Tech Tent: Do Covid apps work?

At the beginning of the coronavirus pandemic, governments around the world decided the smartphone could be a key weapon in their battle to stop the spread of Covid-19.

On this week's Tech Tent, we ask whether their various approaches, from Bluetooth contact-tracing apps to smartphone surveillance, have made a difference.

• Listen to the latest Tech Tent podcast on BBC Sounds

• Listen live every Friday at 08.00 GMT on the BBC World Service

The technology adopted has varied according to local cultures. South Korea used smartphone data along with information from credit card payments and CCTV to track the movements of people infected with the virus.

Victim of success

Singapore used what appeared to be a less invasive method in developing a Bluetooth contact-tracing app but it ended up collecting a lot of data for the government.

Many European countries ended up with decentralised contact-tracing apps which handed over very little data to government or health authorities.

This was the approach eventually chosen for the NHS Covid-19 app in England and Wales, after an initial trial of a centralised app proved both controversial and technically disappointing.

Susan Landau, professor of cyber-security at America's Tufts University, examines the various methods in her book People Count: Contact-Tracing Apps and Public Health. She tells Tech Tent how each has fared.

"South Korea has done quite well in controlling the disease. But one has to say that there are cultural aspects to this - as well as the technology, the willingness to wear masks."

That, she explains, would not work in the United States.

Singapore's centralised app has worked well, she says, because citizens have been required to use it in offices, shopping centres and schools. But they have also been forced to hand over lots of very sensitive data.

"It has been used for criminal investigations," she says. "If you're a journalist and people know proximity information, then they know who you've been talking to. And that, of course, can be really dangerous for human rights workers."

As for the decentralised apps, their effectiveness "has become more clear with time". Prof Landau points to a study in the journal Nature which showed that the NHS Covid-19 app had averted hundreds of thousands of cases of the virus.

But in the last week, the app seems to have become the victim of its own success. With cases of Covid rising rapidly, the number of alerts telling app users to go into isolation has soared.

With many businesses angry that employees are being sent home - in their view unnecessarily - by pings from the app, there is mounting pressure to change the way it works.

Politicians seem to be responding, briefing that its sensitivity may be tweaked, although at the same time the Department of Health says: "The app is doing exactly what it was designed to do."

The technology being used for contact-tracing apps is far from perfect - Prof Landau points out that Bluetooth doesn't know whether a contact takes place outside or inside, where the danger of infection is much higher.

But overall, automated contact tracing appears to be a useful addition to the old-fashioned manual variety - after all, people's recollections of who they were in contact with a few days ago and at what distance may also be unreliable.

Susan Landau points out that there is also an economic and social context to be considered: "For people who are bus drivers, restaurant workers, food service people, an exposure notification - where they have to stay home from work, they're not getting paid, they may, on the third exposure notification, lose their position - can be very expensive."

It turns out that contact-tracing smartphone apps have been an experiment not just in technology but in psychology, politics and economics.

But whatever their faults, it seems likely that they will remain a weapon in the public health armoury, ready to be deployed when the next pandemic comes along.