Efforts to get a handle on the spread of COVID-19 in the U.S. have been hampered by the lack of widespread testing. Most public health experts, in fact, say that absent the emergence of a treatment, vaccine, or herd immunity, having data from a robust national program should be a precondition to reopening the nation.
Trying to begin the task of gathering data on the spread of the disease is the goal of a new crowdsourcing app, How We Feel, which launched this month in a collaboration involving Pinterest CEO Ben Silbermann and researchers from Harvard and other institutions, including Feng Zhang of the Broad Institute and MIT, Xihong Lin, professor of biostatistics at the Harvard T.H. Chan School of Public Health and a professor of statistics at the Faculty of Arts and Sciences, and Gary King, director of the Institute for Quantitative Social Science (IQSS) at Harvard.
The idea for the nonprofit came from Silbermann and Zhang, best known for pioneering the CRISPR gene-editing technology. The two longtime friends met in high school in Iowa, and after the outbreak began they started talking about whether their expertise in technology and biology, respectively, could be leveraged to find a way to make up for the lack of reliable testing data.
“Since high school, my friend Feng Zhang and I have been talking about the enormous value of connecting citizens to the global health and research community,” Silbermann said in a statement. “When we saw how quickly COVID-19 was spreading, it felt like a critical moment to finally build that bridge between citizens and scientists that we’ve always wanted. How We Feel is an important first step.”
This is how it works. Users can download the free app for Android or iOS. They are then asked to anonymously enter their ZIP code, sex, age, race and other demographic information, health conditions, pertinent lifestyle data (such as whether they smoke), any coronavirus test results, and, most importantly, how they are feeling on a daily basis. If participants report they aren’t feeling well, it prompts more specific questions about symptoms.
Researchers then analyze that data using novel statistical methods developed at Harvard to pinpoint hotspots and predict the percentage of people who have the disease in different regions of the U.S. The information could shed light on key blind spots if enough people sign up and use it daily.
“Although symptoms alone cannot diagnose the disease in any one person, our statistical methods make it possible to accurately estimate the prevalence of the disease in different populations, locations, and other subgroups,” said King, who is also the Albert J. Weatherhead III University Professor.