Should be on Morning Report around 0845 talking digital contact tracing

eep 0745, glad I woke up early

make that 0740, which is now in the past

In summary: we aren't sure what our manual contact tracing capacity is, but sometimes it does get too hard (e.g. people going through crowded spaces) so digital technology could help be more comprehensive - both more complete in the information and in the number of people traced

I categorise the tech options in two ways: those that use existing data like cellphone location data or transaction data at a high-level, and those that work at the individual level to get people to log their locations or interactions with people.

The latter include apps that effectively help people keep a diary of who they've talked to using manual entry or QR codes, using GPS to keep track of location, and the Bluetooth-signal apps that detect when people are in close proximity to each other and log those as contacts.

Every solution has its strengths and weaknesses and there is no silver bullet that is an obvious answer to everything. There are technologies being rolled out around the world and we should take what works and apply it in the NZ context, not just copy-paste Singapore or Germany.

A key thing we need to know is what level of coverage is needed to provide effective contact tracing for the whole pop. Manual contact tracing is not 100%, cellphone data might be 90-95% but a huge privacy cost. If we tell people to download an app, maybe 20-30% will participate?

A study based on a model from Oxford earlier this week suggests that speed is much more important, and getting contact tracing time down from 3 days to a few hours significantly increases our chances of containing an epidemic, even with only partial uptake.

And I think that's all I got into the interview... but the government has some difficult decisions around this. Do we need to track location, or just interactions with people, or both? Are we just using this for contact tracing or also enforcing physical distancing?

What solution would best complement the manual processes? How quickly can we get a system deployed that actually works without too many errors? If we have a solution that relies on smartphones, who will miss out?

Then there's privacy, and we need to ask that if we do need to effectively surveil everyone, how can we mitigate the harms? Who has access to these systems, and when do we shut them down?

And we need to do all of this in a way that keeps up morale and trust in government, in order to encourage compliance and participation. It's definitely not easy!

So folks like @MrCraigSquire and @Tom_Bcgh have made the very good point that we need a range of voices. We need epidemiologists to understand what impact improved contact tracing has. We need the experiences of manual contact tracers and how this might help them (or not).

and then from a technical perspective we need to understand what is actually achievable in the short development time frame, what solution will be most reliable without too many bugs and errors

and we need to integrate the perspectives of end users and understand whether they have access to smartphones and data, and what actions people are willing to take. This doesn't have to take a long time, and groups like Māori and the elderly need to be included

and we have to trade-off between public health and privacy. Just because these are extraordinary circumstances doesn't mean we immediately suspend all civil rights. An inter-disciplinary discussion is needed to help us assess the options and find the right balance for NZ.

Okay so what's already happening? There is a contact tracing team in MOH and the COVID taskforce is actively thinking about the different options. Minister for Health said they are "very actively" considering options at the moment.

I think given we've gone into lockdown, and in essence that buys some time, we shouldn't squander it but we also shouldn't jump at the first shiny thing that /might/ work. We can and should evaluate all options (including sticking with manual contact tracing) for the NZ context.

But I would suggest that it's not a bad thing to start socialising the idea that people might need to download apps to participate in the fight against COVID-19, or that more surveillance may be necessary to effectively contain the epidemic. /FIN