Let’s talk about contact tracing

Published on the 28/04/2020 | Written by Newsdesk


Contact tracing app

InternetNZ online event serves up food for thought on the use of technology for contact tracing…

Contact tracing is one of the four key pillars critical to reducing restrictions due to Covid-19. Along with medical preparedness, testing efficiency and border security it is a key capability to eliminate the pandemic threat.

But it is a vexatious issue that means different things to different people. From police checking in on you while in isolation to government agencies tracking your every move, contact tracing technology triggers an emotional response.

These ‘musts’ are alongside a vaccine as necessary precursors to society being able to return to normal.

Now, with Australia having launched a tracing app over the weekend based on Singapore’s TraceTogether (relying on Bluetooth proximity) and New Zealand authorities seemingly still some way off (and avoiding open dialogue on plans), it is an issue that has polarised views and policy responses.

To help moderate the wide range of perspectives, InternetNZ hosted an online meeting on April 17 where a panel representing academia and public agencies presented perspectives and issues with contact tracing and where technology fits into the picture.

The webinar attracted a broad attendance and range of input across public and private sector, with many experts, including iStart, contributing to a collaborative document post-event.

In our view, the right solution must include an app to augment established methods, the app must run in the background, it must sense the proximity of other devices and it must maintain the privacy of the user. Importantly for uptake, the solution must be endorsed and promoted across all Covid-19 authorities just as the lockdown has been. These ‘musts’ are alongside a vaccine as necessary precursors to society being able to return to normal.

The key themes and questions arising from the online meeting and the InternetNZ document developed subsequently are below. Note that this is an excerpt of the summary produced by InternetNZ:

Public Health

  • Around the world there are proposals for technology options to replace or to augment manual contact tracing.
  • There was a shared view that in New Zealand, technology for contact tracing needs to augment existing manual contact tracing techniques.
  • Manual tracing is time consuming and resource intensive; technology can speed this up and help address community transmission, as it can reduce the tracing time and reach more people.
  • The critical thing is to define the problem the tech will solve, before deciding on the tech.
  • There is no perfect solution. We just need to decide in a timely way, implement and deploy.
  • There are different lenses through which to look at technologies:
    • Some tech looks at location, including GPS and locations from telco networks. Some tech uses contacts rather than location and figures out who you have been spending time with. Public health officials can use this to alert people you may have been in contact with. Some argue this is better for privacy.
    • Some solutions don’t need active participation from the user (information from banks; telcos) others do (apps; diarise).
    • Some solutions are voluntary others are not. The tech solution can be designed either way.

Technology options

  • There are several ways of looking at technology options for contact tracing:
    • Whether location or contacts are reported;
    • The type of underlying technology (Bluetooth versus GPS or telco data);
    • Whether an approach is on-by-default or opt-in;
    • Whether use of tracing technology is tied to coercive legal or social measures such as requiring use of an app to leave home.
  • Any technology proposal should be assessed against a clear purpose, with clearly set out design goals and risks.
  • Any technology proposal should be approached in a way that proportions impacts to needs. If community spread is limited, it may not be needed.
  • Technology and other approaches need to be integrated and to work in context, looking at people’s social, economic, and health situation. Requiring people to stay at home may not be realistic or helpful if their home is crowded, damp, or insecure.
  • Tech solutions should also consider what else might want to be addressed, eg enforcing quarantine, allowing streamlined access to official information.

Social Licence

  • Social licence is about the community giving continued permission and acceptance to a proposed solution. This means asking people, but there are ways of doing this quickly in ways that include even the most marginalised.
  • It is important that the social license refers to the problem/solution that exists at a particular time – it can expire.
  • Social licence is about trust. If something is voluntary, people may opt out if the explanation of what the tool is and why they should use it doesn’t gain their trust. If it is mandatory it can spur backlash.
  • Governance is important – being transparent and clear about management, monitoring, oversight etc.
  • Consultation is important. New Zealanders have eight questions they want answered about use of their data – see the data wheel. If these questions are answered, then social licence is more likely.

Security

    • Government has historically struggled with large technology projects. Contact tracing is likely to require quick work on a broad scale which increases these challenges.
    • Security decisions involve trade-offs and real-world considerations that depend on what is at stake and what risks it raises.
    • While security design is important, aiming for perfect security may require delays and barriers to uptake that would block effectiveness.
    • Independent assessment is needed to understand and manage potential risks technology systems at scale and under real conditions.

Privacy

  • Existing privacy law is robust but flexible, and can be applied while allowing the use of contact tracing technologies.
  • Participants asked whether data could be minimised in how much is collected, how long it is held, how widely it is shared, and by requiring deletion after the need for contact tracing passes.

Human rights

  • Measures that have an adverse impact on human rights including privacy must follow the principles of legality, necessity, and proportionality.
  • One of the key issues is the longevity of any proposed measure. How do we roll back the mechanisms being introduced during the crisis? Government needs to articulate how long the measure will be in place.
  • Another key issue is accessibility. Proposed solutions could be prohibitive for some, eg in order to go back into the world you need to sign up – this could put barriers up for people who don’t have access to technology etc.
  • Governments could use the tech to target minority groups or dissidents.
  • There is a body that is accepting complaints – special rapporteurs at the UN who are monitoring the situation – can take cases to them.

Conclusion
We think a few key points are worth highlighting:

  • People were highly interested and engaged, which indicates the depth of public interest in this issue. We were pleasantly surprised to have 160 registrations and over 80 people contributing to our shared Google Document, which is still developing thanks to continued work by people in the community;
  • People are open to technical solutions, but also want to understand how they complement other social, legal, and public health interventions. People are also concerned that proposed tech solutions will be ineffective if they do not take into account the realities of people’s lives including the digitally excluded;
  • A range of concerns were raised, including technical challenges, security, privacy, honouring te tiriti and human rights concerns. People want to know how these will be managed, and there was clear support for early engagement and steps to uphold transparency;
  • It is vital to test, build and maintain trust in the wider community and in specific affected communities, through a clear governance framework, early consultation and ongoing engagement. It seems better address concerns openly and early rather than risking later and more costly conflicts;
  • We could only start this conversation. There are people and communities we could not reach, including vulnerable people who may face the highest risks, or be less likely to trust technical or government processes. More is needed.

Panellists:

  • Dr Sayali Pendharkar, University of Auckland
  • Dr Andrew Chen, Koi Tū
  • Dame Diane Robertson, who led the Data Futures Partnership
  • Donna Cormack, Te Mana Raraunga
  • Kate Pearce, TradeMe and on the InternetNZ Council
  • Joy Liddicoat, University of Otago and on the InternetNZ Council
  • Lucie Krahulcova, AccessNow
  • Tom Barraclough, Brainbox Institute

Thanks to InternetNZ for facilitating and documenting the discussion.

The full text is available in a public document here: Let’s talk about technology and contact tracing in New Zealand.

Questions or comments...

  1. Andrew Pengelly

    Remember that these solutions do record person to person contact, they only record phone to phone contact. And then only if the require radio receiver (eg Bluetooth), is turned on. So if I go out without my phone any interactions I have with another person will not be record. I may be unusual in that respect but I often go out without my phone. I also have my Bluetooth turned off by default. This is to conserve battery power so again this technology will be unsuccessful at registering my social interactions.

    What would be useful is if the generic GPS location tracking offered by phones could be accessed in such a way that if a person is identified as having COVID-19 then I could match my movement history against their movement history to see if there was any overlap. If this could be done anonymously by sharing my tracking database with tracking databases of known infected persons to look for overlap instances then that would be really useful

    Reply
    1. Hayden McCall

      Thanks Andrew – let’s hope the updates from Apple/Google (iOS/Android) bluetooth protocols embed better tools for proximity recording in terms of not needing an app to be open to work (as with iOS currently) and for battery conservation. It will never cover 100%, but it is a better alternative than GPS which would prompt tracing of everyone on every floor in an office building for example.

      Reply

Post a comment or question...

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.


Follow iStart to keep up to date with the latest news and views...



MORE NEWS: