In April 2020, our team published an article on our old blog, Orion-society.com, advocating for a mass testing strategy. Originally it was titled ‘lower sensitivity tests’, which was a mistake and we should’ve thought harder and highlighted the positive of rapid speed and scalability, but the key debate point was the sensitivity issue and we felt we had to show why this was a red herring.
As published April 2020:
Here is a paper by Gergo and Gaurav concerning how to potentially end lockdown with proposals along the lines of the Peto proposal. They originally posted it at Gergo’s GitHub – https://gbohner.github.io/coexist/ Here is the pdf of the full paperDownload
The UK has just announced a test-and-trace strategy to end the lockdown.
How many tests, and of which type, are sufficient to end the lockdown safely?
Large scale testing strategies have universally relied on RT-PCR tests, which are exquisitely sensitive. When performed perfectly, these tests are capable of detecting tens of viral RNA molecules in a given patient sample.
The UK has had enormous difficulty scaling this test. Modifications of RT-PCR which make the test easier to scale now exist – including pooling samples, skipping the RNA extraction step, and collecting samples with spit rather than swabs. To our knowledge, tests with these modifications are not being deployed in the UK.
The easiest tests to scale are likely ‘antigen’ tests. These tests detect the presence of viral protein rather than RNA, and can be performed at the point of care using lateral-flow-assays, the same technology that is used in home pregnancy tests. The tests can therefore be deployed at scale without the construction and organizational overhead of large centralized testing facilities.
There is concern that antigen tests and modifications of RT-PCR will be less sensitive than the tried-and-true RT-PCR test. Is this concern justified?
It is important to first note that competing tests must be compared with RT-PCR as deployed in practice; due to handling errors and RNA degradation, RT-PCR tests have been observed to have a relatively high false negative rate in the clinic. Assuming that there is a tradeoff between sensitivity and scalability, which kind of tests will get us out of lockdown safely?
We find that the number of daily tests carried out is much more important than their sensitivity, for the success of a case-isolation based strategy.
Our results are based on a Susceptible-Exposed-Infectious-Recovered (SEIR) model, which is age-, testing-, quarantine- and hospitalisation-aware. This model has a number of parameters which we estimate from best-available UK data. We run the model with variations of these parameters – each of which represents a possible present state of circumstances in the UK – in order to test the robustness of our conclusion.
We implemented and investigated a number of potential exit strategies, focusing primarily on the effects of virus-testing based case isolation.
The implementation of our model is flexible and extensively commented, allowing us and others to investigate new policy ideas in a timely manner; we next aim to investigate the optimal use of the highly imperfect antibody tests that the United Kingdom already possesses in large numbers.
There are a number of heterogeneities that our model does not capture. Most important among these may be the effect of exposure level on disease progression. There is evidence that the severity of COVID-19 correlates with the exposure level to SARS-CoV-2; this may significantly impact the effect of home quarantine policies on the spread of severe disease. Additionally, our model does not account for the compliance rate of a given government policy. It is possible that the use of a relatively lower-accuracy test will lead to low compliance with home quarantine instructions.