Every year in the US, influenza causes hundreds of thousands of hospitalizations and tens of thousands of deaths, but the timing and magnitude of the flu season varies, making it difficult to prepare for its impact. FluSight Challenge 2022/23 is Metaculus’s second annual tournament contributing forecasts to the yearly FluSight forecasting efforts of the Centers for Disease Control and Prevention (CDC).
Since 2013, the CDC’s yearly competition has engaged academic and private industry forecasting teams to submit weekly predictions supporting more effective public health responses. Flu forecasts by Metaculus and other forecasting teams can be invaluable to healthcare providers, and inform
- Flu vaccination and antiviral treatment timing
- Hospital preparation for large influxes of patients
- Community mitigation strategies
- The distribution of healthcare staff and resources
Metaculus is proud to again contribute predictions to the CDC FluSight Challenge while strengthening ongoing collaborations with computational scientist Thomas McAndrew, who leads Lehigh University’s Computational Uncertainty Lab, and biostatistician Evan Ray, who develops infectious disease forecasting models at the University of Massachusetts at Amherst. For FluSight 2022/23, Metaculus is also collaborating with Georgia Institute of Technology biomedical engineer and lecturer Todd Fernandez.
A $5,000 prize pool will be awarded after the challenge concludes in May of 2023.
Tournament Structure
Participants will forecast weekly confirmed flu hospitalizations for six states:
- New York
- Florida
- California
- Texas
- Arizona
- Pennsylvania
For each state, forecast questions will elicit flu count predictions for dates spaced two weeks apart. These forecasts will serve as inputs to a computational model that will project flu counts for every state and territory in the US. See the Model Description section for more detail.
Thank you to the forecasting community for sharing your predictions. Your contributions can lead to a more effective public health response and fewer flu-associated hospitalizations and deaths.