Can we relax lockdown without Coronavirus proximity contact tracing?
I want to start a conversation about the effectiveness of and alternatives to the current onslaught of Bluetooth contact tracing apps. Would empowering citizens with real time information and providing manual contact managing tools drive more adoption and therefore effectiveness? And how might we incentivise adoption of new and untested technology without coercion, fear or drama? Can we use a carrot rather than a stick? So, here’s an exploration of these questions with a wondering whether this topic requires a lot more research (before we rush in). Comments are most welcome below and text modification suggestions are most welcome in the original Google Doc.
Contact tracing apps are currently being developed at a rapid pace across the globe. Some apps are being built by established organisations/teams and some are seemingly springing out of nowhere. Apple and Google should be praised for collaboratively creating a cross-compatible underlying Bluetooth API [1] for their respective mobile OSes that app builders should now integrate with — rather than trying to roll their own hacks.
However, there are many fundamental questions on the use of Bluetooth contact tracing apps, not least: are they effective at all [2]? There is a lot of research into contact tracing being a good thing [3] — but the technologies used to implement it and the methods used to drive adoption have not been tested en mass outside of China and some other Asian countries [4].
The main aim of contact tracing apps is to alert a person that they may have come into contact with someone who has tested positive to Coronavirus, so that they can then move into self isolation. Now seeing that anyone could be carrying the Coronavirus and not exhibit any symptoms it follows that, without consistent and extensive mass testing, we will never know who does and does not have the virus in their system. So, until any one person has been tested, shouldn’t we just assume we/they are a virus carrier? You will see time and time again medical professionals crying out for testing. Is it worth contact tracing in the absence of mass testing?
It is also worth noting that contact tracing is after the fact and, in some cases, could be way way way after the fact. But could we be more preemptive and proactive rather than reactive? For instance, like most of us, before having sexual relations with a new partner, we share when we last had a STI test, the results and who and what and how we’ve done since then. We can then both make assessments on if and how to continue. Shouldn’t we be relaxing lockdown with the same degree of rigour and concern, preemptively?
There is also a big question around moving a whole country from complete lockdown back to ‘normality’ in one huge step. Is a gradual, managed approach far more sensible? Perhaps based on local needs rather than a one-size-fits-all model? [5] Is national government or even local government/councils actually capable of orchestrating such a mammoth task of piecemeal lockdown releases based on local feedback and criteria? Can citizens play their own part in mitigating potential viral spread as lockdown releases?
Is national government and even local government/council limited to taking too high a view of potential risks, whereas citizens themselves can work at a much more granular level? Can citizens be given tools to assist them in making informed decisions within their own communities and families [6]? Can we create and incentivise a much more participatory system where we are all collaborating and sharing in real time; alerted to and managing hotspots as they arise? Much of this management has happened before within local communities managing local, albeit minor outbreaks of infestations, without the need for centralised monitoring or control.
In China, I think it’s fair to say, people are used to government interventions in much of their lives. I also think it is fair to say that within Europe and the Americas people are more vocal about their civil liberties and individual rights. There is a lot of caution around the implementation of mass surveillance apps. Even if the caution is unjustified, the rumours alone can affect the adoption of these systems — perception is everything. So we need to work doubly hard to make sure that we create the right apps to do the right job. The more empowered we can make people, the more emotional buy-in we will create, the more responsible people will be, and the more willing adoption we will have and the more safe we will be.
High levels of distrust by the general public in big business, centralised systems and government, are going to make it harder for apps to be adopted [7]. In order to build confidence and drive adoption of solutions we need to investigate effective tools that are totally disconnected from vulnerable centralised storage and control.
Could we build tools that would assist citizens to reduce the risks of contracting or passing on the virus and also help manage contact awareness and alerting. Could these tools be totally GDPR compliant and also not rely on any centralised services or authorities for the contact notification system to work effectively? Could these tools enable citizens to make informed decisions about whether to self isolate or to risk viral transfer with their own contacts, based on recent potential transfer history, upstream test results, general risky behaviour, and so forth?
The younger and more healthy a person is, the less likely they are to pose a risk of developing life threatening symptoms and thereby adding load to the NHS. Can we trust citizens to make their own decisions as to the risks involved, based on their existing health level (smokers take extra care here), and hence expectations to increase load on the NHS?
The human drive to preserve and protect our loved ones, the family and the tribe, is strong. Can we employ this drive as a motivator for using tools which also protect us on a larger scale: our NHS, our country and our planet wide society? We share our personal information willingly and openly between our trusted contacts. Trust between these contacts is vital. And within families and friends trust is already high and sharing private information is the norm.
So, as we release the lockdown, how do we increase the size of our viral contact containment pods; our isolation units; our extended households; our potential contagion teams; our ‘quaranteams’, in a safe and managed way? And how could apps help us do that? And what information would need to be available to us in order to make decisions about maintaining connection with certain contacts or disconnecting from other contacts based on perceived/known risks to them or us? The more open we can be with our trusted contacts, the more everyone can make their own decisions freely with no drama needed.
For example, at a simple level, let’s just say that 2 families have been self isolating apart. Then, at the start of ‘lockdown release’, they decide that it is safe to have a barbecue together and they are not cautious about sanitising everything. Effectively, they are now one self isolation team. One member of this team goes off for work. Let’s call her Sally. On the way back from work Sally receives a message saying that one of her work colleagues has just been positively tested for the virus. Sally calmly sends a message to her team and tells them that she may be compromised. The team can then have a sensible discussion on what to do next. They can ask Sally to isolate alone. They can decide to split their team up into the original two family units. Or they could make a judgment call based on their perceived health and other known risks and just continue to operate as their combined team.
So, extending this model, the logical contagion teams can expand and contract over time based on extremely local and timely data, and assessment based on the personal choices of those most likely to be affected within any team.
Of course, individuals could, if they wished, inform central authorities about the state of their health and the decisions they are making to self isolate or not. But each person needs to be in control of what data they share on their own terms. Central authorities may want to make it attractive to share such data if they so wish. Also guidance from larger institutions and government could be much more targeted to app users.
This may all get really complex and may end up either not being buildable or not buildable in the timescale needed. However, if we start small and do simple functions first, let us just take it one step at a time and see how far we can get, eh? Is this all worthy of more research? Is this a potential way to move from total lockdown? How are we going to get out of lockdown [8]?
I should point out that all of this talk of contact tracing apps is all academic if we cannot be constantly testing ourselves. So, in the absence of these home tests we need other tools that we can work with.
More work needs to be done. If this line of investigation appeals to you please let me know.
Be happy and stay safe!
Cheers Daniel
References
[1] Apple and Google partner on COVID-19 contact tracing technology | Apple
[2] Coronavirus: UK confirms plan for its own contact tracing app | Leo Kelion | BBC wrote “Ross Anderson… flagged doubts about the use of Bluetooth” and cited Contact Tracing in the Real World | Ross Anderson
[3] Covid-19: why is the UK government ignoring WHO’s advice? | Allyson M Pollock | BMJ wrote “The mathematical model used by the UK government clearly shows that rigorous contact tracing and case finding is effective” and cited The Efficacy of Contact Tracing for the Containment of the 2019 Novel Coronavirus (COVID-19)
[4] Coronavirus Contact Tracing Apps
[5] Mass testing is the only way to stop the virus — it’s long overdue | Anthony Costello | The Guardian wrote “Rather than adopting a one-size-fits-all national policy, we need to devolve power to our local authorities and their public health outbreak teams.”
[6] It’s time to take back control of the epidemic | Molly Scott Cato | Facebook wrote “we may be forced [to] organize to protect our own communities through setting up local monitoring and contact-tracing systems.”
[7] The Challenge of Proximity Apps For COVID-19 Contact Tracing | Andrew Crocker | EFF wrote “Fear of disclosure of such proximity information might chill users from participating in expressive activity in public places.”
[8] Wanted: An exit strategy from the overt surveillance of smartphone contact tracing | Simon Sharwood | The Register wrote “regulators need to plot an exit strategy from this new form of deeply personal and intensive surveillance.”
Social media
If you’d like to like or share, here’s where I’ve posted:
https://www.linkedin.com/pulse/can-we-relax-lockdown-without-coronavirus-proximity-contact-harris/
https://www.linkedin.com/feed/update/urn:li:activity:6656516772998504448
https://www.facebook.com/dahacouk/posts/10158386966695972
https://twitter.com/dahacouk/status/1250753434330021888
Daniel Harris
Daniel has worked extensively in the UK Internet industry. In 1996, Daniel co-founded Cerbernet and served as CEO till late 1998. Under his watchful eye Cerbernet grew to become the UK’s leading independent broadband Internet Service Provider. Cerbernet was acquired for over 4M GBP by First Telecom early in 2000 — at the height of the dotcom bubble.
Daniel is currently leading Kendraio, which he founded in 1999, to promote an open content marketplace for the Internet. The aim is to drive interoperability between and within every link in the content value chain — enabling artists to manage their digital assets, metadata and workflow more efficiently — driving increased revenue to content owners and a more pleasurable buying experience for consumers.
Previously Kendraio investigated ways to prove that a story was true and received a 50K EUR grant from Google News Initiative to do so. Kendraio has received over 1M EUR as a partner in European funded projects: P2P-Next and Saracen, and currently Bloomen. Kendraio has also received a further 100 GBP grant from the UK for researching a rights management app which has evolved to become a generalised ‘data browser’, Kendraio App. With this app we created the Kendraio Coronavirus dashboard graphing live data from multiple APIs.
