Determinants of Stunting Among Under-Five Children in Katsina State: A Multilevel Analysis

Stunting, a marker of chronic undernutrition, is shaped by factors operating at individual, household, community, and systems levels, necessitating analytical approaches that capture this complexity. This study employs multilevel modelling to identify individual and contextual determinants of stunting among children under five in Funtua and Malumfashi local government areas of Katsina State, North West Nigeria. Katsina State consistently records stunting prevalence exceeding national averages, yet determinant analyses with appropriate multilevel frameworks are absent in the current literature. A community-based cross-sectional design was used with 1,200 children aged 0 to 59 months and their mothers recruited through multi-stage cluster sampling across 30 communities. Child height was measured and height-for-age z-scores calculated using WHO Anthro software, with stunting defined as a z-score below negative two. Individual-level determinants assessed included dietary diversity, breastfeeding history, birth weight, illness episodes, and immunisation status. Household-level factors included maternal education, household food security using the Food Insecurity Experience Scale, and socioeconomic quintile. Community-level variables included safe water access, health facility proximity, and sanitation coverage. A two-level logistic regression model was fitted using maximum likelihood estimation to partition variance between individual and community levels. Intraclass correlation coefficients were calculated to quantify the extent to which community-level clustering explains stunting variation. This multilevel approach provides a more precise and policy-relevant understanding of stunting determinants in Katsina State than conventional single-level analyses. Keywords: stunting, multilevel analysis, determinants, Katsina State, child malnutrition.