Coronavirus contact tracing apps flop in France and Australia

Coronavirus contact tracing apps are not all they’ve cracked up to be and those being used in France and Australia have flopped, experts warn. 

In France, two million people downloaded the StopCovid app but only 14 people received alerts that they were at risk in three weeks of the app being in operation. 

Australia faced the same problem, with six million people downloading COVIDSafe but in Victoria, a state currently in the grip of a new outbreak, it produced no useful information that the authorities couldn’t already get from human contact tracers.

The smartphone apps are intended to keep track of who people have been in contact with in case one of them becomes infected with the virus.

In Britain an app was for months hailed as a crucial to stamping out the coronavirus in the population, but it soon become clear the plan wouldn’t work when officials realised their app didn’t work on iPhones.

Now other countries’ major teething problems suggest a successful app may be further away than people thought.

Software developers have been so wary about scaring people and sending out too many notifications that they’ve gone too far the other way, scientists say. 

People have complained of the apps – which rely on Bluetooth connections between people’s smartphones – draining their batteries.  

As a result, many are now uninstalling the apps because they create more problems than they do solutions, MIT Technology Review reported.

France’s app, StopCovid, has reportedly been downloaded two million times – about three per cent of the country’s population – but only alerted 14 people to a coronavirus risk in its first three weeks of operation

Australia’s CovidSafe app was downloaded by around a quarter of the population there but did not prove at all useful to authorities in Victoria, whose human contact tracers produced the same amount of information, according to Gizmodo

The reason for disappointing numbers of people being warned about possible infection by the app may just be because not enough people use it, and people still aren’t having many close contacts. 

One expert said that if case numbers are low and people are following rules and social distancing, there shouldn’t be many notifications at all. 

‘It’s simple math,’ Cambridge University’s Professor Jon Crowcroft told MIT Technology Review.

‘If one per cent of people have Covid-19 and they are all tested, and only 1 per cent of people run the app, you have a one in 10,000 chance of having both the tested person and the exposed person having the app.

‘So your notification rate will be 10,000 times lower than the case rate.’

Uptake of the apps appears to be higher than Professor Crowcroft suggests – around 25 per cent in Australia, according to Gizmodo, but three per cent in France.

But another reason the apps aren’t notifying people could be that developers have programmed them to act cautiously.

Huge numbers of warnings being fired out every day might cause mass panic, so apps have reportedly been programmed to pick only the highest risk contacts. 

But one of the biggest problems with the apps is the fact that the Bluetooth they completely depend on just doesn’t work in the way it should.

Health officials around the world had hoped that people would simply be able to download the app and it would do all the work in the background with no effort, automatically connecting to and logging every phone within two metres of someone.

Officials in the UK abandoned the NHS’s attempt at making its own app in June when they realised it didn’t work on iPhones (Pictured: The app in development stages)

But in reality, the Bluetooth connections seem to drop out when someone locks their phone and the app fades into the background.

Google and Apple, who have developed app systems of their own, have limits on how much Bluetooth activity can continue when someone isn’t using an app.

France and Australia have both attempted to make their own software and apparently been hit by the same issue Britain faced, in which trials showed the app could only detect four per cent of connections when it was run on an iPhone.

The Australian version has a higher success rate but only makes around 25 per cent of connections successfully, The Guardian reported.

One anonymous researcher told the MITTR: ‘This effectively means for a contact tracing app to work without using their system, a user has to walk around like a Pokemon Go player, with their phone out, the app open, and not use their phone for anything else.’       

Software developed by Apple and Google is now forming the basis for many country’s apps around the world.

Officials in the UK, who abandoned their own attempt to make an app last month, said the one made by Apple and Google wasn’t yet good enough to use either.

Health Secretary Matt Hancock and the NHS Test and Trace chief, Dido Harding, admitted on June 18 that they were scrapping plans for an NHS-made app.  

The health chiefs said the app, which was originally promised for mid-May and the NHS spent months developing, was unable to spot 25 per cent of nearby Android users and a staggering 96 per cent of iPhones in a trial on the Isle of Wight.

Meanwhile, the Apple and Google technology could spot 99 per cent of close contacts using any type of smartphone – but it could not currently tell how far away they are, officials claimed. 

Health bosses said the Apple/Google technology couldn’t differentiate someone 3m (9’8′) away with their phone in their hand from someone 1m (3’3′) away with it in their pocket; 

The leaders of Britain’s test and trace system said neither app is fit for purpose and Mr Hancock appeared to point the finger at Apple for the failure, saying: ‘Our app won’t work because Apple won’t change their system’. 

Developers in the NHS are now said to be working alongside the tech giants to try and roll their detection software and the NHS app’s distance-measuring ability – which they said was significantly better – together to make a hybrid app that actually works. 

British officials have refused to put a time scale on when the UK’s app might be ready – one politician has suggested it will be winter at the earliest.

But they insist it is still ‘urgent and important’ and will be able to do jobs that human employees can’t.

Simon Thompson, managing director of the NHS Covid-19 App, told a House of Lords committee on Monday: ‘I think that when we have a look at the particular benefits that the app can bring to the programme there are three areas that we’ve really focused on. 

‘One is the speed, so the ability to communicate with the user in minutes.

‘The second one is precision, which is the ability to have confidence around the distance and time. Our sense is it needs to be a really good standard but we believe it will definitely be better than what a human could manage. 

‘And in terms of reach – the ability to know people that you have met who you did not know you had met – we believe that the app can make real inroads there.’