A study was recently brought to my attention that basically advocates supplementing with formula within the first two weeks of birth in order to reduce likelihood of cow’s milk allergy. The first author of the study, in interviews by the press, stated his recommendation to breastfeed while “simply complementing it with cow’s milk early on.” He says, “Let Dad enjoy some midnight infant bonding, while he delivers a dose or two of cow’s milk protein.” He specifies in the conclusions of the study that “The data should not be interpreted as discouraging breastfeeding.”

This is not a study that I’d seen before, but I do find it interesting from an immunological perspective. The study essentially goes to tolerance. Basically, your immune cells need to learn “self” vs “non-self”, and tolerance is used to tell them not to attack “self”. Sometimes this gets messed up, and you end up with an autoimmune condition – essentially, your immune cells attacking some part of “self” that it should have had tolerance to. There are also plenty of “non-self” things though, that you don’t want the immune system to attack – food, pollen, a fetus… The body should therefore also be able to tell “danger” from “non-danger”. When the immune system mounts an attack on something that we consider “non-danger”, that’s typically what we call an allergy. It’s a tricky business for the immune system to learn what it should and should not attack, and immunologists are very interested in learning how it works, which is still not fully understood. If we could manipulate this system, it could lead to many medical breakthroughs in autoimmune disease, allergies, transplant medicine, and the like.

Unfortunately, we still don’t know exactly how it does work. This study is postulating that exposure to the allergen within the first two weeks of life, seems to result in a higher rate of tolerance to that allergen (in this case, cow’s milk protein). This study though, is purely observational and does not even begin to propose a mechanism for how and why this would work. Without a mechanism, one observational study is not going to be enough to influence the medical associations and public health boards to change their current recommendation of exclusive breastfeeding for at least 6 months – exclusive breastfeeding being defined as no water, juice, formula, cereal, or foods.

Some other thoughts on this study:

**The reported incidence of cow’s milk allergy (CMA) is 2-5%, yet this study only found it in 0.5% of their population. They either have a skewed demographic or they have different diagnostic criteria than other clinicians. In addition, 381 cases were reported to the researchers by the parents as CMA, but the researchers diagnosed only 66 of these as actually allergic using their criteria. If parents did not report CMA, the researchers assumed CMA was not present. Only those who reported CMA were tested for it. Also of note, this study did not find any correlation of rate of CMA with other factors previously shown to affect likelihood, including gender, type of delivery, and genetic background.

**The study purports to relate first exposure of cow’s milk protein with incidence of CMA, however the researchers excluded any formula that was given to newborns while still in the hospital (estimated to be ~70% of all Israeli babies) as inconsequential.

**The study also excludes babies who were not exposed to cow’s milk or cow’s milk formula at all within the first year. The general recommendation from medical and public health associations is to breastfeed exclusively for 6 months, followed by continued breastfeeding with the introduction of complementary foods until at least 12 months, with introduction of cow’s milk after 12 months. Therefore, all babies whose parents actually followed this advice were excluded from the study and there is no data on their rate of CMA. These babies were excluded presumably because babies typically outgrow a cow’s milk protein allergy after the first year anyway. Babies who were delayed introduction to cow’s milk until after this time are not at a disadvantage. Cow’s milk is not a necessary part of a human diet, particularly a human infant whose diet consists of human milk.

**The study found a 0.5% rate of CMA for introduction of cow’s milk protein at most time-points. The publicized result was that the rate was 10 times lower when introduced at less than 14 days. However, it is also worthy to note that the incidence was 3.5 times higher when cow’s milk protein was introduced between 3.5-6.5 months. And, as mentioned above, the rate for those who were not exposed to cow’s milk protein until after 12 months was not given because these infants were excluded from the study.

**The study measures the first (non-hospital) exposure to cow’s milk protein in days, but it does so through parental interview at two-month intervals. Recall rate down to the exact day for introduction of cow’s milk protein from an interview up to two months later seems prone to inaccuracy given that parents were not asked to keep a daily diary. Though the study calls itself prospective because subjects were enrolled at birth, the methodology for data collection was retrospective.

**The study did report on dairy product consumption by the mother, however less than 1% of mothers did not consume dairy. Of note, all of the mothers who did not consume dairy had babies who did not have CMA, but the numbers were very small and there was no statistical significance.

**The study was funded by the Israel Dairy Board, a potential conflict of interest. Note that the conclusions of the study are to begin supplementing breastfeeding with formula made from cow’s milk within the first two weeks of birth, which would be of financial benefit to the Dairy Board.

My personal conclusions: Benefits of introducing cow’s milk formula within the first 2 weeks may include reducing the likelihood of cow’s milk allergy by inducing early tolerance. Costs of introducing cow’s milk formula though are loss of exclusive breastfeeding (and associated health implications) and increased likelihood of earlier weaning in general. Cow’s milk allergy is typically outgrown within the first year anyway, and delay of introduction of cow’s milk or cow’s milk formula until after 12 months also reduces the likelihood of cow’s milk allergy while also avoiding the other risks inherent in formula feeding. Introduction of cow’s milk formula between 3.5-6.5 months may be the worst scenario with respect to cow’s milk allergy. Even if reduced risk of cow’s milk allergy was only to be achieved through early use of formula, I do not think that this would be worth risking the other health implications of not exclusively breastfeeding.