The accuracy of dietary recall of infant feeding and food allergen data

Background Research investigating the association of infant dietary factors with later health outcomes often relies on maternal recall. It is unclear what the effect of recall bias is on the accuracy of the information obtained. The present study aimed to determine the extent of recall bias on the accuracy of infant feeding and food allergen data collected 10 years later. Methods Mothers were recruited from a prospective birth cohort from the Isle of Wight. When their child was 10 years of age (2011/2012), mothers were requested to complete a retrospective infant feeding questionnaire asking the same questions as those solicited in 2001/2002. Results In total, 125 mothers participated. There was substantial agreement for recollection of any breastfeeding (κ = 0.79) and the duration of breastfeeding from 10 years earlier (r = 0.84). Some 94% of mothers recalled accurately that their child had received formula milk. The exact age at which formula milk was first given was reliably answered (r = 0.63). The brand of formula milk was poorly recalled. Recall of age of introduction of solid food was not reliable (r = 0.16). The age of introduction of peanuts was the only food allergen that was recalled accurately (86%). Conclusions The present study highlights the importance of maternal recall bias of infant feeding practices over 10 years. Recall related to breastfeeding and formula feeding were reliable, whereas recalls related to age of introduction of solid or allergenic foods, apart from peanut, were not. Caution should be applied when interpreting studies relying on dietary recall.