Infant feeding method and special educational need in Scottish schoolchildren: a national, retrospective cohort study

Abstract

BackgroundAlthough special educational needs (SEN) are increasingly recorded among schoolchildren, infant breastfeeding has been associated with reduced incidence of childhood physical and mental health problems. This study investigated relationships between infant feeding methods and the risk of SEN. MethodsA retrospective population cohort of schoolchildren in Scotland was constructed by linking together health (maternity, birth, and health visitor records) and education (annual school pupil census) databases. Inclusion was restricted to singleton children, born in Scotland from 2004 onwards with available breastfeeding data and who attended local authority mainstream or special schools between 2009 and 2013 (most up to date data available). Ethical approval was not required. Generalised estimating equation models with a binomial distribution and logit link function investigated associations between infant feeding method at 6–8 weeks and any-cause and cause-specific SEN, adjusting for sociodemographic (child age, sex, deprivation quintile, and ethnicity) and maternity (age, smoking, parity, 5-min Apgar score, estimated gestation, sex-gestation-specific birthweight centile, marital-status, and mode of delivery) factors. FindingsOf 191 745 children meeting inclusion criteria, 126 907 (66·2%) were formula-fed, 48 473 (25·3%) exclusively breastfed, and 16 365 (8·5%) mixed fed. Overall, 23 141 (12·1%) children required SEN. Compared with formula feeding, mixed feeding and exclusive breastfeeding were associated with decreased any-cause SEN (OR 0·90, 95% CI 0·84–0·95; and 0·78, 0·75–0·82), and SEN attributed to learning disabilities (0·75, 0·65–0·87; and 0·66, 0·59–0·74), and learning difficulties (0·85, 0·77–0·94; and 0·75, 0·70–0·81). Compared with formula feeding, exclusively breastfed children had less communication problems (0·81, 0·74–0·88), social-emotional-behavioural difficulties (0·77, 0·70–0·84), sensory impairments (0·79, 0·65–0·95), physical motor disabilities (0·78, 0·66–0·91), and physical health conditions (0·74, 0·63–0·87). There were no significant associations for mixed-fed children. Feeding method was not significantly associated with mental health conditions or autism. InterpretationOnly the 6-8 week feeding method was available, precluding differentiation between never-breastfed infants and those who stopped breastfeeding before 6 weeks. The findings provide further evidence of potential benefits of breastfeeding for child health, lending weight to the need to support mothers to breastfeed. Resources have increased recently but should remain a national priority. FundingHealth Data Research UK (grant reference number MR/S003800/1).