Aim

It has been postulated that breastfeeding (BF) reduces the risk of allergy in children. However, recent studies failed to show a protective effect or suggested an increased risk of atopy with BF. We examined the association between BF and atopy, measured by skin prick test (SPT), in a large population-based cohort, stratifying by maternal hayfever.

Method

We assessed BF in the first years of life in children from Leicestershire, UK. At school age (9-14 years), we performed SPT for 4 allergens (cat, dog, dust, grass) in 1515 children. A child was defined as atopic if any wheal size was >3mm. We used logistic regression, adjusting for age, sex, ethnicity, birth season, family history of atopy and socio-economic status. We stratified by maternal hayfever and investigated potential reverse causation in analysis that excluded children with eczema or wheeze in infancy.

Results

465 (31%) children were not breastfed, 435 (29%) had BF <3 months (mo), 256 (17%) BF=4-6 mo and 348 (23%) BF >6 mo. 437 mothers (29%) had hayfever. We found no evidence for differences in risk of atopy by BF; the odds-ratios (95% confidence intervals) for BF<3 mo, BF=4-6 mo and BF>6 mo were 0.97 (0.69,1.37), 1.07 (0.72,1.59) and 1.19 (0.83,1.71), respectively, compared with no BF (p-trend=0.297). We did not find differential association by maternal hayfever (p-interaction=0.489) nor evidence for reverse causation.

Conclusion

Using objective measures and controlling for reverse causation, our study did not find evidence for an effect of breastfeeding on atopy in children of mothers either with or without hayfever. We can continue recommending breastfeeding for atopic and non-atopic mothers.