Data Collection and Retention

In Australia, the contact tracers get the data and are subject to non-disclosure laws, which is okay, I guess. Why not trace the tracers? Give them the data they need, have it automatically delete? The state and territory health officials can take their own copy and basically do with it as they will. I view this as a giant data sieve.

I don't think that any studies should be done on data that is collected through mandatory means. Universities have ethics boards that review how studies on people are done. All of the studies, first and foremost, have to allow the user to opt out at any time and have their data removed. If this data is mandated specifically to maximize the speed and reach of contact tracing, then it cannot be used for any other analysis of the pandemic. The ethics of mass surveillance would forbid it, yet we're doing it out of respect for public health.

Who has access to this data? Is contact tracing a good source for analysis of the pandemic? or should that data be acquired in another manner to maximize trust and participation in tracing? Is aggregated data risky to the individual in the hands of public health officials? might that risk simply be trustworthiness?

No, they are not auditted and they certainly can be a leaky sieve of data, but there are subject to disclosure rules and laws like HIPAA in the US. So there is some governance, but your point is exactly valid. If we are going to trace people? Shouldn't we all trace the tracers?

Do we know of any audits that occur for traditional "human contact tracers"? I would have assumed that was already covered, and we were covering the technical side of tracing using a phone app. Because if we have to fill that niche as well? This becomes so much harder.

So Australia for example, which is collecting Postcodes has committed publicly to NOT use the app to enforce social distancing or quarantine, but they could... because they have the ability to do so. COVIDSAFE, is a high human touch approach where positive contacts are referred to human contact tracers (who can build their own data files - YUCK), but these human contact tracers are applying their skills to at-risk age groups an zip codes (hot spots) first. It is a choice they made, using a centralized governmental approach, with a human touch. DP^3T on the other hand, has no human touch, whatsoever, and is extremely limited on data collection (no location for example) and basically is super privacy-friendly. Our challenge is how to help both achieve best results and still advocate for true best practices

I agree - it's outside the scope. All you're going to get is where people are congregating in a confined space, and what can health authorities learn from that? People will either listen to recommendations or they won't, which falls under risk communication and social sciences.

Location data might provide the ability for special cleaning, or locational quarantine. Do you see those as outside the scope of contact tracing then? One of the mute points, in many of what I the different contact tracing systems, is the use of aggregated data by health authorities to "reach conclusions and learn". I am concerned about this mission creep for contact tracing.

I don't think location data is relevant. Contact tracing by definition is who did I come into contact with, not where was I probably infected? We know how the infection spreads: through prolonged close contact with an infected individual, or through contact with high traffic surfaces before washing hands. The only way to determine a true hotspot would be to prove that a person was infected at that location. Otherwise we're just guessing based on probabilities, which could cause more harm than good.

If phone numbers are used, is any other data required besides the details of the "transaction"? Is the location of the "transaction" relevant? Should contact tracing be tracing people AND hot locations? Are hot locations even MORE valuable as a method to alert the public without creating hypochondria and increase usage?