Peer-reviewed article

Impact of Central Asia Stunting Initiative on Stunting and Anemia Among Children and Women of Afghanistan

Imran A. Chauhadry, Aminah Jahangir, Muhammad Umer, Khakerah Rashidi, Habibullah Haziq et al. · 1 Oct 2026

Abstract

ABSTRACT The Central Asia Stunting Initiative (CASI) is a program designed to address malnutrition among women of reproductive age and children under five in Badakhshan Province, Afghanistan, a region facing significant nutritional challenges. This study used a pre‐post design to assess the impact of CASI interventions, with cross‐sectional surveys conducted at baseline and midline among non‐pregnant women of reproductive age (15–49 years) and children under five using a two‐stage stratified random sampling approach; data collection included anthropometric measurements and hemoglobin assessments. Between the baseline and midline surveys, access to safe drinking water improved from 80.9% to 90.7%, and sanitation improved from 37.4% to 88.3%, while the severe food insecurity rate dropped from 94.7% to 30.5% ( p < 0.001). Exclusive breastfeeding among children aged 0–5 months increased from 64.0% to 80.9%, and dietary diversity among children 6–24 months rose from 7.5% to 20.7% (both p < 0.05). Child under nutrition declined significantly, with underweight decreasing from 35.0% to 15.5% (Cohen's h = 0.46) and stunting from 52.3% to 37.3% (Cohen's h = 0.30). The prevalence of moderate anemia among children fell from 59.7% to 41.1%, while normal hemoglobin levels rose from 39.2% to 58.1% (both p < 0.001). Among women, underweight prevalence declined from 20.0% to 9.4% ( p < 0.001), while overweight prevalence rose from 5.4% to 10.7% ( p = 0.003). Adjusted models confirmed lower odds of stunting, underweight, and anemia at midline compared with baseline. These findings suggest that CASI was associated with substantial improvements in maternal and child nutrition indicators in Badakhshan, and policymakers should focus on scaling up effective interventions and ensuring continuous monitoring. However, given the single‐group pre‐post design without a comparison group, these findings should be interpreted as temporal associations rather than causal effects.