For the past few years, the CDC has not been the only place to go for tracking incidences of the flu. Google has also been tracking illness trends through identifying key search terms associated with illnesses like the flu, allergies, and sunburns. Google claims to have found a close relationship between the number of people who search for specific flu related topics, and the number of people who actually have flu symptoms. By counting the search queries, including specific terms, Google claims to be able to identify flu trends in real time with good accuracy as compared to CDC data. See Google’s tracking against the CDC data on one graph here.
This year, however, the Google model has varied from the model that the CDC puts out each week. Both models can be seen in the National Journal article here. The Google model shows that the U.S. is undergoing the most severe season in years, while the CDC model shows that the severity of this year’s season is still less than the H1N1 pandemic in 2009. So which model should be trusted? Even though the Google model tracks in real time, a week ahead of the CDC model, the National Journal suggests that the CDC is more accurate.
The CDC model can control more factors than the Google model. For example, if the CDC sees an increase in people going to their doctors for flu-like illness without an increase in lab samples testing positive for flu, it can look for signs of other respiratory viruses. The CDC will also call the state labs or individual physicians to try to get to the bottom of the discrepancy. Google, without these resources, may be tracking other illnesses with symptoms similar to the flu.
Still, the fact that the Google Trends model has been commonly accepted as reliable has the CDC thinking. In the future, the CDC hopes to get more information electronically and in real time. As more doctors get Electronic Health Records, more information can be shared faster, without taking up the valuable time of office workers. Electronic death certificates may also work to improve the accuracy of the CDC’s flu mortality rate. No matter how the information is transmitted, however, the CDC does not want to give up the personal phone calls and conversations with state labs and local physicians that help it identify odd or unusual patterns.