Spikes in fevers across the US predicted deaths from the novel coronavirus in most states just days beforehand.
Kinsa Health, a medical technology company based in San Francisco, has been tracking daily fever readings using data from smart thermometers connected to the Internet.
Of the 47 states that have experienced deaths from COVID-19, the disease caused by the virus, the company found unusually high rates of fevers at least five days prior to the first fatality in 41 states.
In fact, on average, Kinsa saw illness anomalies in US states 14 days before they reported their first death from the virus.
Data from daily temperature readings shows that spikes in fevers predicted the first fatalities from coronavirus in 41 states
Increases in fevers were seen at least five days before the first death, and an average of 14 days out (above)
The data comes from Kinsa Health, which has been collecting at least 162,000 daily temperature readings from smart thermometers. Pictured: EMTs move a coronavirus patient from Mt Sinai Morningside Hospital, April 9
Kinsa has distributed more than one million thermometers and gets about 162,000 temperature readings a day.
The thermometers upload the temperatures to a database (similar to Apple iCloud) and users can add other symptoms into an app.
A new graphic shows that illness anomalies were observed in at least 41 states before its first coronavirus death.
In New York, the epicenter of the outbreak, Kinsa’s smart thermometer data showed a spike in fevers about 10 days before the first fatality on March 14.
And in West Virginia, the last state to report any cases of the virus, anomalies were seen 25 days before the first death on March 29.
Only six states had no anomalies prior to the first deaths and, in four of these states – Minnesota, Wisconsin, North Dakota and South Dakota – no anomalies were seen at all.
In the US, there are more than 588,000 confirmed cases of the virus and more than 23,600 deaths
‘This means that Kinsa’s atypical illness signal can serve as an early detection system in future outbreak monitoring,’ the company wrote on its website.
‘Though a work in progress, Kinsa’s atypical illness signal has effectively identified COVID-19 outbreak epicenters in advance of reported deaths.’
Prior to using the tool to track COVID-19, Kinsa’s tool has mostly been used to track where seasonal flu outbreaks are occurring.
Traditionally, the company’s predictions have been two or three weeks ahead of those compiled by the Centers for Disease Control and Prevention.
With the coronavirus pandemic a new feature has been added to the map, which the company calls ‘atypical’ illnesses
This tracks illnesses that don’t match up with typical flu patterns and are likely due to the novel coronavirus.
Influenza-like illness levels are depicted in orange and red and where they are expected to be depicted in blue. Almost every state has seen decreases.
Additionally, earlier this week, fevers across the nation were dropping except for a county here and there.
In fact, fevers are about 92 percent lower than expected at this time of year. Only about 0.17 percent have temperatures above 99F.
‘Social distancing is slowing the spread of feverish illnesses across the country,’ Kinsa wrote on its website on April 1.
‘Note: This does not mean that COVID-19 cases are declining. In fact, we expect to see reported cases continue to surge in the near term.’
In the US, there are more than 588,000 confirmed cases of the virus and more than 23,600 deaths.