Safe Blues — The Measurement Tool for Responsible Easing of Social Distancing

Six weeks ago, a few academic colleagues and I decided to join the battle against COVID-19. Armed with mathematics, statistics, programming skills, and a general desire for fighting the virus, we started to think where we can be useful. After all, as you are probably aware, a key part of the battle against COVID-19 relies on data, measurements, modeling, and computations.

One thing that we quickly realized was that COVID-19 itself was not the only enemy. The economic and social effects of social distancing are also dangerous. This doesn’t mean that social distancing measures shouldn’t be imposed. However it means that the battle against COVID-19 also includes the goal of striking the right balance between public health and economic health. This is especially important in the end game where successful governments will be able to ease their social distancing measures with grace, while others not.

With that, we weren’t immediately sure exactly how we can be useful. So for about two weeks we postulated about different directions. We analyzed basic data, and we discussed where and how we can perhaps help. It wasn’t clear to us from onset that we could add something to the table that wasn’t there already. After all, public health professionals have been preparing for such a Pandemic forever.

We even allowed for the possibility that we won’t do anything at all. This is because we did not want to overload the busy systems with additional distractions that we may cause. However, things turned out differently. After much thought and analysis, an idea emerged and developed. With further discussions and research, we gained more confidence in its strength. Then after a while it even earned a name. We called it Safe Blues.

We are now 6 weeks later, with a 30 page, 8 author, paper on our Safe Blues website. It is framework ready to be deployed in contact tracing apps, or independently. It is now up to governments to choose if to employ such an idea.

Frankly, after this quick and intense journey, our team members strongly believe that Safe Blues is nothing short of a killer idea. We believe that governments that will adopt frameworks similar to the Safe Blues idea, will be able to do very well in the end game dealing with COVID-19. They will be able to decide on social distancing directives in a much more effective way than what they can today. Real time, privacy preserving measurements will guide them. Safe Blues provides such measurements.

Before I tell you how Safe Blues works, let’s review a key problem that governments are facing when trying to release social distancing measures:

At any given point in time, there is hardly any knowledge about how many new infections occurred in the past two weeks.

So if you are government easing social distancing measures, the only responsible thing that you can do is ease them off extremely slowly. One step at a time. Anything else will imply that you may turn off social distancing too quickly and the spread of the virus will go crazy before you know it. But what is “too quickly?”. And what about “too slowly?”. Without Safe Blues one can’t tell. But with Safe Blues the picture is different.

Safe Blues measures near real time population response to government directives. This allows to adjust directives on the go.

The Safe Blues philosophy is as follows: In an imaginary world, we would develop a harmless biological virus that spreads just like COVID-19 but is traceable via cheap and reliable diagnosis. Then by spreading such an imaginary virus throughout the population, the spread of COVID-19 could be estimated because the benign virus would respond to population behaviour and social distancing measures in a similar manner to COVID-19. However, such a benign biological virus does not exist. Instead, Safe Blues is a privacy-preserving digital alternative.

The basic idea is that mobile devices pass tokens via Bluetooth in a similar way to how a disease spreads. However in contrast to the actual disease which can be lethal and is hard to measure. Safe Blues is measured instantly and does not cause any harm.

At this point you may have many questions. Here are some main questions that you may ask:

  1. What about privacy?
  2. What if only a small fraction of the population uses Safe Blues?
  3. How is it related to contact tracing apps?

I’ll focus on the answers to these questions in my next post about Safe Blues. But for now I’ll just say that Safe Blues does not have the same ambitious data goal as contact tracing apps.

A contact tracing app needs to locate individual people in case they may have been infected with COVID-19. Safe Blues only deals with aggregate data.

Hence by design, Safe Blues is much more privacy preserving than contact tracing apps. It also requires fewer members of the population to participate. You can read more about Safe Blues in our website.