We appreciate the interest and comments of Plagemann et al. (1) regarding our study (2) on maternal status as a potential modifier of association of breast-feeding on childhood obesity. As noted, for the contrast of breast milk only versus formula only, the 95% CI excluded the null value, thus necessarily including values >1.0. From a statistical perspective, however, the best estimate for this contract is an odds ratio (OR) of 0.79, not a value >1.0. Furthermore, the test for dose response suggested a statistically significant trend in the direction of protection by breast-feeding for both groups. Finally, there is no indication of a differential effect of breast-feeding according to maternal status. Specifically, for both exclusivity and duration of breast-feeding, the interaction term from fully adjusted models was P = 0.50 and P = 0.66, respectively. Thus, our interpretation of the data is that there is no evidence of a deleterious effect of breast-feeding according to maternal obesity or diabetes status.
With regard to the second point raised by Plagemann et al. (1), the OR of 1.11 for overweight among children of diabetic mothers who were breast-fed <1 month compared with those who were formula fed was in the same direction (i.e., potentially deleterious) as observed in the previous work by Plagemann et al. (3). We note that for this specific contrast, the 95% CI was quite wide (0.22–5.60), making interpretation difficult. Interestingly, the OR for the same contrast for nondiabetic mothers with BMI >25 kg/m2 was also >1.0 (OR 1.46 [95% CI 1.01–2.13]). We noted in our original article (2) that to interpret this finding, one must consider potential circumstances related to the decision to stop breast-feeding at such a young age, as well as the infant-feeding behaviors in response to those circumstances. Here, while we agree with Plagemann et al. (1) that there is a need for further work in this regard, we note that the context of breast-feeding duration during the neonatal period (i.e., choosing to stop or to continue breast-feeding) is extremely important to consider, rather than simply focusing on this time period in isolation.
Finally, our work was, in fact, very specifically motivated by that of Plagemann et al., and thus we certainly agree that this topic is of considerable importance. Our comments in our study (2) regarding potential explanations for differences in our findings were speculative; thus, we have no further comments in this regard.
