The IDEA model: A single equation approach to the Ebola forecasting challenge

Area of Science:

• Epidemiology
• Mathematical Biology
• Public Health

Background:

• Mathematical modeling is crucial for informing disease control policy during emerging infectious diseases.
• The 2014-2015 West African Ebola epidemic highlighted the need for effective forecasting approaches.
• Limited knowledge exists regarding the comparative performance of different forecasting models.

Purpose of the Study:

• To evaluate the performance of various mathematical models in forecasting simulated Ebola outbreaks.
• To assess the utility of a simple, phenomenological single-equation model (IDEA) against more complex methodologies.

Main Methods:

• The RAPIDD Ebola Forecasting Challenge (REFC) involved eight mathematical models simulating Ebola outbreaks.
• The IDEA model, a single-equation approach, utilized only case count data.
• Model fitting was conducted using a maximum likelihood estimation.

Main Results:

• The IDEA model demonstrated reasonable performance, ranking 4th or 5th on average among participating models.
• The model showed better suitability for long-term forecasts compared to short-term predictions.
• Confidence intervals improved when assuming a negative binomial distribution for case counts over a Poisson distribution.

Conclusions:

• The IDEA model, despite its simplicity, provides forecasts with accuracy comparable to more complex methods.
• Its ease of implementation in standard software makes it accessible for frontline public health personnel.
• The model offers a practical tool for generating forecasts to inform infectious disease control strategies.