Why CDC Supports Flu Forecasting
When and where flu increases will occur, how large the impact of flu season will be, and when flu season will peak varies from season to season, making the preparation for and response to flu difficult. Flu forecasting can change that by offering the possibility to look into the future and better plan ahead, potentially reducing the impact of flu.
To help support the development of the science of flu forecasting and its application for public health, CDC, through the Epidemic Prediction Initiative (EPI), has organized FluSight challenges to forecast the timing, intensity, and short-term activity of influenza seasons since the 2013-2014 season. These challenges have provided the scientific and public health community experience in real-time forecasting, the ability to evaluate forecast accuracy, and experience in communicating and applying these forecasts in real-world settings. For example, forecasts are currently used to help inform CDC’s activity summaries provided to public health officials and CDC leadership and public messaging regarding the timing of the influenza season and how the public can protect themselves and their family.
How can flu forecasts be used prior to and during outbreaks?
The potential uses of flu forecasts extend beyond communication, both in seasonal and emergency situations. Flu forecasts can potentially be used to prepare for and prevent illness, hospitalization, and death, as well as the economic burden, experienced during the epidemic. When forecasts accurately predict flu activity, the ability to more effectively plan for public health responses to seasonal flu epidemics and future influenza pandemics is possible. Flu forecasts can inform messaging to health care providers regarding influenza vaccination and antiviral treatment for patients. Forecasts also can help to prepare for an influx of illnesses and hospitalizations, potentially helping inform the distribution and placement of health care staff and treatment resources. Finally, forecasts can be used to guide community mitigation strategies, such as school closures.
The flu forecasting challenges have provided experience in real-time forecasting, as well as experience in communicating and applying the results. These challenges also have offered a unique opportunity to evaluate forecast accuracy across different targets, seasons, geographic locations, and methods. Results from these evaluations enable researchers to prioritize future inquiries and help decision-makers better understand the strengths and limitations of current forecasts. These experiences are critical to developing a network of forecasters capable of providing results that public health officials can use.