Lucy Sweeney* says digital contact tracing seems to have gone the way of the dinosaurs as the coronavirus pandemic has evolved, but overseas, educators are thinking about new ways to share COVID information that could change the way we approach future outbreaks.
While the world has been busy ramping up booster shots and procuring enough tests and masks to handle the ongoing threat of Omicron, another tool once touted as crucial in controlling the spread has become something of a relic.
According to some, “contact tracing is so 2020.”
Several states in Germany, where the Corona Warn app was seen as a world leader, gave up contact tracing just before Christmas, in the face of an overwhelming wave of infections.
Across Europe, contact tracing apps managed to track 5 per cent of confirmed cases up to November last year, according to an analysis published by Spanish international affairs publication El Orden Mundial.
Parts of the US and Canada are also winding back efforts due to a lack of resources, while those contact tracers still hitting the phones are reporting fatigue and overwork.
After announcing a controversial plan to abandon contact tracing altogether, South Africa has instead refocused its efforts on healthcare facilities and vulnerable settings.
Public health centres in Japan have also been stretched to the limit, and are scaling back their contact tracing capacity, while Singapore remains committed to its TraceTogether app particularly in higher-risk settings.
In Australia, the federal government’s $9 million COVIDSafe app has been plagued with issues, with recent calls for it to be scrapped entirely.
States and territories continue to carry out manual contact tracing — a labour-intensive and costly exercise — but most are now focused on household transmission.
Many jurisdictions have abandoned listing exposure sites to alert “casual” contacts, prompting concerned citizens to take matters into their own hands and share their location histories on local Facebook groups.
It’s clear many of those testing positive are anxious to let the people around them know.
And as states and territories finesse their approaches to logging positive results from rapid antigen tests, people are looking for ways to manage their own health risks.
If people are still hungry for COVID information, what are the alternatives?
Australian epidemiologist Mary-Louise McLaws, who advises the World Health Organisation on outbreak management, has called for governments to consider the alternatives.
Given the strain on PCR testing and the pivot to reliance on rapid antigen tests, she’d like to see a centralised system for reporting RAT results, similar to those used in the UK and Canada, as well as a new approach to contact tracing.
While the Canadian government, like Australia’s, does not keep track of rapid test results at a national level, there are a number of citizen-led projects offering solutions.
Rapid Screening Consortium, led by the non-profit Creative Destruction Lab at the University of Toronto, has been developing guidelines and protocols since August 2020 to help businesses and educational institutions screen for COVID-19 and securely log results from rapid tests.
The consortium’s approach aims to tackle the economic problem presented by COVID, which CDL says really comes down to sharing information.
“You don’t know if I have it, I don’t know if you have it, so we lock things down or use blunt instruments because we don’t have information,” executive director Sonia Sennik told Toronto magazine The Local.
This kind of information sharing is also the basis of another app Professor McLaws has championed as a key tool in managing COVID-19 into the future.
Self-described as a new approach to solving pandemics, NOVID “flips the perspective” on COVID apps from primarily helping people protect those around them to instead helping them protect themselves.
The app was launched by US mathematician Po-Shen Loh in 2020, but hasn’t taken off, which he says is because of technological teething issues and the global pivot to vaccination as everyone’s ticket to freedom in 2021.
But both Professor McLaws and Paul Garrett, an Australian researcher who led efforts to understand the efficacy of various digital contact tracing apps, think the concept has legs.
NOVID acts as an early warning system to let people know that COVID might be creeping closer, before they’ve even been exposed to a positive case
It relies on game theory and network theory to create a radar-like map of the user’s frequent contacts and show them how many relationships away they are from a positive case — like six degrees of Kevin Bacon, but for COVID.
“If the government is not going to really take the numbers in control … they could be using something like Po-Shen Loh’s app to help people help themselves and advise others that they’re infected,” Professor McLaws said.
How does NOVID work?
Professor Loh says the principle is simple.
“If you think about what contact tracing apps were about, those are apps to tell you after you are already potentially dangerous to other people, so that you can protect other people from you,” he said.
“NOVID turns that upside down.
“It’s an app to tell you when you might want to take some caution, before you’re the dangerous one, so that you don’t get sick.”
Why would this be any more effective than other contact tracing apps? This is where things get a bit nerdy.
The app is underpinned by a part of non-cooperative game theory called Nash’s equilibrium, which is the state where players will always make the decision that best serves them, regardless of what others do.
Essentially it boils down to the assumption that people can be reliably expected to act in their own self-interest.
Professor Loh argues NOVID leverages this concept because by downloading the app to protect themselves from getting COVID, users are also contributing to a benefit for the broader community by reducing the spread of the disease.
Or, to quote Russell Crowe as John Nash in A Beautiful Mind: “The best result will come from everyone in the group doing what’s best for himself, and the group.”
(It’s worth noting that the bar scenario Russell Crowe uses to explain this theory is not really an example of Nash’s equilibrium, but he gets points for trying.)
NOVID also uses network theory to construct its broad map of users’ relationships.
It does this in a similar way to contact tracing apps like COVIDSafe, using Bluetooth and WiFi to scan for other phones nearby that have installed NOVID.
The app takes notice of the phones each user is in frequent contact with, then, when someone tells the app they’ve tested positive for COVID or have been identified as a close contact, that message is passed on to their network anonymously.
So for example if you regularly catch up with your brother and his housemate’s co-worker’s wife’s elderly mum got sick, you’d see that someone five degrees away from you was positive, but you wouldn’t know who.
“If you can see the network around you, and know that this is coming, then you can use the masks dynamically, decide whether a rapid antigen test is a good idea for you, decide if you want to wait three hours in line for a PCR.
That changes the calculation entirely,” Dr Garrett said.
Yes, there are technical flaws and privacy concerns
As with any technical solution, there are limitations and caveats.
NOVID doesn’t use the Google-Apple Exposure Notification (GAEN) system that allows other contact tracing apps to work seamlessly across different device types.
This has been among the major criticisms of the federal government’s COVIDSafe app, partly because without this integration, it’s extremely difficult for apps to catch interactions between two locked iPhones.
But Professor Loh says missing some of those interactions doesn’t reduce NOVID’s capabilities.
Rather than record every single encounter, his app is trying to build a network based on the people who you spend the most time with.
So it’s OK if it misses the barista who poured your coffee, or the instances where you spoke to your colleague at the kitchen sink but left your phone in your pocket.
NOVID’s betting that, given the average person checks their phone 85 times a day, if it’s someone you see a lot of, at some stage you’ll wake up the screen and allow the app to work.
As with COVIDSafe, there are also privacy and security concerns.
NOVID doesn’t collect names, locations, emails or phone numbers from its users.
It randomly generates a pseudonymous ID for each user and sends encrypted information about which other users they come into contact with back to a centralised database.
The University of Melbourne’s Dr Garrett says the ultimate solution from a privacy perspective would include pseudonymous IDs that are updated at regular intervals, and that aren’t stored on a centralised database.
But there are ways to address some of these concerns for future use, such as releasing part of the source code for experts to inspect.
NOVID did release its source code for an independent review by partner university Georgia Tech in 2020, and Professor Loh said the company would be open to doing the same with future partners.
In the meantime, Professor Loh says ultimately it’s up to users to weigh up the privacy cost, which he sees as minimal, with the overall benefits of an app like NOVID.
Could something like this work in Australia?
Just like contact tracing apps, this type of early notification system becomes more effective the more people there are using it.
So far, NOVID has been adopted in a few universities and cities around the United States, with varying degrees of enthusiasm.
At Georgia Tech, users peaked at 3,723 out of an estimated 12,370 on campus, while over at Carnegie Mellon the figure was closer to 500 out of 5,000.
Both universities said that while the app was useful early on, activity waned as the pandemic changed shape.
“The intention here is to provide people with data and trust that they can make intelligent decisions themselves.
“I think that’s the real unique aspect of it,” Georgia Tech’s Shannon Yee said.
“There’s other features that we never even got to see, because we kind of wound down the effort of promoting it here on campus.
“People could self-report if they got vaccinated, that way you could see within your group, are you with vaccinated individuals or unvaccinated individuals? So I think there’s still a lot of power there.”
Other public health experts who spoke to the ABC said that without widespread uptake, an app like this would ultimately do very little to reduce the current spread.
“Efficacy isn’t just determined by how well the app works, but also, its acceptance by the public and whether they’re willing to use it,” Dr Garrett said.
“If an app is 100 per cent effective when used, but no-one uses it, then there’s no social licence and it doesn’t work.
“It’s not so much the usefulness of the app that dictates whether people will use it, it’s what the cost benefit analysis is.
“On the cost side, there’s privacy concerns.
“On the benefits side …until the rapid antigen tests and masks [are more freely available], the benefits are going to be quite significant.”
Dr Garrett said if it were to be adopted in Australia or other parts of the world in future outbreaks — COVID or otherwise — then private organisations may be best placed to pick it up, in the same way Canada’s Rapid Screening Consortium has been working.
Professor Loh is hoping that by sharing the concept with the world, people will rethink the way smartphones can be used to help manage contagious diseases.
“The main value of what we’re trying to do is to put a new concept into the public consciousness, so that if something emerges which has [a higher fatality] rate, everyone can [use this],” he said.
“The only thing I care about is that we have caused the world to realise that how you fight a pandemic with a smartphone is by making an app to help the user not get sick themselves.”
*Lucy Sweeney is a digital journalist and producer with the ABC’s International team.
This article first appeared at abc.net.au.