Multiscale influenza forecasting

Influenza forecasting in the United States (US) is complex and challenging due to spatial and temporal variability, nested geographic scales of interest, and heterogeneous surveillance participation. Here we present Dante, a multiscale influenza forecasting model that learns rather than prescribes spatial, temporal, and surveillance data structure and generates coherent forecasts across state, regional, and national scales. We retrospectively compare Dante's short-term and seasonal forecasts for previous flu seasons to the Dynamic Bayesian Model (DBM), a leading competitor. Dante outperformed DBM for nearly all spatial units, flu seasons, geographic scales, and forecasting targets. Dante's sharper and more accurate forecasts also suggest greater public health utility. Dante placed 1st in the Centers for Disease Control and Prevention's prospective 2018/19 FluSight challenge in both the national and regional competition and the state competition. The methodology underpinning Dante can be used in other seasonal disease forecasting contexts having nested geographic scales of interest. Influenza forecasting in the United States is challenging and consequential, with the ability to improve the public health response. Here the authors show the performance of the multiscale flu forecasting model, Dante, that won the CDC's 2018/19 national, regional and state flu forecasting challenges.

Automated summary

Influenza forecasting in the United States is complex, but the multiscale flu forecasting model Dante has shown superior performance compared to other competitors, with sharper and more accurate forecasts that have potential for improving public health response. Dante's success in winning multiple competitions highlights its utility in seasonal disease forecasting with nested geographic scales of interest.