Greensboro, N.C.-based Cone Health is gathering data from several resources including Facebook to better monitor North Carolinians social interactions and predict future COVID-19 cases across the state, according to CBS affiliate WFMY.
Led by Cone Health's Chief Data and Analytics Officer Richard Pro, the health system's data science team is using the Facebook socialization data to better predict trends and possible case outbreaks in North Carolina weeks ahead.
"We use the data to better understand how (people) move around, the socialization data enables us to do what we call nowcasting," Mr. Pro told the network, adding that the data has been very accurate and that Cone Health often shares it with North Carolina's state and local health departments if they are able to identify future clusters.
Cone Health is also using automated contact tracing to support case predictions. The method tracks GPS signals on cell phones to see when two people encounter each other for more than five minutes. The data from the interactions can then be extended to help researchers predict future cases in a certain city or region.
Socialization data collected in July almost reflected North Carolina's biggest one-day spike, according to the report. Social interactions based on phone records peaked at 3.2 on July 2, compared to 1.4 interactions during the lockdown; sixteen days after the peak of social interactions, North Carolina recorded the highest number of reported cases to date.