New Hyde, N.Y.-based Northwell Health created a new predictive tool that can anticipate a spike in COVID-19 cases at its hospitals by mining user data patterns from the system's website, according to a Sept. 30 news release.
Four things to know:
1. The digital dashboard gathers 15 different indicators from the Northwell.edu website and inputs them into a machine learning algorithm to identify patterns in website traffic, including searches for emergency department wait times and physician page clicks.
2. The website indicators are anonymized for privacy and create a composite view of daily web traffic, which then translates into the "public mood" of the website.
"You're feeling sick, start doing the research online, but you aren't so ill that you need to see a doctor immediately," said Ramon Soto, senior vice president and chief marketing and communications officer at Northwell Health. "We found a very strong relationship with website activity to in-facility COVID-19 patients two weeks later."
3. The predictive model generates accurate results because of the traffic volume on Northwell's website; since the first week of March, Northwell has recorded more than 20 million page views coupled with the geographic concentration of its 19 hospitals, 800 outpatient facilities and 52 urgent care centers across New York City, Long Island and Westchester, N.Y.
4. Northwell plans to release the source code to other health systems to help them predict future COVID-19 cases.