Confronting models with data: the challenges of estimating disease spillover

Area of Science:

• Ecology
• Epidemiology
• Disease Ecology

Background:

• Strategic disease control necessitates understanding pathogen spillover dynamics.
• Estimating spillover risk involves correlating events with covariates or using mechanistic models.

Purpose of the Study:

• To highlight challenges and solutions in estimating spatio-temporal spillover risk at wildlife-livestock interfaces.
• To explore mechanistic approaches for predicting pathogen spillover.

Main Methods:

• Utilized case studies from the wildlife-livestock interface.
• Discussed challenges in aligning and integrating multi-species datasets.
• Explored the use of hierarchical models for spatio-temporal predictions.

Main Results:

• Data misalignment in space, time, and resolution poses significant challenges.
• Error propagation in predictions and computational demands for fine-resolution, broad-scale models are key issues.
• Confronting mechanistic predictions with observed events is crucial for understanding spillover.

Conclusions:

• Integrated, multi-disciplinary data collection is essential for robust spillover risk assessment.
• Mechanistic modeling, despite limitations, is vital for advancing pathogen spillover research.
• Addressing data integration and modeling challenges is key to improving disease control strategies.