LAUSR

OBJECTIVE The study's objective was to investigate multiple underlying social, economic, and agricultural determinants of stunting among under-five children in three distinct ecological areas in rural Myanmar. DESIGN Repeated cross-sectional surveys in three states of Myanmar. SETTING Rural households in Chin (mountainous), Magway (plains) and Ayeyarwady (delta). PARTICIPANTS From two purposively selected adjacent townships in each state, we randomly selected 20 villages and, in each village, 30 households with under-five children. Households in the first survey in 2016 were revisited in late 2017 to capture seasonal variations. RESULTS Stunting increased from 40.4 % to 42.0 %, with the highest stunting prevalence in Chin state (62.4%). Univariate Poisson regression showed factors contributing to child stunting varied across the regions. Adjusted Poisson regression models showed that child's age and short maternal stature (aRR=1.14 for Chin, aRR=1.89 for Magway and aRR=1.86 for Ayeyarwady) were consistently associated with child stunting across three areas. For Chin, village-level indicators such as crop consumption (aRR=1.18), crop diversity (aRR=0.82) and land ownership (aRR=0.89) were significantly associated with stunting. In Magway, the number of household members (aRR=1.92), wealth status (aRR=0.46), food security status (aRR=1.14), land ownership (aRR=0.85), and in Ayeyarwady, women's decision-making (aRR=0.67) and indicators related to hygiene (aRR=1.13) and sanitation (aRR=1.45) were associated with stunting. CONCLUSIONS Area-specific factors were associated with stunting. Maternal short stature and child age were consistent determinants of stunting. A multi-sectoral local approach, including improvements in transport, is needed to address the intergenerational malnutrition problem.