Feeding practices and growth among young children during two seasons in rural Ethiopia

BACKGROUND: The use of indices of infant and young child feeding practices to predict growth has generated inconsistent results, possibly through age and seasonal confounding. The aim of this study was to evaluate the association of a dietary diversity score (DDS) and infant and child feeding index (ICFI) with growth among young children in a repeated cross-sectional and a follow-up study in two distinct seasons in rural southwest Ethiopia. METHODS: We used a repeated cross-sectional design comparing child feeding practices to nutritional status in 6–12 month old children during harvest (HS; n = 320) and pre-harvest season (PHS; n = 312). In addition, 6–12 month old children from the HS were reassessed 6 months later during PHS. In addition to child anthropometry, child feeding practices were collected using 24-h and 7-day dietary recalls. RESULTS: The mean (±SD) length-for-age z-score (LAZ) of the 6–12 month old children was −0.77 (±1.4) and −1.0 (±1.3) in HS and PHS, respectively, while the mean (±SD) of the follow-up children in PHS was −1.0 (±1.3). The median DDS (IQR) was 2.0 (1.0, 3.0.), 2.0 (2.0, 3.0) and 3.0 (2.0, 4.0) for the children in HS, PHS and the follow-up children in PHS, respectively. The DDS in HS was positively associated with LAZ at follow-up (β = 0.16; 95% CI: 0.01, 0.30; P = 0.03) after controlling for confounding factors. ICFI and DDS were not associated with mean LAZ, weight-for-height z-score and weight-for-age z-score within season. However, the odds of being stunted when having a DDS ≤ 2 was 2.3 times (95% CI: 1.10, 4.78; P = 0.03) higher compared to a DDS > 2 child in HS and 1.7 times (95% CI: 1.04, 2.71; P = 0.04) higher for the pooled sample of 6–12 months old children in HS and PHS. CONCLUSIONS: The DDS was found to be an indicator for child stunting during the Ethiopian harvest season. The DDS can be an appropriate tool to evaluate the association of child feeding practices with child growth irrespective of season. Inclusion of other dimensions in the construction of ICFI should be considered in future analysis as we found no association with growth. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s40795-017-0158-y) contains supplementary material, which is available to authorized users.