thoughts by vr

Toward the End of Shelter in Place

May 06, 2020

In the last couple months, the world has mobilized itself in incredible ways to combat COVID-19. Governments have enacted policies that were politically impossible months ago. Companies have rebuilt entire supply chains to meet unprecedented needs. Even Google and Apple, fierce competitors in different times, are working together on a novel solution to contact tracing.

While shelter in place policies in particular have been effective in limiting the spread of the virus, the economic hardship these measures created has been devestating. Every week the economy is closed, more people are laid off and financially squeezed. In response, governors are laying out plans to gradually reopen their state economies. While each plan is different, their shared goal is balancing the efficacy of shelter in place with the pressure it puts on the economy.

The primary way these plans will be implemented is on a per industry basis. Industries that can maintain social distancing in the workplace, like manufacturing, will reopen first. While places like sports arenas, for example, will reopen later. By gradually easing restrictions, the healthcare system will be better positioned to meet demand. California’s plan will see businesses like bookstores and florists open in the coming days.

Given the tools that we have to work with, these proposals can make sense. They apply economic relief to industries that do not pose as high a risk at a time when public health officials are gaining confidence in their ability to handle a surge in cases. That being said, they do leave much to be desired. First, many healthy people in high-risk industries will be unable to return to their jobs for a long time—possibly many months. Second, the virus does not discriminate based on a person’s job, so infected people will still be able to spread the disease even if they work in lower-risk industries.

Fortunately, there has been an outpouring of ideas coming from a diverse group of scientists, policy makers, and technologists to make reopening the economy more efficient. One such solution that caught my eye was documented in a recent report in the WSJ.

Daily Certification of Symptoms All employees and students must certify (via smartphone app), before leaving home, that they are not experiencing enough of the following COVID-19 symptoms to exceed a calculated risk, weighted by symptom frequency, of being infected with SARS-CoV-2.

This certification should detect the vast majority of symptomatic cases, including mildly symptomatic ones, among those who accurately respond. None of these individual symptoms are specific to COVID-19, but in aggregate they can be used to assess an individual’s risk of being infected with SARS-CoV-2, and even if caused by other pathogens are a prudent basis for staying at home. The acceptable level of calculated risk may differ among occupations (for example, nursing home caregivers could be subject to a very low risk threshold).

Since COVID-19 tests are still in short supply, using known symptoms is one of the best ways to estimate the likelihood of infection at a large scale. Realtime symptom data of a population will enable policymakers and public health officials to make more informed decisions on a daily basis. They can restrict the interactions of people who present the highest risk of being infected while accelerating the reopening the economy. Further, an app gives authorities fine-grained control in the easing of restrictions. As conditions on the ground change, the app can be updated daily to adjust to the current risks.

This solution is no panacea. Its primary purpose is to serve as an efficient way to reintroduce people back into the economy in a speedy and safe way, but the work being done to increase testing capacity, develop contact tracing, procure PPE, and manage hospital load is still imperative. Since a large number of cases are asymptomatic and the virus can be transferred before symptoms are manifested, anything reliant on experienced symptoms can only be a part of the puzzle.

A targeted rollout will enable us to bring back segments of the economy that can save the most jobs while controlling for public health risk. Anything less forces us to tradeoff safety for economic recovery. By gathering data daily via a widely distributed app, we can make deliberate decisions backed by the ground realities.

So far the response to COVID-19 has been marred by a lack of preparedness. I hope that with tools like this we can change that story.


You can find the discussion thread on Twitter!