Oakland, Calif.-based Kaiser Permanente developed the "COVID Hotspotting Score" to use data from routine clinical care to predict the onset of a COVID-19 surge up to six weeks in advance, according to a July 26 study published in BMJ Open.
Vincent Liu, MD, a physician-scientist with Kaiser's research division, said in a July 27 news release that "shorter-term predictions — looking only 1 to 3 weeks out — left little time to respond adequately. By examining hundreds of millions of data points across our entire health system, we have now identified key leading indicators that forecast impending COVID-19 surges as much as 6 weeks before they hit our medical centers."
Six things to know:
1. Kaiser tested dozens of data elements from its EHR to look at indicators from seasonal flu surges as far back as 2015.
2. The tool Kaiser developed includes four major variables: cough and cold calls, subject headers from patients' emails, COVID-19 positivity rates and the current COVID-19 hospital census. It also includes six minor variables: COVID-19-related calls to the call center, routine and urgent clinic visits with respiratory infections, clinic and urgent care COVID-19 visits, and respiratory virus tests ordered.
3. Researchers said the tool could also predict when a COVID-19 surge was waning.
4. Health systems can use the tool to prepare hospital space and supplies in anticipation of a jump in the number of incoming patients. The developers wanted to make sure the tool could be duplicated by other health systems and would be freely available.
5. The tool launched in June 2020 and was evaluated against actual COVID-19 patients through Dec. 31. Researchers found the accuracy very strong. It peaked at predicting 28-35 days into the future, but correlation remained when tested out to six weeks, the release said.
6. The tool was tested at 20 Kaiser Permanente hospitals and was strong at each hospital, despite varying influxes of COVID-19 patients, the release said.