Researchers combined EHR data from athenahealth with historical flu outbreak patterns and a machine-learning algorithm to estimate flu activity in near real time. The estimates created using the cloud-based EHR approach had two to three times fewer errors than older models. Additionally, the algorithm correctly estimated the timing and magnitude of the national peak week during three flu seasons.
Ultimately, the study showed EHR data, historical patterns of flu activity and a robust machine-learning algorithm can monitor infectious diseases at the national and local level.
“Our methodology provides timely flu estimates with the accuracy and specificity of sentinel systems like the CDC’s [influenza-like illnesses] surveillance network,” the authors concluded. “This demonstrates the value of cloud-based electronic health records databases for public health surveillance at the local level.”
More articles on the flu:
US flu activity at lowest point in 5 months
Flu vaccine produces better outcomes when administered in the morning
As cases of influenza A fall, B strain infections rise
