Latest
Latest
4m agoAsk Rocio: How can I make life after death less stressful?
Latest
3h agoTaxpayer faces Thames Water bailout as boss warns of 40% bill rise
Latest
3h agoMajor Tory donor and party treasurer Mansour knighted in surprise honours list

Covid symptom tracker app suggests 350,000 could have untested coronavirus

The self-reporting app has provided an insight into lesser known symptoms of the disease

An app that tracks self-reported symptoms of coronavirus among the general population suggests that there are more than 350,000 people in the UK who would be likely to test positive for Covid-19.

The Covid Symptom Tracker app has provided “unprecedented amounts of data” according to researchers at King’s College London (KCL), who are working on the app with healthcare data science start-up Zoe.

Live data from the app projects that as many as 354,690 people across the UK could currently have symptoms of the disease that would probably result in a positive test – down from a projected peak of 2,127,663 cases.

This is more than double the official figure of just over 140,000 positive tests in the UK, recorded by the Department of Health and Social Care, as testing is not routinely offered to people experiencing symptoms that do not require hospitalisation.

The app has been downloaded by more than 2.4 million people, with users asked to self-report daily on what, if any, symptoms they are experiencing.

The Covid Symptom Tracker app has provided "unprecedented amounts of data"
The Covid Symptom Tracker app has provided “unprecedented amounts of data”

Researchers at KCL have been analysing the “real-time” data it provides to track trends of the disease across the country – and believe it could play a role in helping to ease the lockdown.

Head of the School of Biomedical Engineering & Imaging Sciences at KCL, Professor Sebastien Ourselin, said: “Through the Zoe app, we are gathering a wide amount of data that is key to better understand the disease and how to better manage it.”

New symptoms

While fever and persistent cough are the best-known symptoms of the disease, Head of Genetic Epidemiology at KCL, Professor Tim Proctor, told Sky News that the app was also useful for flagging up other predictive symptoms of coronavirus.

The loss of taste and smell was one particularly notable symptom that can be a unique identifier for coronavirus.

He told Sky News: “Around 12 per cent of people who are displaying symptoms have this very strange symptom and it comes on very suddenly.

“There’s not many other things that cause it and so it’s highly predictive that you’ve been infected and at the moment, the official advice is still not to take it that seriously if you don’t have a fever.

“Yet many people are in that position – they have mild disease, they’ve lost their taste and they should really be isolating from other people.”

Severe fatigue is another symptom that could be indicative of a positive test, he said.

The app asks people to provide information about their symptoms (Photo by Cindy Ord/Getty Images)
The app asks people to provide information about their symptoms (Photo by Cindy Ord/Getty Images)

‘A radar system’

The app was providing data faster than hospital records and is helping to track the patterns and spread of the disease in “real-time”, he explained.

“Many new symptoms are coming to light and not being reflected in the hospital statistics and the death records,” Prof Spector said.

“Because we’re seeing things at least a week, maybe 12 days, before they get into the hospital and so we’re able to track these trends.”

He said the app could act as a “radar system” for tracking Covid-19, with “millions” submitting health reports every day, leading to an “unprecedented amount of data”.

As such, it could be a useful tool for quickly noticing “hotspots” for the disease across the country, helping to find and deal with new clusters of people with symptoms when the lockdown is eased, the professor said.

There are limitations, however, as the current users of the app are “disproportionately young, female and of a higher socio-economic status than the general population.”

Prof Spector is keen for a bigger and more diverse range of people to use the app to report their daily health status and symptoms.

“The more of the public that also use the app, the better the real-time data we will have to combat the outbreak in this country,” he said.

How the app works

The app asks users to submit basic health data, risk factors and their location when signing up and ensures they give consent for their data to be shared with the NHS.

They are then asked to answer health questions on a daily basis, regardless of if they feel well or ill.

Researchers are able to predict if users are likely to test positive for coronavirus if they took a swab test, based on the data they submit.

They also use statistical tests to analyse which symptoms are likely to be linked to a positive test.

Researchers can also extrapolate this data to make predictions about UK-wide coronavirus figures and spread, as well as making global predictions.

In future, researchers at KCL hope the data can be used to make predictions about which symptoms and factors are more likely to lead to someone being hospitalised with Covid-19.

Most Read By Subscribers