Our findings must be viewed in conjunction with several limitations of the study design; parental recall could have affected the accuracy of data on breast-feeding patterns, particularly if memory varied with the length of the recall period. Breast- and non-breast-feeding mothers might recall events differently. Moreover, parental recall bias may lead to underdetection of the association between breast-feeding and asthma in the event that mothers whose children are asthmatic subconsciously emphasize having breast-fed out of guilty.