Relationship between dietary patterns and stunting in preschool children: a cohort analysis from Kwale, Kenya

Objectives: Stunting is a significant cause of poor cognitive performance and lower school achievement. Stunting is observed among pre-school children in several areas in Africa; however, not all children are affected, and children with and without stunting are seen in the same communities. Therefore, this study aimed to identify nutritional and other factors that prevent stunting that may exist in local communities. Study design: This is a prospective cohort study. Methods: Data were extracted from the Health and Demographic Surveillance System conducted in Kwale County, Kenya. The cohort consisted of all households with children less than five years old, within a radius of 2.2 km from a local health centre. A dietary pattern (DP) survey with a semi-quantitative food frequency questionnaire was conducted on caretakers of children who were voluntary participated from the cohort between June 2012 and August 2012. Using cluster analysis, the children were assigned to a DP group. Logistic regression analysis was applied to calculate the adjusted odds ratios (aORs) of DPs for stunting controlling for other factors. Results: In total, 402 children were included in the analysis. By cluster analysis, three DPs were identified: protein-rich DP; traditional DP; and traditional DP complemented by breastfeeding. The aOR of a child becoming stunted from a normal height during the study period among children who received a traditional DP compared with those who had a protein-rich DP was 2.78 (95% confidence interval [CI]: 1.02–7.55). However, the aOR for children who were already stunted at the start of the study and had a traditional DP was 1.49 (95% CI: 0.82–2.72). Increased aORs of stunting were observed among children aged over 12 months compared with children aged 6–11 months, and the effects of DPs were modified by age in months from 12 to 35 months; however, the effects were near the null value for children over 36 months of age, although these were not statistically significant. Conclusions: We found that the traditional DP showed a higher risk for stunting compared with the protein-rich DP, and the most vulnerable age range for stunting was between 12 and 35 months. Interventions to prevent stunting should focus on providing 12- to 35-month-old children with locally available, protein-rich foods.