Materials associated with the Analytic Solutions for Real-Time Biosurveillance: Infectious Disease Forecast Modeling consultancy held October 29-30, 2015 in Falls Church, Virginia.
Problem Summary
New efforts are underway to design and evaluate forecasting models for diseases such as dengue and influenza. Critically, these are common diseases with ample data to both design and test models. The interagency Dengue Forecasting Project was designed to use forecasts to address critical healthcare system decision-making associated with dengue epidemics. While clear definitions for comparing models have been implemented, it remains unclear how to measure their utility for decision-makers. Essentially, how good does a forecast have to be to be useful for decision-makers? And how can that be quantified? This is both a specific question for the Dengue Forecasting project, but also a more broad effort to set objective bars for forecasting implementation. Using dengue as an example, there will now be numerous forecasts and metrics to determine which is "best." However, it will not be clear if any of these should be implemented for public health use, nor how good they would need to be for that to happen.
Attachments
- Use case summary
- Consultancy agenda
- List of attendees
- Final report