Te Tari Taiwhenua
Department of Internal Affairs
Purpose
1. Your office has confirmed your oral item on the draft Cabinet paper
Using the
CovidCard to improve contact tracing. This aide memoire provides you with
information and suggested talking points to support you at the Cabinet Social
Wellbeing Committee (SWC) meeting on Wednesday, 29 July 2020 at 9.30am.
2. Attached as
Appendix A is the draft Cabinet
Using the CovidCard to improve contact
tracing paper as at 3.00pm Tuesday 28 July 2020.
Appendix B provides you with
suggested talking points for your discussion at SWC. Q&As are also provided as
Appendix C.
Background
3. Effective contact tracing is a critical component of the strategy to isolate COVID-19 in
New Zealand, by identifying close contacts, isolating them quickly, and stopping the
spread of the virus. Digital technology can complement manual contact tracing by
speeding up the process, filling in memory gaps, and recording contact with strangers.
It could also allow contact tracers to access contact information more rapidly,
improving the chances of quickly identifying and isolating potential cases of COVID-19.
4. On 20 April 2020, Cabinet noted that officials are assessing a Bluetooth-enabled card
to support contact tracing and will report back to the Minister of Finance, the Minister
of Internal Affairs, the Minister of Health, and the Minister for Government Digital
Services [CAB-20-MIN-0175 refers]. On 13 May 2020, the Minister of Justice was added
to this group and SWC noted that a report-back to Cabinet on the consumer
application Bluetooth assessment, and the Bluetooth-enabled card, was scheduled for
June 2020 [SWC-20-MIN-0043 refers].
5. The paper seeks Cabinet agreement to proceed with the next phase of the CovidCard,
including further trials, as part of keeping technology options open to improve contact
tracing. If Cabinet agrees to this recommendation, Cabinet would be provided advice
fol owing the further trial by the Minister of Health and the Minister for Government
Digital Services, after the completion of the next phase of work, on whether to
proceed with a full national deployment option for CovidCard and alternative options
for deployment, or use of other developing technologies.
Cabinet paper summary
One simple and effective technology to improve contact tracing has not been identified
around the world
6. Many countries around the world have tested or developed technologies to support
contact tracing for COVID-19. These technologies are mostly smartphone apps.
Officials have not identified a country that has developed one simple and effective
technology for contact tracing.
7. The NZ COVID Tracer app has had good initial uptake and adds to the accuracy of the
National Health Index data, yet it relies on people scanning QR codes. The Ministry of
Health is planning to undertake a trial of an app based on the Apple and Google
Exposure Notification Framework technology with the NZ COVID Tracer app.
RESTRICTED
Page 2 of 11
Te Tari Taiwhenua
Department of Internal Affairs
The Public Private Partnership group produced a report recommending the Government
consider deploying the CovidCard
8. As directed by Cabinet, the Government Chief Digital Officer (GCDO) has worked with
the Public Private Partnership group (PPP) to investigate an alternative to Bluetooth
mobile apps, the Bluetooth-enabled CovidCard. The PPP is a group of private sector
volunteers and contractors and is not a formal entity.
9. The PPP produced a report recommending the Government consider deploying the
Bluetooth-enabled CovidCard to improve contact tracing. Initial assessments have
shown that the card works technical y, but there are issues with the card’s security and
false positives (when someone is incorrectly classified as a close contact) that require
further investigation.
10. The report states the CovidCard would take six months to deploy from a decision to do
so, at a cost of $98.5 million in the first year and $64 million in the second year. These
are indicative costings and do not account for the funding of policy, regulatory or
enforcement activity associated with a nationally rolled out CovidCard.
The CovidCard has potential advantages, as wel as disadvantages
11. The CovidCard does not have any connections to the internet or mobile networks.
From a digital inclusion perspective, the cards also do not require any skills to use or
for anyone to already have a device. However, this technology is expensive, largely
unproven in real world environments, and it is not certain if uptake of the cards would
be better than any other technology solution.
Initial assessments have shown that the CovidCard works technical y
12. Following the PPP’s report, the GCDO led cross-agency work to independently assess
the CovidCard and the benefits of the CovidCard for contact tracing. Results from these
assessments are promising but there are issues that require further investigation,
including the CovidCard’s security and the extent of false positives.
13. Independent model ing of the epidemiological value of the CovidCard was completed
on 24 July 2020. The model ing reported that Bluetooth apps and card-based proximity
detection systems perform similarly at comparable levels of uptake, but other factors
such as usability need to be considered.
There are wider issues with the deployment of the CovidCard or another technology
solution
14. There are broader issues to consider about the deployment of technology solutions.
The success of any technology solution requires strategic policy decisions to address
social licence and public uptake of the contact tracing technologies. Additionally, it has
not been addressed where the CovidCard would fit in with the wider context of New
Zealand's contact tracing system and plans for using technology to improve contact
tracing.
We suggest that you propose that the CovidCard is investigated and trial ed further
15. In order to keep multiple technology options open to improve contact tracing, we
suggest you propose to proceed with a field trial of the CovidCard. This will also involve
research into public sentiment about using technologies to improve contact tracing. A
tentative site for the trial has been identified in the Rotorua region, pending
consultation with the Caring for Communities workstream as part of the Al -of-
RESTRICTED
Page 3 of 11
Te Tari Taiwhenua
Department of Internal Affairs
Government COVID-19 response. Your expectation is that this would be completed by
early October 2020, subject to PPP engagement.
16. The findings of this trial wil be reported back to Cabinet to inform any decision on
whether to deploy CovidCard as part of New Zealand’s contact tracing system.
17. There is uncertainty about the PPP’s continued involvement under this approach.
Without the PPP, Government would need to bring in people with the right skil s to
facilitate the trial, which would impact timeframes, costings and outcome. The
Minister of Health and the Minister for Government Digital Services will engage with
the PPP and confirm next steps.
New arrangements would be needed for the next phase of work on the CovidCard
18. The GCDO has worked with the PPP on the proof of concept for the CovidCard. It is
important to continue to develop the CovidCard design while determining the value it
adds to the objectives of contract tracing and how it could be integrated into the wider
contact tracing system. Therefore, the next phase will be jointly led by the GCDO and
the Ministry of Health.
9(2)(j)
Financial implications
20. The Ministry of Health and the Department of Internal Affairs for the GCDO have
reviewed options for reprioritising funding from within the respective agencies’
2020/21 baselines.
21. Subject to Cabinet agreement, the Ministry of Health will require additional funding of
$1 mil ion for 2020/21. This would require a drawdown of $1 million from the Ministry
of Health COVID-19 Contact Tracing – Tagged Contingency.
RESTRICTED
Page 4 of 11
Te Tari Taiwhenua
Department of Internal Affairs
Appendix A: Draft Cabinet Paper Using the CovidCard to Improve Contact Tracing
RESTRICTED
Page 6 of 11
Te Tari Taiwhenua
Department of Internal Affairs
Appendix B: Suggested talking points for the Cabinet Social Wel being Committee
meeting on Wednesday, 29 July 2020
Digital technologies can support contact tracing
1. I am seeking support from Cabinet to begin further testing and a field trial of the
CovidCard to support manual contact tracing.
2. This paper seeks agreement to draw down $1.000 million from the COVID-19 Contact
Tracing – Tagged Contingency,
3. Effective contact tracing is a critical component of the strategy to eliminate COVID-19
in New Zealand.
4. Digital technologies can complement manual contact tracing by speeding up the
process, filling memory gaps, and recording contact with strangers that would have
been missed in the current process.
5. In April, Cabinet noted that the GCDO and Public Private Partnerships group were
assessing a Bluetooth-enabled card, called the CovidCard to assist with contact tracing
in New Zealand.
Technical overview
6. This card uses Bluetooth technology to record when it into contact with another card
that is within two metres for 15 minutes.
7. The CovidCard does not have any connections to the internet, mobile networks, or
GPS, requires no skil s to use and does not store personal information. The close
contact data on the cards is kept for 21 days and after that time it is deleted.
The card has had some initial testing
8. The Public Private Partnership has conducted technical trials of the card as well as a
small trial in Nelson Hospital. They delivered a report which found the cards to be
technical y sound and provide epidemiological benefit.
9. This report has been reviewed by the Defence Technology Agency agreed the card
does work technical y, but also noted it needed more testing.
10. Independent model ing of the epidemiological value of the CovidCard was completed
on 24 July 2020. The modelling reported that Bluetooth apps and card-based
proximity detection systems perform similarly at comparable levels of uptake, but
other factors such as usability need to be considered.
11. The GCDO is also working with the Ministry of Health’s contact tracing experts to
investigate how the data from the CovidCard can add value to the existing contact
tracing approach. The next step would be to jointly work with Public
12. While the card does work technical y, we do not know how the public wil accept the
card or how well it would perform in a real-world scenario.
Further trials of the cards are required
13. Further trials would include a community trial in Rotorua region with 250-300 cards,
further technical testing, and market research.
RESTRICTED
Page 7 of 11
Te Tari Taiwhenua
Department of Internal Affairs
14. This research will allow us to understand how the cards would work in a real-world
scenario, if they are compatible with the contact tracing system, and if the public
would accept and use the cards if they were rol ed out.
9(2)(j)
16. The Minister of Health and I wil work with the Public Private Partnership to confirm
next steps.
Report back
17. The Minister of Health and I wil also report back to Cabinet fol owing the trial to
provide advice on whether to proceed with a full national deployment of the
CovidCard.
RESTRICTED
Page 8 of 11
Te Tari Taiwhenua
Department of Internal Affairs
Appendix C: Q&As to support you at the Cabinet Social Wel being Committee
meeting on Wednesday, 29 July 2020
Technical questions
1.
Wil the card work with NZ COVID Tracer? No, the card as designed by the PPP, would
not have the functionality to interact with any phone app, including NZ COVID Tracer.
2.
Does the card col ect personal data? No, the card only stores the encrypted serial
numbers of other cards it comes into contact with. All contact information for the
cards are stored on a separate government database.
3.
Can the cards be updated? Once the cards have been deployed, they cannot receive
updates.
4.
How long wil the battery last? It is estimated that the battery would last about 12
months, this could vary from card to card, depending how often it used. The battery is
not replaceable so a new card would need to be issued once the battery ran out.
5.
Is the data secure? The data on the cards wil be encrypted. The Defence Technology
Agency have reviewed the security of the cards and believe it to be secure. Ensuring
data security will also be part of the further technical tests.
6.
How is this different to an app? The CovidCard does not require any skills or a device
to use or require any interaction with it, which makes it very accessible. Some apps,
such as NZ COVID Tracer require people to scan QR codes, the cards would only
require you to wear them. However, unlike an app this is a more pre-emptive option
and would have a significantly longer time to rol out than an app. The CovidCard
would take 6 months to rol out nationwide.
Trial questions
7.
Why does the card need more testing? While we believe the card will work from the
technical perspective, we are not sure how it would work in a real-world scenario.
We don’t know if people would want to wear, how often people forget the card, how
accurate it is if worn in a pocket. A real-world trial and market research will help us to
understand how the cards could perform if they were launched nationwide.
8.
Does a trial mean we are committed to rol ing out the cards? No, the Government to
not out the cards after the trial. The Minister for Health and I will provide advice on
whether the CovidCard should be deployed.
9.
Where wil the trial be? The proposed location is the Rotorua region. Some regions
already have governance structures, including established leadership groups which
operated extremely well before and through COVID-19, some emerged during COVID-
19 and others have reached out for help connecting to different leaders within their
region.
10.
Why Rotorua? Through the COVID19 engagement, this region has the knowledge, the
understanding and trust to address the practical questions of proposed COVID options.
The regional leadership group membership includes Council, all iwi, MSD, Police, and 2
chairs of DHBs. Goodwill will be achieved through involving this regional leadership
group to be part of the solution - by identifying who in their communities could test
the COVID card.
RESTRICTED
Page 9 of 11
Te Tari Taiwhenua
Department of Internal Affairs
11.
How long will the trial go for? After a period of community consultation, the trial is
planned to take place over seven to ten days over a two-week period.
12.
How wil the trial work? Each participant would receive a card for that seven to ten
day period, which they wil use over that trial period.. A few participants would be
randomly selected to be positive cases to test the processes around col ected the data
from the cards and translating it into information to assist contact tracers.
13.
Why not do a larger trial? The trial is proposed to be for 250 to 300 participants in
order to keep the trial manageable.
14.
What about a trial in a isolation facility? That is possible, and the Ministry of Health
and the GCDO will consider it during trial planning.
Operational questions
15.
Why should the Ministry of Health now take the lead with the GCDO? The Public
Private Partnership and the GCDO have led the work to understand if the cards would
work from a technical perspective. However, any technology solution that is adopted
will have to work with the wider contact tracing system and the Ministry of Health is
best placed to make that assessment with the assistance of the GCDO and Public
Private Partnership.
16.
How long would it take to complete a nationwide rol out? The PPP estimates that it
would take roughly 6 months to manufacture and send out the cards to everyone in
New Zealand.
17.
Would the cards be mandatory? At this stage I am not recommending that the cards
be mandatory. While this could increase uptake, it has significant policy, human rights,
and Te Tiriti challenges.
Health efficacy of the cards
18.
Would the cards prevent another outbreak? It is unlikely that the cards would prevent
another outbreak of COVID-19. The cards would more likely assist with containing an
outbreak and improving the manual contact tracing process.
19.
Could the cards be targeted at digitally disadvantaged groups? From a digital
inclusion perspective, the cards also do not require any skills to use or for anyone to
already have a device. This means the cards, unlike some other technologies, are
accessible to everyone. Officials have worked with the All-of-Government Caring for
Communities workstream to inform the trial location.
20.
Could the cards be rol ed out to a target group ? They could, but a more targeted
approach is likely to be less effective at improving contact tracing. For the cards to be
effective a strong uptake is required so that the cards are more likely to save close
contacts. If the cards were only given to the elderly in care facilities, for example there
would be no additional benefit over the current contact tracing process unless the staff
and visitors had also been using a CovidCard.
9(2)(j)
RESTRICTED
Page 10 of 11
Te Tari Taiwhenua
Department of Internal Affairs
Public perceptions of the CovidCard
22.
What are public perceptions of the CovidCard? Market research is required to fully
understand the publics appetite for the CovidCard. The All-of-Government insights
research suggests that any new contact tracing initiatives in the current environment
would be hindered by a low perceived threat, misinformation and privacy concerns.
23.
How could public perceptions of the Card be increased? The All-Of-Government
insights research suggests that the CovidCard be established as a social norm early on
as well as widespread transparent messaging regarding contact tracing to reduce
misinformation.
RESTRICTED
Page 11 of 11