Tuesday, December 04, 2012

Forecasting Flu Outbreaks With Computer Modeling

With winter upon us, flu outbreaks will gradually increase over the coming weeks as colder weather continues to march through North America.   A collaborative between Columbia University and the National Center for Atmospheric Research has developed a model that operates similar to the GFS or European weather forecast models and helps scientists get an idea of where flu may develop.

The study ran for five years (2003-2008) and was centered on New York City and was recently published in the Proceedings of the National Academy of the Sciences.

"Analogous to weather prediction, this system can potentially be used to estimate the probability of regional outbreaks of the flu several weeks in advance," said Alicia Karspeck, NCAR scientist and study co-author.

Karspeck, along with Jeffrey Shaman from Columbia University, found that flu outbreak peaks could be predicted with as much as seven weeks of lead time.

Using data that is refreshed from Google Flu Trends (a site I didn't know existed until I wrote this blog up), the modeling system is refreshed with data on a continuous basis...which allows Karspeck and Shaman to get a handle on where flu outbreaks are prone to increase over coming weeks as they can utilize real-time data.

If you hadn't seen their flu page, it's pretty impressive.  A look at flu outbreaks by state or by metropolitan area, with comparisons to the past five flu seasons prior to this one.  For those who care, flu cases at this point in the calendar are running higher than any of the last six years at this point and second highest on a cumulative, only bested by the early flu peak in 2009-2010.

Much like meteorological modeling, data is critical.  The more information that is available, the more accurate those flu outbreaks can be predicted.  Given that there are non-meteorological variables involved with the flu, the need to continually ingest data is critical in helping improve forecast outbreaks.

"Flu forecasting has the potential to significantly improve our ability to prepare for and manage the seasonal flu outbreaks that strike each year," Irene Eckstrand, a program director at the National Institutes of Health, said in a statement.

Shaman warns, however, that this is a work in progress.

"There is no guarantee that just because the method works in New York, it will work in Miami," Shaman says.

However, much like any other computer modeling, the science behind it will continue to improve the results over time.  It may miss badly, just like meteorological models have in the past...but it will show where the potential for outbreaks of flu in areas that researchers may not anticipate it occurring.