Interplay of individual and food environment factors shaping diets in the Kisumu urban informal settlements

Introduction Malnutrition persists in Kenyan urban informal settlements, disproportionately affecting women and children. This study explores the complex interplay between individual-level and food environment factors and the on influence dietary outcomes for women and children in these settings. Methodology We focus on women of reproductive age (15–49 years) and children aged 6–23 months in four informal settlements in Kisumu, Kenya. Using a cross-sectional design, we mapped food vendors and conducted household surveys ( n = 510) to assess socio-demographic characteristics, food shopping behavior, and dietary intake. Dietary outcome indicators included Dietary Diversity Scores, Dietary Species Richness, and fruit and vegetable (ProColor) diversity. Results Results show that <50% of women and children met the recommended Minimum Dietary Diversity, with diets heavily reliant on starchy foods. Our regression analyses significantly associate child age, maternal education, and marital status with child dietary outcomes. Marital status, nutrition attitudes, and food vendor density within the households’ immediate food environment are positively associated with some maternal dietary outcomes. Moderation and mediation analyses reveal that these associations are both conditional and indirect. The association between food shopping frequency and Children’s Dietary Diversity Scores (CDDS) was significantly stronger among older children (interaction effect = 0.01, p = 0.031), indicating an age-moderated effect. Additionally, wealth status and maternal education interaction significantly predicted both CDDS (interaction effect = 0.10, p = 0.047) and Children’s Dietary Species Richness(interaction effect = 0.30, p = 0.013), suggesting that maternal education modifies the influence of wealth on children’s diet. Among married women, food shopping frequency was more strongly associated with fruit and vegetable consumption (interaction effect = 0.12, p = 0.044). Conversely, significant negative interaction between vendor density in the immediate food environment and marital status predicted women’s Dietary Species Richness (interaction effect = −0.06, p = 0.001), favoring greater species richness in the diets of unmarried women. Mediation analysis identified nutrition knowledge and nutrition attitudes as primary mediating pathways for dietary outcomes. Conclusion These findings emphasize that improving diets in resource-poor settings requires holistic interventions that integrate structural access, socio-demographic realities, and cognitive drivers to foster resilient, equitable, and inclusive food systems.