Scanning for answers to a pandemic

From Bill Gates Blog. Thanks to Steve L. for sending this in.

In any fight, it’s important to know your enemy. Unfortunately, in our battle against COVID-19, there’s a lot that we still don’t know.

How many people are infected with the virus, including those without symptoms?

Is it seasonal or weather dependent?

And how will we know when it might be safe to get back to work and send kids back to school?

These are important questions. More testing, of course, will help us answer them. But with tests in short supply in many parts of the world, including the U.S., it is impossible to test everyone—at least for now.

That’s why I’m excited about a new disease surveillance program in the Seattle area to detect cases of COVID-19 and help guide public health responses. Not only will it help improve our understanding of the outbreak in Seattle, it will also provide valuable information about the virus for other communities around the world.

The greater Seattle Coronavirus Assessment Network—or SCAN—is a first-of-its-kind disease surveillance platform for COVID-19 that allows participants to use a self-swab test to collect their own nasal samples and send them to a lab without leaving home. As a surveillance program, SCAN’s goal isn’t to test every person or serve as a replacement for medical care. Instead, SCAN is testing a sample of people in the Seattle region, including those who are healthy as well as those who are feeling sick. The test results and other data (like a person’s age, gender, race, zip code, and any underlying health conditions) are used by researchers, data modelers, and public health officials to paint a clearer picture of how COVID-19 is moving through the community, who is at greatest risk, and whether physical distancing measures are working.

One of the biggest questions puzzling public health officials is exactly how many people are infected with the virus. Think about the pandemic like an iceberg, says Dr. Jay Shendure, scientific director of the Brotman Baty Institute, one of SCAN’s lead partners. With ongoing COVID-19 medical testing, which has been largely focused on people with symptoms, we have been able to see the tip of the iceberg. Just below the surface, however, there is the part of the iceberg we don’t see—the unknown number of people who are infected but experience mild symptoms or no symptoms at all. Dr. Shendure compares SCAN to “a set of sonar pings where we’re skimming over the water and pinging to see what lurks beneath.”

I want to be clear that SCAN does not replace the widespread testing that is still needed in communities. But it has the potential to become an important tool for health officials seeking insights about the spread and behavior of the virus. Early results from SCAN found many cases of COVID-19 in Seattle that might otherwise have gone undetected among individuals who had experienced some symptoms (fever, cough, or shortness of breath) but had not yet sought medical care. As SCAN gathers more test results in the weeks ahead, researchers expect the new data to provide a better sense of the number of infections and serve as one source to help answer other questions, like when physical distancing measures can be relaxed.

SCAN is a partnership between Public Health—Seattle & King County, the Brotman Baty Institute, University of Washington Medicine, Fred Hutchinson Cancer Research Center, and Seattle Children’s Hospital. It relies on data modeling support from the Institute for Disease Modeling (IDM) and receives support from my private office, Gates Ventures, and our foundation.

SCAN is an outgrowth of a research study started before COVID-19. It’s been clear for years that there was a lot the scientific community didn’t understand about respiratory viruses, such as how they spread through a community, and the best ways to stop them. So, in 2018, my office teamed up with the Brotman Baty Institute and other partners to launch a study of respiratory illnesses, including the seasonal flu. That effort, the Seattle Flu Study, aimed to recruit 10,000 volunteers in Seattle who showed cold symptoms to provide a simple nasal swab at kiosks set up in health centers and through at-home tests.

One of the early discoveries of the study was the impact of high-intensity physical distancing measures on reducing the flu. In the winter of 2019, a major snowstorm in Washington state led to week-long school and workplace closures. Analyzing the data from that flu season, researchers found that the snowstorm’s dramatic disruption of social contact led to a drop in the transmission of the flu and other respiratory illnesses.

The Flu Study team hoped that these and other findings would help researchers develop tools to curb and even prevent the spread of the flu—and maybe one day help public health officials prepare for a future pandemic.

That day arrived sooner than anyone ever imagined with COVID-19.

In late January 2020, the first confirmed case of the novel coronavirus appeared in the U.S., just outside Seattle. The patient had visited Wuhan, China, the origin of the outbreak. 

Several weeks later, the Seattle Flu Study team started picking up signs of the coronavirus’s genetic signature in their flu study survey samples. In February, they identified the first known case of transmission in the U.S.—a teenager living outside Seattle who had not traveled to China and had no link to anyone with the virus. The Flu Study team sequenced its genome, and quickly realized that the virus had been spreading undetected for weeks.

This finding and the discovery of dozens more coronavirus cases suddenly put the Seattle Flu Study at the center of the area’s response to the coronavirus. We were fortunate to have this existing surveillance platform and an experienced team to quickly shift their focus to the outbreak. In March, we formed the new SCAN partnership with King County’s public health agency to track the spread of COVID-19.

One of the innovations of SCAN is an easy-to-use, at-home, self-swab test kit. The key advantage of this at-home testing approach is that people don’t need to go to a clinic, where they risk exposing themselves or others to infection. People interested in participating in SCAN can enroll online, and, once approved, they get a test delivered directly to their home. After individuals complete the nasal swab test, a courier service picks it up from their home and returns it to SCAN for processing. Anyone who has a positive result gets contacted by a public health worker who provides guidance on how to care for themselves and their families. And all participants can check their results online.

SCAN is currently testing 300 people per day, but actively working to test more. Those test results are then analyzed by disease modelers to map virus transmission chains. By examining the genetic signature of an infection, they can determine whether it represents a new introduction to the region or is part of a local transmission. They can also use the data to estimate disease prevalence and build models to look at how the virus is responding to certain measures—like school closures and physical distancing. You can learn more about their work on Nextstrain and the Institute for Disease Modeling’s research site.

As the SCAN team collects more data, I’m looking forward to learning more about their insights into many of the questions we have about this pandemic and how we can prevent the next one.