Pregnancy and child

Evidence of breastfeeding increasing food allergies risks not clear-cut

"Breastfed babies may face more than double the risk of developing food allergies," the Mail Online reports.

The headline follows a Japanese study that questioned how parents fed their 6-month-old babies, whether they:

  • never breastfed
  • exclusively breastfed
  • partially breastfed – combined with using formula milk

They then asked whether their child had sought medical help for food allergy up to 5.5 years of age.

They found that exclusively breastfed babies had a 50% greater chance of having a food allergy compared with never breastfed babies. However, allergies were uncommon, affecting only 5-10% of babies, so this reflected only around a 3% absolute risk increase.

The researchers then looked into whether this was food allergy with or without eczema as an indicator of whether the child could be prone to allergies. They found increased risk with breastfeeding for food allergy without eczema, but a decreased risk (protective effect) of food allergy from partial breastfeeding in children who also had eczema.

The full results give an unhelpful and confusing picture. There are also many limitations, including the fact that results in Japanese women may be less applicable to the UK.

Even if there is a direct link between breastfeeding and food allergies, the small risk is likely to be outweighed by the many other benefits of breastfeeding. These include a reduced risk of your baby getting an infection while lowering your own risk of certain cancers, such as breast cancer.

Read more about the benefits of breastfeeding.

Where did the story come from?

This study was conducted by researchers from Okayama University in Japan. No sources of funding are reported and the authors declare no conflict of interest. The study was published in the peer-reviewed journal Allergology International and is freely available to access online.

The Mail Online coverage takes these findings at face value without considering the limitations. The statement that "infants may develop allergies if they are not exposed to foods early enough" is speculation.

The study has not assessed food introduction and, in any case, only analysed women and their babies up to 6 months when weaning with solid foods is not recommended.

And the doubled figure quoted by the news website only applies to women who exclusively breastfed and whose babies did not go on to develop eczema. This represents a smaller subset of the overall population studied.

What kind of research was this?

This was a population-based cohort study using data collected as part of the Longitudinal Survey of Newborns in the 21st Century in Japan, which started in 2001. The primary aim of this cohort was reported to be to investigate measures that could counter the declining birth rate in Japan.

Cohort studies are useful for looking at whether a particular exposure (breastfeeding in this case) is linked with a health outcome (allergies). However, they cannot prove direct cause and effect between the 2 as many confounding factors may be influencing the link. Particularly as this study was not set up to look at the link between infant feeding and later allergies, the data collection may not have taken account of all relevant factors.

What did the research involve?

The Longitudinal Survey specifically recruited babies born 10-17 January or 10-17 July in 2001. A total of 53,575 of these families were sent questionnaires when their babies were 6 months old; 88% responded (47,017) and were followed up with annual questionnaires.

At the first questionnaire, when the baby was 6 months, parents were asked about how they fed their child.

The following 5 annual questionnaires, up to when the child was around 5 years old, asked "whether the child had attended an outpatient clinic for various common diseases such as food allergy".

Parents where then said to have marked up on a questionnaire any food allergies their child had attended with in the past 12 months.

Information on eczema up to the age of 18 months was collected at the second annual questionnaire.

The researchers analysed the link between the assessed feeding practices and food allergy, taking account of issues that can influence both factors:

  • gender
  • whether they were of twin or multiple birth
  • prematurity or low birthweight
  • number of other siblings/children of the parents
  • mother's age at birth
  • educational level
  • maternal smoking status and history
  • area of residence
  • day-care attendance to 18 months

After exclusion of those with missing data on infant feeding, the researchers had a starting sample of 46,616, three-quarters of whom continued follow-up to 5.5 years.

What were the basic results?

Mothers of breastfed babies tended to have higher educational level and were less likely to be smokers. Their babies were less likely to be preterm or low birthweight or to have attended daycare.

Food allergies were uncommon, affecting around 5% of babies up to 18 months and 9% up to 5.5 years. Compared to babies never breastfed, those exclusively breastfed had 52% increased risk of food allergy up to 18 months, that just reached statistical significance (risk reduction [RR] 1.52, 95% confidence interval [CI] 1.02 to 2.25). This related to an absolute risk rate of 3.6% among the never breastfed group and 6.3% among those exclusively breastfed.

A similar risk increase was seen up to 5.5 years: rate 7.1% never breastfed vs 10.4% exclusively breastfed (RR 1.50, 95% CI 1.08 to 2.10).

The researchers then carried out sub-analysis according to whether babies who did develop food allergy also developed eczema.

Babies exclusively breastfed and partially breastfed had an increased risk of food allergy without eczema up to 5.5 years. Meanwhile for food allergy with eczema, partially breastfed babies now had decreased risk while those exclusively breastfed had no difference in risk.

When breaking down by duration of partial breastfeeding they tended to observe that the risk increase (for food allergy without eczema) was seen with the longer duration of partial breastfeeding to exclusive breastfeeding.

How did the researchers interpret the results?

The researchers conclude: "Our results showed that breastfeeding had prophylactic effects on food allergy only among high-risk children with eczema whereas prolonged breastfeeding increased the risk of food allergy."


The findings of this rather odd analysis should not deter mothers from breastfeeding on the basis that this may increase the risk of their child having a food allergy.

In the main analysis there was a 50% relative risk increase of child food allergy up to 5.5 years with exclusive breastfeeding compared with formula feeding, but:

  • we do not know if breastfeeding has directly caused the risk increase because unmeasured hereditary, health and environmental factors could be influencing the link
  • even if there is a true direct link, the absolute risk increase was small at only around 3%, which may be negligible when balanced against the benefits of breastfeeding
  • any discussion around relation to age of food introduction has limited relevance; the study has not questioned whether babies had food before 6 months, and weaning is not advised until 6 months of age anyway
  • we do not know anything about the type of food allergy – only that all of those examined required attendance at an outpatient clinic
  • the study is in a Japanese population, which may have distinct dietary practices and other characteristics meaning the findings cannot easily be applied to the UK or other populations

The researchers then looked at whether this was food allergy with or without eczema as an indicator of whether the child could be prone to allergies. But the recorded results were not particularly illuminating.

The main finding – "if your baby is prone to allergies, some breastfeeding – but not all – may protect against food allergies" – is confusing and unhelpful. Then with the above limitations, we also do not know if this is a true and accurate finding.

Breastfeeding is known to have many well-established benefits for both mother and baby. Ultimately, how you choose to feed your baby – whether exclusively breast, formula or a mix of the both – is personal choice.

Exclusive breast (or formula) feeding is recommended up to 6 months of age.

This will provide your baby with enough nutrition and support food introduction when they are better able to cope with solids and start to explore food themselves.

NHS Attribution