Integrated approaches to improve the nutritional status of children under five: a positive deviance hearth model in Bomet county, Kenya

Kenya experiences the triple burden of malnutrition, and despite a notable decline in stunting over the last decade, Bomet County continues to report very high stunting levels. This study sought to assess the effect of the PD-Hearth model on the weight of children 6–24 months and explore contextual determinants of malnutrition in Konoin, Bomet County. An exploratory sequential design was used, comprising a qualitative Positive Deviance Inquiry and a quantitative quasi-experimental non-randomized study with a non-equivalent control group. Phase one targeted adults aged ≥ 18 years, and phase two targeted underweight children aged 6–24 months and their caregivers. Phase two sample size was determined using G*Power software. Data was analyzed using SPSS version 22. Differences in weight, MUAC, and nutritional status between baseline and endline, and between study groups, were assessed using t-tests, chi-square tests, and difference-in-differences analysis, with p < 0.05 at the 95% confidence interval considered significant. Phase 1 showed that 85% of respondents had post-primary education, with 67.1% being farmers. In Phase 2, out of 110 children sampled, 53.6% were male. Caregivers demonstrated awareness of childcare practices, though these were limited by poverty, workload, and cultural beliefs. At baseline, the study groups showed no significant differences in age, weight, height, or MUAC, with p-values > 0.05 and a 95% confidence interval. The difference-in-difference analysis indicated that the proportion of well-nourished children increased by 7.3% in the intervention group compared to 1.8% in the control group (DiD = + 5.5%) for the weight-for-age z-score. For the weight-for-height z-score, the proportion of well-nourished children increased by 20.0% in the intervention group versus 3.6% in controls (DiD = + 16.4%). This analysis confirmed better outcomes regarding underweight and wasting in the intervention group, suggesting a meaningful positive impact of the intervention on these outcomes. Additionally, the mean weight gain was higher in the intervention group (0.15 kg, SD 0.32) than in the control group (0.10 kg, SD 0.14) over the study period. The PD-Hearth model demonstrated a positive short-term effect on weight and nutritional status among children aged 6–24 months, highlighting its potential as an effective community-based approach for addressing underweight in high-burden settings such as Konoin.