To the Editor: International recommendations discourage feeding infants anything other than breast milk for the first 6 months of life1,2—in part, because of concern that supplementation would reduce a mother’s breast milk supply as a consequence of reduced suckling. Conversely, perceived “insufficient milk” is commonly reported by mothers throughout the world3–6 as a reason for early supplementary feeding or discontinuation of breastfeeding. In a recent commentary in EPIDEMIOLOGY, Kramer and colleagues reflect on a “formidable obstacle to causal inference”3(p. 793) regarding infant feeding: do differences in infant feeding behaviors cause differences in infant growth or do growth patterns instead affect infant feeding choices (“reverse causality”3,7)? Detailed prospective data with which to address these questions are rare. A study from Argentina provides some insights. Five hundred mothers who planned to breastfeed for 9 months, with no supplementary feeding for 4 months, were enrolled.8 Here, we consider the patterns of supplementation (milk or other foods) during the infants’ first 4 months of life (which was the duration of unsupplemented breastfeeding recommended at the time this study was conducted). The Table presents infant weight gain in relation to timing of supplementation, using the notation in the directed acyclic graph depicted by Kramer et al3 (eFigure, https://links.lww.com/EDE/A763; Kramer’s Figure 4). One-third of mothers who were observed at 4 months had already begun supplementation. For each month, weight velocity during the period leading up to the follow-up visit (which occurred at time j + 1) was highest for infants who were not yet supplemented, somewhat lower for those who had begun supplementation before this interval (before time j), lower still for those who began supplementation during this interval (j to j + 1), and lowest for those who began supplementation at the time of this visit.TABLE: Number of Subjects, Mean Daily Infant Weight Velocity Since Previous Follow-up Visit, and Percentage of Mothers Reporting Decreased Milk Production,a by Month and Timing of Supplementary Feeding (Described According to the DAG in the Article by Kramer et al3; see eFigure, http://links.lww.com/EDE/A763)At time j + 1, mothers who had begun supplementing before time j reported the highest percentages of perceived decreased milk production—presumably because supplementary foods had been providing some of the infants’ nutritional needs for at least 1 month. Once supplementation began, the reported daily breastfeeding frequency and average duration per feeding both declined, suggesting that these mothers were producing less milk. Of concern is the observation that, more than a month after supplementation had begun, these infants continued to grow more slowly than unsupplemented infants. Among mothers who began supplementing at time j + 1, maternal reports about decreased milk production expressed at that visit—as well as the slowest infant growth of the 4 groups—clearly preceded supplementation. (International recommendations1,2 instead encourage more frequent breastfeeding to stimulate the milk supply.) For mothers who began supplementing during the interval j to j + 1, we cannot be sure which mechanism is at work. However, we do know that at the previous visit (time j) few of these mothers had reported decreased milk production, and infant weight velocity had been relatively high. These results suggest that, for at least some of these mothers, the decision to begin supplements between these two visits was not prompted by perceived decreased milk production or slow weight gain but rather led to these problems. Among mothers who had not begun supplementation, very few reported that milk production was “unsatisfactory” (Table). These infants had the highest weight velocity. One limitation of our study is the small sample sizes in the three supplementing groups, due in part to study attrition; thus, confidence intervals are broad and overlapping. Nonetheless, similar patterns in all 4 months lend credence to these results. Analyses of larger data sets are warranted to explore the longitudinal associations of mothers’ reports of insufficient milk, supplementary feeding of young infants, and infant weight gain. Margaret F. McCann FHI360 Durham, NC, [email protected] Angel Victor Moggia Department of Obstetrics School of Medicine Buenos Aires University Buenos Aires, Argentina
Search from 250 M+ research papersSearch from 250 M+ research papers
Open Access
Apr 2, 2015
Jenna J Morgan + 1
