Dissimilarities across age groups in the associations between complementary feeding practices and child growth: Evidence from rural Togo.

Adequate complementary feeding (CF) practices are essential for achieving optimal growth but challenging to measure comprehensively. This paper describes CF practices in 2,034 children aged 6-23 months and investigates their relationships with length-for-age z-score (LAZ) and stunting, using cross-sectional data collected from May to July 2014 in rural Northern Togo. The World Health Organization infant and young child feeding indicators were computed, along with ancillary indicators on feeding style and timing of introduction of complementary foods. The associations between those indicators and children's LAZ and stunting were assessed using linear and logistic regressions after stratification by age group and adjustment for children, maternal, and household characteristics. CF practices were suboptimal, and their associations with child's growth varied across indicators and age groups. In children aged 6-11 months, reaching the minimum dietary diversity and the minimum acceptable diet was associated with higher LAZ (p < .05). In 18- to 23-month-old children, only the consumption of iron-rich food was associated with both LAZ (p = .02) and stunting (p = .05). The late introduction of family foods was associated with higher odds of being stunted and lower LAZ in children aged 12-17 months (p < .001). The untimely introduction of porridge was associated with higher odds of stunting in children aged 9-23 months (p < .05). Unexpectedly, helping the child to eat was negatively associated with linear growth in all age groups. These findings nurture the ongoing process of review of the World Health Organization infant and young child feeding indicators showing that, in their current version, they hardly capture the links between CF and child's growth at different ages.