Breast milk sodium content in rural Gambian women: between -and within-women variation in the first 6 months after delivery

It has been suggested that infancy is a particularly sensitive period with respect to the effect of dietary sodium on future risk of hypertension. One difficulty of researching the effects of early sodium intake on later health is accurately measuring sodium intake from breast milk. In observational studies, sodium content has been calculated by estimating breast milk volume consumed and assuming a fixed sodium concentration for all women at all times (a standardised measure). The objectives of this study were to investigate the variation in breast milk sodium concentration in the first 6 months postpartum within women and test whether the pattern of change in sodium concentration differs between women. The study population was 197 rural Gambian women. Multilevel models were used to investigate whether the sodium content of breast milk changed over time within and between women. Fractional polynomials were used to identify the best-fitting functions of age to be included in the within and between variance functions. Sodium levels decreased with time; the reduction was initially rapid (levels decreasing by 17.7% between 30 and 60 days after delivery). Immediately after birth, there was substantial variation in breast milk sodium content between women but this reduced with time. Our results suggest that it is not appropriate to use a standardised measure of breast milk sodium content when direct measurement is possible – particularly when there is a research interest in measuring sodium intake in very early infancy.