Action Against Hunger Unveils Tool to Predict Malnutrition Months in Advance

A new forecasting model called MERIAM is producing rolling monthly malnutrition projections at a geographic precision far beyond existing humanitarian early warning systems — but the collapse of nutrition surveillance data since 2025 threatens to undermine the very models the sector is racing to build.

• MERIAM forecasts malnutrition hotspots down to ward level, months ahead of crisis, using climate, conflict, and health data

• Kenya used a MERIAM forecast to release preventive funds after the model's projections matched on-the-ground conditions almost exactly

• Nutrition surveillance tools have "fallen off a cliff" since 2025, creating a critical data gap that imperils all predictive malnutrition models

• Action Against Hunger plans to embed MERIAM into all anticipatory action planning over the next one to two years and envisions a real-time digital platform

The basics: MERIAM — Modeling Early Risk Indicators to Anticipate Malnutrition — is a statistical forecasting tool developed by Action Against Hunger in partnership with David Backer at the University of Maryland. It is designed to complement, not replace, existing humanitarian early warning systems like IPC (Integrated Food Security Phase Classification) and FEWS NET (Famine Early Warning Systems Network), which typically produce roughly one projection per country per year with a six- to 12-month horizon at the state or county level. MERIAM generates rolling monthly forecasts at two-, four-, and six-month intervals and can drill down to ward (municipal) level where data environments permit.