# Assessing the correlation between STAR health risk forecasts with actual health emergencies in West and Central Africa: a three-year analysis

## TL;DR

This study assesses how well the WHO's STAR tool predicts health emergencies in West and Central Africa over three years, finding it effective for some diseases but limited by static data and low geographic detail.

## Contribution

The study provides a three-year evaluation of STAR's predictive accuracy across nine countries, identifying its strengths and limitations in forecasting health emergencies.

## Key findings

- STAR's meningitis forecasts were consistently accurate due to predictable seasonality and established patterns.
- Measles and cholera predictions were less reliable due to factors like immunization coverage and socio-political instability.
- Effective preparedness actions, such as vaccination campaigns, helped mitigate high-risk scenarios in some countries.

## Abstract

This study evaluates the Strategic Tool for Assessing Risks (STAR), developed by the World Health Organization (WHO), for its effectiveness in predicting public health emergencies across West and Central Africa during the period 2022–2024. STAR applies a composite risk scoring system based on four dimensions likelihood, severity, vulnerability, and coping capacity to classify hazards such as measles, cholera, and meningitis into five risk categories ranging from very low to very high.

Using a retrospective observational design, the study integrates quantitative outbreak data with qualitative assessments of preparedness actions across nine countries. The analysis demonstrates that STAR’s predictive accuracy varies significantly by hazard and context.

Meningitis forecasts were consistently accurate, primarily due to the disease’s strong seasonality and well-established epidemiological patterns in the African meningitis belt. In contrast, predictions for measles and cholera were less reliable, influenced by fluctuating immunization coverage, socio-political instability, and environmental factors such as water and sanitation conditions. Case studies illustrate these discrepancies: Burkina Faso’s cholera risk was overestimated, resulting in zero reported cases despite a high-risk classification, while Guinea’s measles outbreak closely matched STAR’s high-risk prediction. The findings also highlight that effective preparedness measures, including vaccination campaigns, hygiene promotion, and cross-border coordination, can mitigate high-risk scenarios, as observed in Gabon and Burkina Faso. Key themes emerging from the analysis include STAR’s strength in forecasting predictable hazards and its limitations due to static inputs and low geographic granularity.

While STAR is not a statistical forecasting model, its participatory, multi-sectoral approach provides strategic value by guiding planning, prioritization, and resource allocation for health emergency preparedness. It enables countries to optimize limited resources, prioritize highrisk hazards (scores 16–25), and implement preventive actions. Recommendations for improvement include recalibrating scoring parameters, integrating real-time surveillance and climate data, enhancing seasonality modeling, and increasing geographic resolution. When combined with dynamic data systems and collaborative efforts, STAR remains a critical strategic tool for strengthening regional public health resilience and supporting WHO’s Health Emergency Framework and all-hazards preparedness planning.