University of Massachusetts Amherst School of Public Health and Health Sciences associate professor of biostatistics Dr. Nicholas Reich will receive up to $3 million in funding over the next five years from the U.S. Centers for Disease Control and Prevention (CDC) to operate a UMass-based CDC Influenza Forecasting Center of Excellence, one of two in the nation. Dr. Reich, whose FluSight Network has produced some of the world’s most accurate models in recent years, leads a team that will work closely with the CDC, identifying new methods and data sources to sharpen the accuracy and improve communication of seasonal and pandemic flu forecasts.
Each year since 2010, influenza has caused between 9 million and 43 million illnesses in the U.S., hospitalized between 140,000 and 960,000 and claimed the lives of between 12,000 and 79,000 people, according to the CDC. These new predictive tools could more effectively target the public health response to a potential flu outbreak, helping to determine the timing for vaccine campaigns, as well as allocating antiviral medications. They could also help hospitals make the most efficient staffing decisions.
In recent years, flu forecasters have spread a wide net for their models, using Google search trends, HealthTweets and other nontraditional sources of information. “There is quite a bit of data floating around that aren’t part of the public health data record,” Dr. Reich says. “These data help us understand what’s happening right now and make more accurate projections.”
Dr. Reich’s group uses ensemble methodology, incorporating 21 models in an open platform that shares data and coding to maximize forecasting capabilities. “Pooling the strength of many models together, collaboratively with multiple teams, results in a more consistent and more accurate forecast,” he explains.Tags: Friday Letter Submission, Publish on October 18