Development and sustainability of interventions to improve child nutrition: a case-study of ethnic minority communities in northern Thailand

Abstract

Background: One third of northern Thai indigenous children aged 0–5 years are stunted. To address this, a Thai research group implemented a nutrition-sensitive agriculture (NSA) intervention in which villagers were supported for 6 months to keep hens and home gardens. The studies undertaken for this thesis were designed to enhance this initiative and its evaluation. Aims: First, to describe and understand infant and young child feeding (IYCF) practices and the underlying circumstances contributing to stunting in children. Second, to assess the outcomes of the NSA intervention in terms of IYCF and stunting, and the prospects for sustainability. Methods: A mixed-methods study involved people from eight villages and staff of the health system serving the area. Prior to the intervention, perspectives on child health and IYCF were explored in a two-day workshop and 30 in-depth interviews. Detailed thematic analysis of transcripts was undertaken. This complemented quantitative data obtained through 172 household surveys which was statistically analysed to profile stunted children and their families. Four of the villages became sites for the NSA intervention and four were controls. Household surveys were repeated after 12 months, permitting statistical assessment of the effectiveness of the intervention. In-depth interviews were undertaken to explore local views about the intervention and its prospects, with thematic analysis carried out. Results: The baseline survey showed that exclusive breastfeeding for six months was rare and children aged 6–11 months had little variety in their diets. Stunting peaked in children aged 12–23 months (affecting 42%). Half of households experienced severe food insecurity. Compared to non-stunted children, stunted children were not as likely to meet minimum dietary diversity (82% vs 63%, p=0.01), however, most household factors did not distinguish between stunted and non-stunted children. Situational and social circumstances shaped infant and young child feeding. Cultural beliefs meant that water and rice were given to children before six months of age. Mothers had to work in distant fields within a few months of giving birth, most often leaving the children to be looked after by grandparents. Villagers considered strength and independence of children to be hallmarks of health. In contrast the health system uses anthropometric measurements to characterise child growth and health. Twelve months after commencement of the NSA, 18% more children living in households receiving the intervention consumed eggs as compared to controls (mean difference 0.18, 95% CI 0.05, 0.30). However, there was no evidence of an impact on dietary diversity, food insecurity or stunting. The intervention was seen as beneficial by the villagers despite activities petering out. Conclusions: Food insecurity is a major problem in the villages. IYCF practices that facilitate women’s return to work within a few months of giving birth also appear to contribute to stunting in children. Growth monitoring was not accompanied by nutrition counselling and its purpose was unclear to locals. The NSA intervention had some short-term benefits. There may be opportunities to increase its impact through engagement with the health system and improvements to infrastructure.