Our coronavirus symptom-tracking app has been used by millions — here’s what we’re learning

The global COVID-19 pandemic caused by the novel SARS-CoV-2 coronavirus is a public health emergency. In just under three months, we have gone from distant reports of a new disease emerging in China to seeing cities locked down and health services brought to breaking point.

Politicians and healthcare leaders are in the unenviable position of having to make impossible decisions about how best to tackle this fast-moving situation. Yet they are missing the one thing that would help them the most: data. Waiting for data on people going to intensive-care units and sometimes sadly dying is like being in a war waiting for bombs to fall without radar.

One of the key pieces of information that we are lacking is a handle on how many people in the UK are infected with coronavirus and where they are living. While some patients who seek medical care or are hospitalized are being tested, there are many more people – possibly millions – with mild or moderate COVID-19 symptoms who are not. And despite the promising headlines, it is unclear how soon we will have antibody tests to see who has already had the disease and is now recovered.

Read: [This startup made a coronavirus knowledge graph to help doctors with diagnosis]

A related problem is the symptoms themselves. The symptoms associated with COVID-19 include fever, persistent coughing, which are the classical symptoms, and also tiredness, breathlessness, loss of smell or taste, hoarseness, muscle aches, chest tightness or pain, abdominal pain, diarrhea, drowsiness and confusion, and loss of appetite.

But many people only experience a selection of mild symptoms or no symptoms at all. So how are we supposed to tell who is infected with coronavirus and who is suffering from a seasonal sniffle or allergy? Some estimates suggest that 5-10% of us may have a fever at this time of year. And our failure to know the difference may lead people to unnecessarily self-isolate when they aren’t infected or inadvertently go out and spread the disease when they are.

Although the elderly and those with certain health conditions are more likely to become seriously ill with COVID-19, there are also plenty of reports of hospitalizations and even deaths in seemingly young, otherwise healthy people. There are various theories as to why some people are more susceptible than others, such as smoking or vaping, but we still don’t understand why the virus affects some people much more than others.

Symptom-monitoring app

Last week, my King’s College London colleagues and I had an idea for how to provide the missing information to answer these tricky questions. We realized that we could use our Twins UK cohort – a group of 15,000 identical and non-identical twins – to understand more about how COVID-19 affects different people. We already have detailed health data about thousands of these twins, including genetic and immune profiling, their medical history and lifestyle, and their microbiome – any or all of which could affect their response to the virus.

Rather than sending out thousands of online or paper questionnaires, we teamed up with health data science company ZOE to develop a simple symptom-monitoring app called COVIDradar. The app was made from scratch in about four days and would normally take four months. Volunteer citizen scientists use it to report their health status daily and note the appearance of any new symptoms. Once we realized that there was nothing similar available in the UK to monitor symptoms on a population-wide level, we decided to make the app freely available to all.

The data from the app is being made available every day to policymakers, NHS services, and academic researchers on a strictly non-commercial basis. As well as helping us understand more about the symptoms of coronavirus and the people who are most at risk of becoming seriously ill, this information will also reveal how fast the disease is spreading and identify geographical hotspots.

By getting real-time updates, decision-makers can plan and allocate limited NHS resources more efficiently and monitor the impact of policy measures, such as social distancing. The data can also help to highlight who needs testing for current or past coronavirus infection.

The response has been staggering. So far nearly two million people have downloaded the COVIDradar app and started to track their daily health, and we hope that many more will join them. The app will be in the US imminently and we hope in other countries, too, if we get support.

So far, we have already shared data from the first million participants with the NHS, revealing a wide distribution of reported symptoms all over the UK (see map), and we’ll be providing regular updates through the COVIDradar website.

We’re going to need all the help we can get over the coming weeks and months as we battle this new enemy that threatens our health and economy. While our heroic NHS staff and key workers keep the country running, the rest of us can join the fight from the safety of our homes, armed with nothing more than a smartphone. Let’s spread the app faster than the virus.

To download the app, go to COVIDradar.org or covid.joinzoe.com and follow the links to the Apple App Store or Google Play.

This article is republished from The Conversation by Tim Spector, Professor of Genetic Epidemiology, King’s College London under a Creative Commons license. Read the original article.