Rapid Responses to:

RESEARCH: Michael S Kramer, Lidia Matush, Irina Vanilovich, Robert Platt, Natalia Bogdanovich, Zinaida Sevkovskaya, Irina Dzikovich, Gyorgy Shishko, Bruce Mazer the Promotion of Breastfeeding Intervention Trial (PROBIT) Study Group Effect of prolonged and exclusive breast feeding on risk of allergy and asthma: cluster randomised trial BMJ 2007; 335: 815 [Abstract] [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

The effect of breast feeding on allergy and asthma

Karen M Silvers, Michael J Epton, Chris M Frampton   (16 October 2007)

Language and breastfeeding

Catherine E Walshe   (22 October 2007)

Weaning

Hugh Mann   (23 October 2007)

The effect of breast feeding on allergy and asthma 16 October 2007
Karen M Silvers, Senior Research Fellow University of Otago, Christchurch; PO Box 4345; Christchurch; New Zealand, Michael J Epton, Chris M Frampton Send response to journal: Re: The effect of breast feeding on allergy and asthma	The PROBIT study is a large randomised controlled study whose intervention was to reduce childhood gastrointestinal infection by promoting breast feeding. Secondary outcomes were childhood respiratory infection, atopic eczema, recurrent wheezing, the prevalence of any breastfeeding at 3, 6, 9 and 12 months, and the prevalence of exclusive and predominant breast feeding at 3 and 6 months [1]. Despite this, this more recent paper has been written as though the primary objective of the study was to test the association between prolonged and exclusive breast feeding, and asthma and allergy at 6.5 years of age. Therefore, when no statistical difference was found between the breast feeding promotion and control groups, the authors have erroneously concluded that breast feeding has no effect on these outcomes. Firstly, this conclusion can not be drawn from this study design, and secondly cannot be extended to other populations with different rates of breast feeding, asthma and allergy. The post-hoc analysis described at the end of the results section, with grouped breast feeding classes, more directly addresses the apparent theme of the paper, but in itself suffers from methodological and interpretation limitations. The effects of breast feeding being somewhat confounded with other effects inherent to the intervention arm. These issues need consideration before conclusions can be drawn about the effect of breast feeding on the development of asthma and allergy. For the obvious ethical and practical reasons stated by the authors, this intervention was a promotion of exclusivity and prolonged breast feeding among women who had already decided they would breast feed. Although valid, this approach is only able to test whether the duration of breastfeeding or exclusion of other allergenic substances in the first months of life reduces the risk of asthma and allergy in the children of mothers with a desire to breast feed. It does not allow other possible hypotheses to be explored, such as differences in the rates of asthma and allergy as a result of a mother’s decision to breast feed, the effect of colostrum and immediate skin-to-skin contact after birth. As expressed by the authors, further caution must be advised when extrapolating the results of this study to populations where asthma and allergy occur more frequently. The prevalences of allergic symptoms and diagnoses were extremely low, for example, the family (parental and sibling) history of atopy, a well established risk factor for asthma and allergy, was <5% compared to >80% seen in New Zealand not including siblings [2, 3]. It should also be noted that the relatively wide confidence intervals for a trial of this size raise questions about whether all the important confounding and predictor variables were included in the multivariate model. In addition, there are major concerns about the quality of the skin prick test which was the only objective measure of atopy in this study. Although it is possible that breast feeding does not protect against asthma and allergy, this study, despite being a large randomised trial, is unable to conclusively address the hypothesis. Instead, it has demonstrated that in a Belarussian population, an intervention to promote breast feeding among women with a desire to breast feed does not alter the risk of asthma and allergy at 6.5 years. Competing interests: None declared
Language and breastfeeding 22 October 2007
Catherine E Walshe, Breastfeeding counsellor WA13 0QP Send response to journal: Re: Language and breastfeeding	Whilst I read this article and the response to this with interest, and acknowledge that there are many issues of methodology and interpretation to be debated, I would like to ask that the BMJ in their publications relating to breastfeeding ask authors to acknowledge that breastfeeding is the physiological norm for human early nutrition. As such then, breastfeeding does not affect the risk of development of atopic diseases, rather it is not breastfeeding (and the introduction of breast milk substitutes) which is the risk factor. Breastfeeding, as the biological norm is not 'protective' (or not as the case may be), rather the alternative 'fails to protect' (or not as the case may be). Perhaps a pedantic viewpoint, but an important distinction. Competing interests: Catherine Walshe is a breastfeeding counsellor. She is responding in a personal capacity.
Weaning 23 October 2007
Hugh Mann, Physician Eagle Rock, MO 65641 USA Send response to journal: Re: Weaning	The purpose of weaning is not to remove milk from the diet. The purpose of weaning is to weaken the maternal-infant bond, so that the infant can be introduced to other people and other food. The maternal bond is the most powerful and enduring relationship in our lives. In fact, we never fully separate from our mothers. We just find maternal substitutes, like food, friends, teachers, spouses, and doctors. Competing interests: None declared