Approximately 10% of all babies worldwide are born preterm, and preterm birth is the leading cause of perinatal mortality in developed countries. Although preterm birth is associated with adverse short- and long-term health outcomes, it is not yet clear whether this relationship is causal. Rather, there is evidence that reduced foetal growth, preterm birth and the long-term health effects of both of these may all arise from a suboptimal intrauterine environment. Further, most infants born preterm also experience suboptimal postnatal growth, with potential adverse effects on long-term health and development. A number of interventions are used widely in the neonatal period to optimise postnatal growth and development. These commonly include supplementation with macronutrients and/or micronutrients, all of which have potential short-term risks and benefits for the preterm infant, whereas the long-term health consequences are largely unknown. Importantly, more rapid postnatal growth trajectory (and the interventions required to achieve this) may result in improved neurological outcomes at the expense of increased cardiovascular risk in later life.

The impact of variations in current infant feeding practice on bone mineral accrual is not known. We examined the associations between duration of breast-feeding and compliance with infant dietary guidelines and later bone size and density at age 4 years. At total of 599 (318 boys) mother–child pairs were recruited from the Southampton Women's Survey. Duration of breast-feeding was recorded and infant diet was assessed at 6 and 12 months using FFQ. At 6 and 12 months the most important dietary pattern, defined by principal component analysis, was characterised by high consumption of vegetables, fruits and home-prepared foods. As this was consistent with infant feeding recommendations, it was denoted the ‘infant guidelines’ pattern. At age 4 years, children underwent assessment of whole-body bone size and density using a Hologic Discovery dual-energy X-ray absorptiometry instrument. Correlation methods were used to explore the relationships between infant dietary variables and bone mineral. There was no association between duration of breast-feeding in the first year of life and 4-year bone size or density. ‘Infant guidelines’ pattern scores at 6 and 12 months were also unrelated to bone mass at age 4 years. We observed wide variations in current infant feeding practice, but these variations were not associated with differences in childhood bone mass at age 4 years.

It is not known what constitutes an optimal diet in infancy. There are relatively few studies of weaning practice in the UK, and there is a need for prospective data on the effects of infant diet and nutrition on health in later life. We describe the dietary patterns, defined using principal components analysis of FFQ data, of 1434 infants aged 6 and 12 months, born between 1999 and 2003. The two most important dietary patterns identified at 6 and 12 months were very similar. The first pattern was characterised by high consumption of fruit, vegetables and home-prepared foods (‘infant guidelines’ pattern). The second pattern was characterised by high consumption of bread, savoury snacks, biscuits and chips (‘adult foods’ pattern). Dietary pattern scores were correlated at 6 and 12 months (r 0·46 ‘infant guidelines’; r 0·45 ‘adult foods’). These patterns, which reflect wide variations in weaning practice, are associated with maternal and family characteristics. A key influence on the infant diet is the quality of the maternal diet. Women who comply with dietary recommendations, and who have high intakes of fruit and vegetables, wholemeal bread and rice and pasta, are more likely to have infants who have comparable diets – with high ‘infant guidelines’ pattern scores. Conversely, women whose own diets are characterised by high intakes of chips, white bread, crisps and sweets are more likely to have infants who have high ‘adult foods’ pattern scores. The effects of these patterns on growth and development, and on long-term outcomes need to be investigated.