To examine the influence of initial infant feeding on obesity in later life.
A systematic review of published studies investigating the association between infant feeding and a measure of obesity was performed with Medline (1966 onward) and Embase (1980 onward) databases, supplemented with manual searches. Data extraction was conducted by 2 authors. Analyses were based on odds ratios of obesity among initially breastfed subjects, compared with formula-fed subjects, pooled with fixed-effects models.
Sixty-one studies reported on the relationship of infant feeding to a measure of obesity in later life; of these, 28 (298900 subjects) provided odds ratio estimates. In these studies, breastfeeding was associated with a reduced risk of obesity, compared with formula feeding (odds ratio: 0.87; 95% confidence interval [CI]: 0.85-0.89). The inverse association between breastfeeding and obesity was particularly strong in 11 small studies of <500 subjects (odds ratio: 0.43; 95% CI: 0.33-0.55) but was still apparent in larger studies of > or =500 subjects (odds ratio: 0.88; 95% CI: 0.85-0.90). In 6 studies that adjusted for all 3 major potential confounding factors (parental obesity, maternal smoking, and social class), the inverse association was reduced markedly (from an odds ratio of 0.86 to 0.93) but not abolished. A sensitivity analysis examining the potential impact of the results of 33 published studies (12505 subjects) that did not provide odds ratios (mostly reporting no relationship between breastfeeding and obesity) showed little effect on the results.
Initial breastfeeding protects against obesity in later life. However, a further review including large unpublished studies exploring the effect of confounding factors in more detail is needed.