts will reduce substantially the risk of peanut allergy in infants and children.

“This does raise some other thoughts. To the point of dogma, parents are advised NOT to introduce solids until 6 months of exclusive breast feeding – despite this advice NOT being underpinned by secure science. So if parents are now told, well, actually you can introduce solids to your baby from aged 4 months, this can lead to a mixed message. (I have generally questioned the advice of 6 months for solid foods – as historically WHO said 4 months for solids then revised its guidance.)

“In my view this brings out a call to re-consider the 6 month ‘dogma’ and I believe this paper, with its elegant modelling, suggests that a genuine change in practice is needed so human infants can have solids from a younger age where needed.

“Important caveat: no processed foods, and the kind of foods which are part of the national tradition of those people. A look at what Turkish children eat (one of the world’s largest producers of nuts) may be enlightening.”



‘Defining the window of opportunity and target populations to prevent peanut allergy’ by Graham Roberts et al. is published in the Journal of Allergy and Clinical Immunology.




Declared interests

Prof Golding: “I worked with the senior author (Prof Gideon Lack) 20 years ago on the epidemiology of peanut allergy.”

Prof Sutcliffe: No conflict declared.

Prof Fewtrell:

  • Unpaid roles: RCPCH Assistant Officer for Nutrition; ESPGHAN General Secretary (and previously Chair of Nutrition Committee when we write an opinion on complementary feeding); EFSA infant nutrition expert group (involved in the Scientific Opinion on the appropriate age-range for introducing CFs); Used to be on the trial steering committee for the EAT study
  • I also receive an unrestricted donation from Philips for infant nutrition research – not related to the topic of the paper.



in this section

filter RoundUps by year

search by tag