A digital "smart" thermometer and a paired phone app is enabling a company to take consumers' temperature data from across the country and provide real-time insights into emerging COVID-19 hot spots.

Researchers at Kinsa Health – which has marketed its Bluetooth-enabled thermometers and a service to direct consumers to health services for several years – now believes its data is able to find unusual instances of people with abnormally high temperatures.

The early results in mapping data from across the country could be important in helping healthcare organizations and potentially government agencies identify areas where there's a higher-than-expected concentration of people with fevers – one of the leading indicators of infection with the COVID-19 virus.

That could give healthcare organizations advance warning and help them anticipate surges in service demands for patients with the virus, said Daniel Doherty, technical product manager at Kinsa. Typically, providers have relied on reports from the Centers for Disease Control to track contagious diseases, but the CDC relies on reports from clinical environments, allowing only a look at past activity. That doesn't help organizations looking for advance warnings of service demands, Doherty said.

The potential for such advanced warnings has drawn coverage for Kinsa's efforts from the popular press, most notably in a New York Times article last week.

In past years, Kinsa has been able to marshal these data capturing and analytical capabilities during flu seasons, which it markets to healthcare organizations as Kinsa Insights. Recently, with the flu season fading away, Kinsa has continued to identify areas with a preponderance of consumers with high temperatures.

Similar to its flu-mapping services, Kinsa is now tracking atypical illness through its digital thermometers. This past week, it released results that indicated correlations between unusually high temperature readings, followed by increases in reports of COVID-19 cases, said Nita Nehru, director of communications and partnerships for San Francisco-based Kinsa.

Consumers with Kinsa thermometers can upload results to the company's free app, which can give them direction for appropriate ways to assess the severity of a potential illness, provides relevant healthcare information or directs them to appropriate healthcare services.

Anonymized data uploaded to Kinsa from the app then provides.

"With our historical data, we can understand the 'illness signature' of a particular geography, and we understand what it should be at a given time of the year," Nehru said. "We can see how it compares with current results and it serves as an early warning system that indicates when something unusual is happening."

With more than one million of its thermometers in use across the country, Kinsa has enough data to map atypical illnesses down to the county level.

It's having a hard time keeping up with current demand for its thermometers, selling more than 10,000 a day, Nehru said. Worried consumers also are taking more temperatures, she added – while the company gets tens of thousands of temperature readings typically, it recently received data in excess of 100,000 readings in one day, she noted.

It's the level of abnormal results "that correlates with COVID," Doherty said. "Checking for fevers is one of the first things that people check," when trying to diagnose infection with the virus. The nature of this pandemic is that early on it doesn't show up as really big numbers. The danger is in how quickly it's doubling – it's moving pretty fast."

Its efforts to repurpose current readings to track atypical illnesses "is 10 months of work that Daniel condensed into a week," Nehru said. "We're in a public health emergency; there's no dataset that's helping the country get ahead of COVID-19. Everything we've seen is reactionary; how do you stop a rise in cases in the next area if you don't know where to be looking? Our team did a tremendous amount of data science work on the back end to do this."

Kinsa Health not only sells its thermometers, but it distributes them to low-income families in federal programs and through sponsorship programs, such as one with Ochsner Health System in Louisiana. As its thermometers eventually are used by more consumers, it hopes to aggregate more data to enable more precise predictive services.

"The more sensors we have out in the field, the more geo-specific we can get with our insights," Doherty said. "Right now, we can track to a county level, but we hope to get to a more granular level, like ZIP codes, over time."