Related Experiment Video
Updated: Mar 8, 2026
10:49
Visualizing Efficacy of Pesticides Against Disease Vector Mosquitoes in the Field
Published on: March 16, 2019
9.2K
Performance evaluation of an operational dengue forecasting system (D-MOSS) in Vietnam
Amy Marie Campbell1, Felipe Colón-González2, Do Kien Quoc3
1London School of Hygiene and Tropical Medicine, London, United Kingdom.
PLOS Global Public Health
|March 6, 2026
Summary
The Dengue forecasting Model Satellite-based System (D-MOSS) in Vietnam shows strong operational performance, outperforming baseline models for dengue incidence prediction up to six months ahead. Its forecasts offer significant value for public health decision-making.
Area of Science:
- Epidemiology
- Public Health
- Remote Sensing
Background:
- Dengue forecasting systems are crucial for public health interventions.
- Few systems have undergone rigorous prospective evaluation under operational conditions.
- The Dengue forecasting Model Satellite-based System (D-MOSS) was implemented in Vietnam in 2019.
Purpose of the Study:
- To comprehensively assess the statistical accuracy and operational utility of D-MOSS forecasts.
- To evaluate D-MOSS performance across various lead times and geographical regions in Vietnam.
- To determine the practical value of D-MOSS for public health decision-making scenarios.
Main Methods:
- Prospective evaluation of D-MOSS forecasts since operationalization.
- Statistical accuracy assessment (incidence, trajectory, peak timing).
- Operational utility assessment using probabilistic outbreak threshold exceedance.
Main Results:
- D-MOSS forecasts outperformed null model baselines across most metrics.
- Accuracy remained relatively high up to six-month lead times, with greatest value-add at 4-6 months.
- Larger errors were observed in central/southern provinces, but with greater value-add over baselines.
Conclusions:
- D-MOSS demonstrates strong predictive ability in an operational setting.
- Forecasts provide significant value for decision-making, especially at longer lead times.
- Insights gained can improve future operational dengue forecasting systems globally.
