As the coronavirus pandemic is sweeping across the globe, most of the United States and many places all over the world are trying to significantly restrict people's movement through "shelter-in-place" or more severe orders. The goal is to "flatten the curve" which means to slow the spread of the disease in order to avoid overwhelming the health care system which would lead to massive casualties from COVID19 as well as preventable deaths from other causes due to hospitals being unable to treat incoming patients.
To avoid this doomsday scenario of hundreds of thousands of people dying at home without proper care, we have to keep these control measures in place until either of the following happens:
- The virus has infected a large enough percentage of population who have recovered that we effectively have herd immunity, which means there are enough people who are or have become immune to the virus that it runs out of hosts to invade and dies out.
- We develop a vaccine against SARS-CoV-2 which works and doesn't have crazy side effects, i.e. it makes it through clinical trials successfully. Then we vaccinate a large enough part of the population to get herd immunity.
Flattening the curve means that option 1 will take a long time because we are trying to limit the spread. How long exactly depends on how many people are already immune, what the hospitalization rate is exactly and a number of other factors. But even optimistic models suggest that it will be at least a year. That's also how long it will take to get a vaccine in the best case.
So, realistically, we would have to keep up with the current control measures for at least a year to avoid overwhelming the health care system and the massive casualties that would result in.
However, restricting movement and closing many businesses for that reason already has a devastating impact on the economy after only a few weeks. If we keep this up for a whole year the economic devastation is likely to be extremely severe and result in a great depression. This will have massive negative impact on people's livelihoods and health as people loose their jobs, their businesses, and their health insurance. Not to mention the psychological and emotional toll we will incur by isolating people for such a prolonged period of time.
South Korea and Singapore have demonstrated that there is an alternative path out of this dilemma: Use severe control measures for a few months until we understand the spread of the infection well enough to contain it via widespread testing, tracing, and isolation of affected individuals only. That would allow us to resume mostly normal economic operations and social functions with some caveats (e.g. wearing face masks, avoiding unnecessary body contact, being more diligent about washing hands and cleaning surface areas, etc) and some limitations (e.g. very large gatherings of people like football games would likely still be too risky).
However, while we are ramping up testing capacity to allow us to implement this model, our current approach to tracing infections to identify people who are at risk and need to be tested or isolated does not scale to a population of hundreds of millions of people. Manually collecting such data and using people's memory to recall whom they have been in contact with is too laborious and error prone.
We need a technical solution to this problem. Most everybody nowadays carries a smartphone or similar electronic device on them at all times. We can use those devices to keep track of whom we have come in contact with and automatically record this data. That would solve the "faulty memory" problem and make it simpler to identify who may have been exposed and needs to be tested. Singapore has announced that they will open-source the application they wrote to do exactly this via bluetooth (i.e. detect which other devices are in close proximity via bluetooth signal). In addition, we need a database to collect all this data and build a contact graph across the entire population so that we can easily determine who is at risk when somebody tests positive or shows symptoms. This would enable authorities to implement effective containment measures that are good enough to allow us to relax the current control measures without causing massive harm.
The prospect of a central database that collects information on who was in contact with whom is scary because it invades our privacy. The goal of this project is to demonstrate that we can implement effective contact tracing with a minimal amount of personal identifiable information and that we can do it at a scale that allows us to roll this out across the population. Developing this an an open-source project allows us to come together and get this done as quickly as possible while providing the greatest degree of transparency.