Using a Community-Based Early Childhood Development Center as a Platform to Promote Production and Consumption Diversity Increases Children's Dietary Intake and Reduces Stunting in Malawi: A Cluster-Randomized Trial.

Background: Children in Malawi face nutritional risks related to low-quality diets and chronic malnutrition.

Objective: This study evaluated the impact of a 1-y early childhood development (ECD) center-based agriculture and nutrition intervention aimed at improving household production diversity, maternal knowledge on child nutrition and feeding practices, and children's diets and anthropometric measures.

Methods: A longitudinal cluster-randomized controlled trial was implemented in 60 community-based childcare centers (CBCCs), covering 1248 preschool children (aged 36-72 mo) and 304 younger siblings (aged 6-24 mo). CBCCs were randomly assigned to 1) a control group providing the Save the Children's ECD program or 2) a treatment group providing a standard ECD program with additional activities to improve nutritious food production and behavior change communication to improve diets and care practices for young children. Primary outcomes were household production and production diversity, preschooler enrollment and attendance, and dietary intake measured by quantitative 24-h recall and minimum diet diversity for younger siblings. Secondary outcomes included anthropometric measures for preschoolers and younger siblings, child development scores for preschoolers, and women's asset ownership and time use (the latter 2 are not discussed in this article). We used difference-in-difference (DID) estimates to assess impacts.

Results: Compared with the control group, preschool children in the intervention group had greater increases in nutrient intakes and in dietary diversity. No impacts on anthropometric measures were seen in preschoolers. Younger siblings in the intervention group had greater increases in height-for-age z scores than did children in the control group (DID: 0.44; P < 0.05) and greater reductions in the prevalence of stunting (DID: -17 percentage points; P < 0.05). The plausibility of the impact on growth in younger siblings was supported by effects along program impact pathways, including production of nutritious foods, caregiver knowledge, and dietary diversity.

Conclusion: Implementing an integrated agriculture and nutrition intervention through an ECD platform benefited children's diets and reduced stunting among younger siblings of targeted preschoolers. This trial was registered on the ISRCTN registry as ISCRCTN96497560.