One of the best weapons we have to throw against Covid-19 may be found in our pockets.

That's according to researchers who say digital data from smartphones offer a powerful way to ramp up contact tracing against a virus whose rate of spread outpaces our painstaking manual efforts.

Experts have told the Herald that Kiwis should prepare for discussions about what trade-offs they're willing to make about digital privacy – and suggested some solutions already in use around the world that addressed some of those worries.

"We need contact tracing to save lives and using digital data could make that much more effective," said Dr Andrew Chen, a research fellow at the university's new think tank, Koi Tū: The Centre for Informed Futures.


"But we also need to protect people's privacy and minimise rights abuses that could have serious consequences."

How an app could work

Chen singled out one app called TraceTogether, which has been used in Singapore since March 20.

People installed it on their phones with Bluetooth enabled. When they were physically close to someone else with the app, the phones exchanged Bluetooth signals and any encounters were logged in the app.

It took just several seconds for the exchange – short enough to capture most interactions but long enough to ignore spurious connections.

Anonymous IDs were used so that phone numbers weren't exposed.

As Bluetooth worked at relatively short-distance, it provided a good proxy for physical proximity and was more accurate than GPS or cellphone-signal methods.

It could also help distinguish between people who had been "close contacts", as opposed to those who were "casual contacts".

Location wasn't necessary, because contact tracing relied mostly on connections between people.


The data was stored on the phone in encrypted form, and could only be sent to the Ministry of Health if the user authorised it after they'd been diagnosed with Covid-19.

Chen said this methodology was promising because it took an opt-in approach: users chose to use the app and knowingly took part in contact tracing.

"More than 600,000 Singaporeans enrolled in a few days with the app seen as a way to protect themselves and to help protect those around them."

Germany now planned to launch such an app within weeks.

Chen said the solution also gave users a sense of agency that was lacking from options that relied on harvesting cellphone network data.

Instead, users actively participated in the system, knew that it was happening, and could feel they were contributing toward a solution.

The Government only tracked consenting people who needed to be tracked, rather than tracking everyone.

The design of this system showed that it was possible to achieve a similar outcome to tracking cellphone location data with fewer implications for privacy.

"Importantly, when a promising intervention meets an established human right, rather than charging ahead anyway, we should consider other ideas that might make the balance easier to find."

UK research shows why app needed

Overnight, researchers at Oxford University published a paper that looked at why such technology was urgently needed.

They analysed key parameters of the epidemic spread, concluding that it was simply too fast to be contained by manual contact tracing.

"Our analysis suggests that about half of transmissions occur in the early phase of the infection, before you show any symptoms of infection," said Professor Christophe Fraser, of Oxford's Big Data Institute.

"Our mathematical models also highlight that traditional public health contact tracing approaches provide incomplete data and cannot keep up with the pace of this pandemic."

But it could be controlled, they said, if the process was faster, more efficient and happened at scale.

They proposed a contact-tracing app which built a memory of proximity contacts and immediately notified contacts of positive cases – something which could achieve epidemic control, if used by enough people.

By targeting recommendations to only those at risk, they said, epidemics could be contained without need for mass lockdowns.

"The mobile app concept we've mathematically modelled is simple and doesn't need to track your location," said study co-author Dr David Bonsall.

"It uses a low-energy version of Bluetooth to log a memory of all the app users with whom you have come into close proximity over the last few days.

"If you then become infected, these people are alerted instantly and anonymously, and advised to go home and self-isolate.

"If app users decide to share additional data, they could support health services to identify trends and target interventions to reach those most in need."

The researchers argued that a mobile app could reduce transmission at any stage of the epidemic, in countries or regions where the epidemic was just emerging, at the peak of the epidemic, or to support a safe transition out of restricted movement or lockdown.

Critically, the researchers suggested a mobile app could help slow the spread of infection until vaccines and antiviral treatments became widely available.

'Matter of urgency'

At a special Government committee meeting yesterday, prominent epidemiologist Sir David Skegg said rapid contact-tracing was needed as a "matter of urgency".

Skegg noted that apps had been used in South Korea and Taiwan for contact tracing, which required special legislation to be passed.

Health Minister David Clark said the Government was "very actively" looking at those options while also weighing up human rights and privacy considerations.

But last week, Privacy Commissioner John Edwards told the Herald that health authorities already had the green light to track coronavirus-infected people's movements via data collected by their mobile phone companies.

Asked if the Ministry of Health could approach a mobile phone company and ask it to hand over the movement data of someone infected with Covid-19, Edwards said this could be done under existing law.

"Although the telcos might want some additional assurance that they would not be liable for providing that information," Edwards said.

"Under the Privacy Act and Telecommunications Information Privacy Code, telcos are able to disclose telecommunications information where they believe on reasonable grounds that it is necessary to prevent or lessen a serious threat to public health.

"It is possible that they could also approach me for a special authority under the Privacy Act if they felt one was needed in the circumstances."

Some capabilities already existed. IT services company Datacom has developed a system that allows police to access location data from Spark, Vodafone and 2degrees customers' mobile phones.

Edwards was in the process of consulting on guidelines around the technology, including the proposal to track a person via their mobile phone without the person's permission if necessary if there is a threat to their wellbeing or public safety.

The New Zealand Telecommunications Forum's chief executive, Geoff Thorn, said he was aware of apps being deployed overseas to help authorities locate people who may have been exposed to the Covid-19 virus.

"To the extent that any technology solution might be considered, as an industry we are guided by what is permitted under current legislation."

Statistics New Zealand has also been using anonymised mobile phone movement data from Spark and Vodafone as part of a people-tracking project.

Tom Barraclough, a researcher and legal expert in synthetic media who has published a blog post on the issue, said contact tracing was going to prove "fundamental" for limiting Covid-19's spread.

"I think it's inevitable that a discussion about the use of digital data for contact tracing is going to occur, and it's important for people to get ready for that discussion," he told the Herald.

"If we're talking about an app that is going to have the potential to track the movements of three to four million New Zealanders, then we need to spread the discussion very widely, to make sure everybody feels heard."

Manual tracing still crucial

Otago University epidemiologist Dr Ayesha Verrall agreed that smartphone apps could "substantially" speed up tracing.

"I wouldn't go so far as to say it is impossible to do manual contact tracing. That depends a lot on the parameters included in the model, which are uncertain," she said of the Oxford study.

"In the end, the practical response is to say we need an up-scaled manual contact tracing system and a smartphone app as well."

Verrall said there were a variety of similar apps around the world "and it goes without saying that we would need to test such a system in New Zealand".

"The ethical and privacy issues listed are also very important. To have maximum public health impact the app would have to be acceptable to most people with smartphones."

As a point of strategy, Verrall said, a strong testing and contact tracing system meant we could manage moderate outbreaks in a targeted way, without going into lockdown.

Otago University epidemiologist Dr Ayesha Verrall agreed that smartphone apps could
Otago University epidemiologist Dr Ayesha Verrall agreed that smartphone apps could "substantially" speed up tracing. Photo / Supplied

Associate Professor Malcolm Campbell, a health geographer at Canterbury University, also pointed out that relying on smartphone-based approaches involved some big assumptions.

"So just imagine you've been out walking about with a smartphone in your handbag, pocket or wherever, the data that is collected from Bluetooth signals can be used to see if you have come into contact with a case of Covid-19," he said.

"Now here is the real important bit, assuming you are told this information, that someone you came into contact with has Covid-19, you'll then need to self-isolate.

"This requires high public trust and buy-in. It also requires people to know they need to be tested, and get tested."

Campbell pointed to three other key considerations: that people not only needed to always be carrying smartphones with them, but also having the tracing app installed and Bluetooth turned on.

"If not, we assume the clever solutions fall over. We know from previous research using technology in a healthcare setting, that not everyone has smartphones, often those with health challenges as well as some groups of older people," he said.

"As long as these people stay isolated, the risk is mitigated, but still, we cannot assume that technology is 100 per cent effective and will save us all."

He emphasised that other "weapons" would be required to combat Covid-19.

"There are five main measures we have to eliminate Covid-19; isolation, contact tracing, quarantine, social distancing and hygiene measures," Campbell said.

Contact tracing was absolutely vital, he said, but so was the current lockdown.

"So, we need a range of virus response weapons, particularly as we don't quite know yet if we are 'flattening the curve'.

"What is key in all this is that we need to be sure that infection is slowing or stopped ultimately. This means an infected person only infects one person or less.

"Then, the epidemic will start to head towards elimination. The more we follow the guidance, the higher the chances of this happening.

"We need to be really careful about looking for magical technological solutions to solve everything.

"The lockdown is really tough medicine, but a runaway pandemic is worse, especially for those in our communities who really need our protection from the virus right now."

• Dr Andrew Chen's full article on smartphone apps and Covid-19 can be read here.