Breastfeeding and risk of overweight in childhood and beyond: a systematic review with emphasis on sibling-pair and intervention studies

Abstract

Breastfeeding is associated with a lower risk of subsequent overweight or obesity, but it is uncertain whether this is a causal relation because most studies have not adequately reduced risk of bias due to confounding. The aim of this review was to examine whether 1) ever compared with never consuming human milk and 2) different durations of human milk consumption among infants fed human milk are related to later risk of overweight or obesity, with emphasis on sibling-pair and intervention studies. The 2020 Dietary Guidelines Advisory Committee, together with the Nutrition Evidence Systematic Review team, conducted a systematic review of articles relevant to healthy full-term infants in countries with a high or very high level of human development. We searched PubMed, Embase, Cochrane, and CINAHL; dual-screened the results using predetermined criteria; extracted data from and assessed the risk of bias for each included study; qualitatively synthesized the evidence; developed conclusion statements; and graded the strength of the evidence. The review included 42 articles, including 6 cohorts with sibling-pair analyses and 1 randomized controlled trial of a breastfeeding promotion intervention. Moderate evidence suggested that ever, compared with never, consuming human milk is associated with a lower risk of overweight and obesity at ages 2 y and older, particularly if the duration of human milk consumption is>6 mo. However, residual confounding cannot be ruled out. Evidence was insufficient to determine the relation between the duration of any human milk consumption, among infants fed human milk, and overweight and/or obesity at age 2 y and older. Further research, using strong study designs, is needed to disentangle the complex relation between infant feeding practices and the risk of subsequent overweight or obesity, as well as the biological and behavioral mechanisms if the relation is causal.