Play Live Radio
Next Up:
0:00 0:00
Available On Air Stations
Contact 90.5 WESA with a story idea or news tip:

CMU Map Helps Predict Spikes Of COVID Illness Across The US

Carnegie Mellon University|Delphi Group
Delphi map displaying the rate of doctor visits in US metro areas.

Carnegie Mellon University’s Delphi Research Group has published several interactive maps that provide insight into the spread of coronavirus at the metro area and county levels. This information could be useful as state and local officials consider easing social distancing restrictions.

The maps reportedly display “real-time" data on doctor visits, medical tests and browser searches related to COVID-19 in the United States. Delphi says its maps can also provide advance warning ahead of a COVID spike in a given area, which could help hospitals prepare for surges of patients.

“All of these signals are just rough indicators of COVID-19,” said Roni Rosenfeld, the co-leader of Delphi, in a press release. “Any one data source may not be conclusive, but if multiple sources indicate the same thing, people can have greater confidence about what is happening or will soon happen in various locales.”

Delphi, which specializes in epidemiological forecasting, collected its data with the help of Google, Facebook, an unnamed “national health system,” and Quidel Corp., a maker of flu tests. The symptoms of COVID-19 and influenza are similar. To rule out influenza, people experiencing COVID-like symptoms are often given a flu test.

In the Pittsburgh metro, metrics show that COVID activity seems to be dropping, with the exception of the flu test data map. There was not a sufficient amount of data on the metro level for Delphi to make an estimate, though statewide flu tests have gone down slightly.

The metrics of east coast metros like Philadelphia and New York City also seem to be on the decline. However, data suggest that illness levels in these locales are far more severe, which is consistent with the surge of COVID patients seen in those cities’ health systems.

Delphi’s use of Google and Facebook data was collected through surveys. Users were asked if they knew someone who was experiencing fever, cough and other symptoms related to COVID-19. Delphi said that some 600,000 Google users responded.

Researchers used the “Google Health Trends interface” to see how often people in a given location searched for COVID-related topics.

Delphi said it plans to include additional data sources in the coming weeks, and that the maps would be updated daily.

WESA receives funding from CMU.

Sarah Boden covers health and science for 90.5 WESA. Before coming to Pittsburgh in November 2017, she was a reporter for Iowa Public Radio. As a contributor to the NPR-Kaiser Health News Member Station Reporting Project on Health Care in the States, Sarah's print and audio reporting frequently appears on NPR and KFF Health News.