Crowdsourcing Data to Fight COVID-19

HowWeFeel appnew article in the Harvard Gazette outlines the findings of a recent study looking at information gathered by an app used by 500,000 people to log daily symptoms, health status, and exposures to COVID-19. The app, which addresses the need for increased data to track and predict the spread of the disease, is the first initiative from the How We Feel Project, a nonprofit created from a collaboration involving Biostats Professor Xihong Lin, Feng Zhang of the Broad Institute; Gary King, director of the Institute for Quantitative Social Science; and Pinterest CEO Ben Silbermann. The hope of study authors is that predictive models can soon be used to help overcome testing capacity limitations and identify disease hotspots.

As summarized in the Gazette, the study results confirm that Black and LatinX users, frontline health care workers, and essential workers have increased risk for infection than other groups, and that those groups also had a higher likelihood to be tested. Other findings were that household and community exposure were major factors in infection, and that the loss of taste and smell is increasingly indicated as the greatest predictor of a positive test. The study also highlighted some of the issues that have made the disease harder to contain; namely, the strong role of asymptomatic cases in disease transmission, and a lack of compliance in prevention measures including mask wearing and social distancing.

For more details, see the article and check out Lin’s Twitter thread summarizing the study.