top of page

Contact-tracing app, a future with potential and problems

To try to prevent another wave of Covid-19 as the lockdown measures are about to be relaxed, healthcare authorities need to identify people who got the virus, who haven’t, and the asymptomatic ones, before they spread the disease and keep the virus in circulation.

Epidemiologists have come up with new techniques to do this.

The form of manual contact tracing where medical workers talk to patients to find out who they might have been in contact with to reach them out and advise them to self-isolate has huge limitations.

This is because the tracing is based on patients’ memory. If they don’t remember who they got in contact with, then this method has very limited accuracy.

In addition to this, it takes a few but crucial days for the authorities to get in touch with the cycle of people the infected patient got in contact with.

To overcome this problem, digital contact-tracing using smartphone apps has been advanced. This method could greatly reduce the risk of another infection wave while lifting lockdowns and resuming commercial activities.

The technology is thought to automatically record who had been in contact with, to then alert people if an individual turns out to be positive to the virus.

Image credit Digital Strategy Consulting

Undoubtedly, these things are much simpler in theory. The biggest issue for digital contact tracing is how to get data on who people have been in contact with.

Some suggestions based on the idea to use geolocation data from smartphones to trace patients have been internationally critiqued by more than 300 scholars.

Academics warned that the GPS data lacks enough accurateness to establish whether people came close enough to each other to transmit the virus or not.

This could lead to a false number of positive individuals which would undermine the hard efforts done until now and the gradual resume to get back to normality. The data would also be unattainable on underground transport, which is a source of transmissions because of the passengers’ closeness to each other.

The most important concern regarding this potential tracing app is that geolocation-based tracking is an invasive route in terms of personal privacy.

Authorities would be able to follow where people are, and at what time, but also who people may have been in contact with.

Even if the public would agree with such a system for the sake of emergency, the potential for it to be preserved and exploited caused concern.

To overcome this issue, some experts encouraged the creation of apps which would use low-energy Bluetooth waves to only detect which devices the smartphones had come near to, and not their geolocation.

But even among this proposal, some mechanisms for surveillance remain, as some Bluetooth could record the networks of whom everybody has physically met.

Image credit Council of Europe

Also, what if wicked actors, such as hackers, were to access this mechanism? They could use it to spy on people’s activities and relationships.

Scientists state that a decentralised system for contact-tracing apps is possible without interference with citizens’ privacy, in which the list of people each smartphone has come in contact with is stored only on the device and kept inaccessible to third parties.

However, is it technically possible?

We're still in the early days of learning about how different countries will be implementing contact-tracing apps, but a clear division between different approaches is beginning to show.

Some governments are thought to favour a centralised system, with one international proposal known as PEPP-PT, or “Pan-European Privacy-Preserving Proximity Tracing” being designed by technologists and scientists in the EU.

But they are being opposed by a coalition of privacy and security experts in a project titled DP-3T, which stands for "Decentralised Privacy-Preserving Proximity Tracing", which has been officially backed by Estonia, Austria and Switzerland.

But PEPP-PT's proponents are still facing a potentially critical technical challenge to their centralised design from two of the largest technology companies in the world: Apple and Google.

Apple and Google are collaborating to allow mobile devices to use Bluetooth in the background and register when they come within proximity of another mobile phone.

The iOS and Android mobile operating systems are run on 99% of the world's smartphones. The companies' technical designs will have a fundamental say in how contact-tracing apps work.

These changes to the operating systems are essential for phones to be able to ping each other using Bluetooth and detect when they are near to another device, even while locked and the app isn't running.

However, some warned that governments which are seeking to build centralised systems are pressuring Apple and Google to "open up their systems to enable them to capture more data".

Apple, which markets itself based on the iPhone's privacy protections, has designed the Bluetooth technology so that contact-tracing apps can only take place as a background process if the data the apps generate isn't sent off the device.

COVID-19 contact tracing apps could be vital in controlling the pandemic. Image credit Ina Fassbender /AFP

This has prompted complaints from government officials, including France's digital minister Cedric O who warned that the security measures mean the contact-tracing app which France is developing wouldn't be able to work on iPhones.

"We're asking Apple to lift the technical hurdle to allow us to develop a sovereign European health solution that will be tied our health system," Mr. O told Bloomberg News.

He said French government ministers had discussed their concerns with Apple but hadn't made any progress.

In the UK, the NHS has confirmed it will also be using a centralised system, different from the DP-3T designs.

Speaking to MPs, the chief executive of the health service's innovation arm NHSx, Matthew Gould, explained that the UK's approach had "a measure of centralisation" as it would be the NHS which sends the alerts, rather than them happening on a phone-to-phone basis.

Despite the French desire to develop a "sovereign European health solution" for contact tracing, it may be that European data protection regulations end up supporting Apple and Google's decentralised model.

New guidelines published by the European Data Protection Board on the use of location data and contact-tracing tools in the context of the COVID-19 pandemic said that a decentralised contact-tracing system was more in line with laws about minimising the amount of data which is collected.

And how to measure distance?

While the privacy issues are well-understood - even if the debate hasn't settled - there is a significant engineering challenge to determine distance using Bluetooth.

Most of the proposed apps are using something called Received Signal Strength Indication (RSSI) as a proxy for measuring the distance between the Bluetooth transmitters and receivers, but this is "notoriously inaccurate" according to Professor Woodward.

This inaccuracy means that some apps which aren't aware of the potential for miscalculating distance will generate "a great many false positives".

But Professor Woodward argued that some of the better apps - he noted DP-3T among them - are logging not just the other Bluetooth tokens, but also how long the phones are within range of each other.

For instance, if the phone detects another device but then a few seconds later doesn't detect any more contact, the app can consider them at lower risk than if they were "close" to someone for minutes at a time, Professor Woodward said.

Professor Fraser, from the University of Oxford, told the Andrew Marr Show: "Obviously you will have to have your phone with you, but your phone will then build a memory of these anonymous IDs of the other phones you come into contact with."

This contact needs to be long enough for there to be a risk of transmission, he explained, and this measure is being developed by engineers in the UK to make sure it is recording "the contacts which are the most likely to result in transmission".

Finally, will this work?

Professor Fraser said that for the app to prevent a resurgence of the epidemic, about 60% of the population would need to be using it.

This has been one of the arguments in favour of a decentralised system. The fear of being spied on could damage the public's trust in the voluntary app and suppress the numbers of people using it, therefore undermining its effectiveness.

Prof Fraser told MPs: "I think this needs to become part of the core message, that as the country looks to reduce the restrictions that it's under at the moment, as the government faces these difficult choices, that the way we can manage that safely is being confident we can rapidly detect and isolate people who come into contact with new COVID cases."

Another major question, such as how to deal with the large numbers of over-55s without smartphones, an especially vulnerable group, is also yet to be answered.

And one of the most significant challenges for the app will be how it handles people falsely claiming to have contracted the virus to send false alerts to others.

Apple and Google say they are working with public health authorities to provide the validation for any diagnoses.

But for many people getting a diagnosis is simply not possible, and though the testing capacities for many countries are increasing, they remain extremely limited.

With so few cases being flagged up by official testing systems, Apple and Google say that public authorities may need to use another method to identify cases.

But what that other method may be, the technology giants have left up to the public authorities.

Source: Sky News


bottom of page