Stanford Medicine is using data and digital tools to predict the next COVID-19 surge

Smartphone apps
Stanford Medicine launched a national daily health survey that uses a chatbot to ask users about any recent COVID-19-like symptoms they’ve experienced to help track a surge in COVID-19 cases. ( Getty/marchmeena29)

Stanford Medicine scientists are working to create an "early warning system" that predicts the next surge in COVID-19 cases.

Using a daily survey that tracks the occurrence of possible COVID-19 symptoms in communities could help raise the alarm far sooner and allow hospitals and healthcare workers to prepare, Lawrence “Rusty” Hofmann, M.D., a professor of radiology at Stanford School of Medicine, told FierceHealthcare.

Hofmann, who is chief of interventional radiology at Stanford Hospital, worked with a team of epidemiologists and data scientists to develop the national daily health survey to help learn and predict which geographical areas will be most impacted by coronavirus based on how survey takers are feeling.

The goal is to use survey responses to provide data that could flag communities at risk for a surge in cases of COVID-19 before they reach the hospital, Lawrence said.

“I was sitting in my office two and a half weeks ago, frustrated that we’d only tested 35,000 Americans,” he said. “Now we’ve tested just over a million Americans, which may sound like a lot, but that’s still less than a third of 1% of the U.S. population. It made me realize that what we’re doing right now just isn’t working.”

He added, "We don’t know what we’re fighting and we can’t fight it effectively until we quantify it."

RELATED: Apple, Google team up on COVID-19 contact tracing via smartphone apps

Given the nine- to 10-day delay between onset of symptoms and hospitalization and the 20% hospitalization rate of patients, tools like this will be necessary to truly track and fight the spread, according to Stanford Medicine scientists.

"I was on call last night and everyone was on eggshells, wondering, 'When is the surge coming? When is it going to hit us? We don’t know and people are running on empty now," Hofmann said. 

The survey, which is voluntary and open to the public, launched April 2 and has enrolled more than 125,000 individuals. The survey uses a chatbot to ask users about any recent COVID-19-like symptoms they’ve experienced and any contact with people who have COVID-19.

If enough people take the survey, it could act as an early warning system for regions of the U.S. with large populations of undiagnosed individuals, Hofmann said.

“My hope is that people see taking this survey as their civic duty and as a way to be involved in fighting COVID-19," he said. 

Leslie Haas, digital health strategy manager at Stanford Health Care, said Stanford is taking a nationwide approach, with the hopes of expanding internationally in the near future. 

"We have visibility into Stanford’s test results, including demographic and condition-specific detail; this will enable us to make better predictions from the national data collected," she said.

The survey’s data, which Hofmann and his team analyze as it comes in, is coupled with local testing results to better predict the true prevalence and severity of COVID-19.  

RELATED: Apple teams up with CDC and White House to roll out COVID mobile app and website

The data will be de-identified and shared broadly with public health officials and healthcare professionals to help inform decision-making and hospital preparedness, according to Stanford Medicine. That information could give hospitals and health systems extra time that's critical for finding bed spaces, organizing staff and securing equipment necessary to care for incoming patients.

Other institutions like Mount Sinai Health System have built apps for the virus in local communities. Friday, Google and Apple announced a joint project to create contact tracing technology. The two tech giants will be able to notify people via smartphone if they've come into contact with someone with the coronavirus.

Stanford's web-based survey is a frictionless process, Hofmann noted. "If you build an app, you have to build one for both Android and iPhone and the user has to download it and log in," he said. 

“This survey is crucial for monitoring symptomatology and understanding what’s happening on a continual basis, rather than a single point in time, like the swab test,” Hofmann said. “We don’t want to be caught flat-footed, because this fight isn’t over until we have a vaccine.”

App to connect first responders to COVID-19 testing 

Stanford Medicine also rolled out a new app, called the First Responder COVID-19 Guide, which was built with the support of Apple to help first responders screen their symptoms and, if needed, schedule a testing appointment.

The app uses Apple’s ResearchKit and CareKit frameworks and connects first responders with high priority drive-thru COVID-19 testing sites in California's Santa Clara and San Mateo counties.

RELATED: Gottlieb, Mostashari propose national COVID-19 surveillance system

Stanford plans to expand high-priority testing to other front-line essential service workers such as grocery store clerks and public service personnel.

"With these apps, we’re expanding the reach of our expertise to provide answers for people who are busy serving others during this crisis. It’s our hope that this technology will ease some of the burden for people on the front lines, and will help inform those who seek a reliable source on COVID-19," said Priya Singh, chief strategy officer and senior associate dean at Stanford Medicine, in a statement.

Stanford Health Care began offering drive-thru COVID-19 testing in early March and has expanded testing to seven locations. To date, more than 3,000 patients have been tested for the coronavirus through Stanford Health Care’s outpatient testing facilities, and the health care system’s sites have the capacity for 2,500 patients a day, officials said.