An influenza season forecasting challenge issued annually by the U.S. Centers for Disease Control and Prevention (CDC) has shown the value of cooperation among international researchers working to forecast influenza outbreaks.
The influenza season forecasting challenge is one issued annually by the CDC, but in the 2015-2016 session, 11 teams participated, submitting 14 predictive models along the way. Now, after review by a team of researchers from all over the world, the successes and failures of those efforts have been revealed.
In all, the teams fared best when tackling seasonal peak intensity and short-term forecast. However, their prediction successes tumbled when concerning timing of the season’s actual onset and when the peak week of outbreak would be. The most important takeaway, though, is the need for cooperation. Forecasting improved among teams that combined more than one model in development of their forecasts. It was also at its highest when all team forecasts were pulled into a single ensemble model, rather than any model relied upon individually.
“The results highlight the continuing challenge of improving forecast accuracy for more seasons and at lead times of several weeks or more; forecasts that would be of even more utility for public health officials,” the researchers wrote.
Combined model approaches and greater experience, the researchers conclude, could only benefit future forecasts. Their results and conclusions were published in the journal Scientific Reports.