40 These differences in immune system differentiation moreover may underlie the higher incidence of allergic disease observed in formula-fed children. Not breastfeeding may also affect disease risk through exposure to foreign antigens in formula. Asthma Multiple studies have examined the association between infant feeding and development of asthma, with mixed results. In a meta-analysis, Ip and colleagues1 found a 1.7-fold risk (95% CI, 1.2�C2.3) of developing asthma among formula-fed children with a positive family history of asthma or atopy and a 1.4-fold risk (95% CI, 1.1�C1.7) among those without a family history, compared with those who were breastfed for 3 months or more. Gdalevich and associates41 compared less than 3 months of exclusive breastfeeding with greater than or equal to 3 months of exclusive breastfeeding and found a 1.

9-fold risk (95% CI, 1.3�C2.9) among those with a family history of asthma or atopy. Atopic Dermatitis Infants with a family history of atopy who were exclusively breastfed for less than 3 months have a 1.7-fold risk of atopic dermatitis (95% CI, 1.1�C2.4) compared with infants who are exclusively breastfed.42 Similar findings were reported in the PROBIT randomized trial of breastfeeding support,17 where infants who delivered in control hospitals were 1.9 times as likely (95% CI, 1.1�C3.2) to develop atopic dermatitis as those who delivered in breastfeeding support intervention hospitals. Type 1 Diabetes Epidemiologic studies have reported an association between exposure to cow��s milk antigen and development of type 1 diabetes, although results have been mixed.

43 Less than 3 months of breastfeeding has been associated with a 1.2- (95% CI, 1.1�C1.4)44 to 1.4-fold (95% CI, 1.2�C1.5)45 increased risk of developing type 1 diabetes compared with more than 3 months of breastfeeding. There is some evidence that differential recall between cases and controls may have biased results.44 A randomized, controlled trial is currently underway to test whether cow��s milk formula increases development of islet-cell antibodies. Infants at high risk of type 1 diabetes have been randomized to supplementation with hydrolysated formula versus cow��s milk formula. In a pilot study,46 exposure to cow��s milk-based formula was associated with higher prevalence of islet cell auto-antibodies, providing tentative evidence for a causal association between cow��s milk exposure and type 1 diabetes.

Childhood Cancer Several studies have examined associations between formula feeding and childhood leukemia based on the hypothesis that immunoreactive factors in breast milk may prevent viral infections implicated in the leukemia pathogenesis.47 Two meta-analyses1,48 found a 1.3-fold higher risk of acute lymphoblastic leukemia (95% CI, 1.1�C1.4) AV-951 among formula-fed children compared with children who were breastfed less than 6 months. Kwan and colleagues48 also found a 1.