Coronavirus symptom checker app predicts infections with nearly 80% accuracy

Boston-based Massachusetts General Hospital and London-based King's College researchers used a COVID-19 symptom-checker app to predict with almost 80 percent accuracy whether a person is likely to have contracted the virus based on a combination of four symptoms, according to a study published in Nature.

For the study, the research team examined whether loss of smell and taste is specific to COVID-19 in 2.6 million individuals who used health science company Zoe Global's app-based symptom checker. The app collects data from both symptomatic and asymptomatic individuals and tracks in real time how COVID-19 progresses by recording symptoms on a daily basis. Participants are not required to share their names or other personal information, only their ZIP codes.

The researchers analyzed data of the 2.6 million app users in the U.S., U.K., and Sweden from March 24 to April 21. Results of the study showed that the loss of taste and smell was the top predictor of whether an individual would get sick with COVID-19; the researcher's mathematical model was able to predict with almost 80 percent accuracy whether someone was likely to have COVID-19 based on their age, sex and a combination of four symptoms: loss of taste or smell, persistent cough, fatigue and loss of appetite, according to The New York Times.

Nearly 15,000 people, or about two-thirds of individuals who later tested positive for COVID-19 had self-reported a loss of taste and smell, while symptoms such as fever and cough ranked fourth and fifth on the list of indicators of infection, according to the report. The study's lead investigator, Andrew Chan, MD, clinical and translational epidemiology chief of MGH, told the NYT that a surveillance app could help health authorities identify individuals at the early stage of COVID-19 who are unaware they have it.

"At the moment, we're mostly gathering data on the tip of the iceberg from those who are really sick and show up at the hospital. But there is a huge iceberg below of people with mild symptoms who we know are major culprits for community spread," Dr. Chan said. "We have no ability to track these people at home and that's a real problem."

More articles on health IT:
Nebraska Medical Center employee wrongly accessed EHR, affecting 1,311 patients
The hospital room of the future: Dr. Alistair Erskine's 5 predictions
AWS' Amazon Kendra now widely available to make internal data more searchable

© Copyright ASC COMMUNICATIONS 2020. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.


Featured Webinars

Featured Whitepapers