Systems Models

COVID Lockdown Mechanics

Disclaimer: I am not an epidemiologist. I do have a pretty good grasp of the behavior of complex systems in a variety of fields. FWIW, I would love to be wrong in this analysis.

I wasn’t planning to start writing posts for my blog quite so soon, but COVID-19 changed that. I’m not yet in a government-mandated lockdown, but I fall close enough to the high risk profiles that my wife and I are entering it ourselves. I expect it will be mandated in Atlanta sometime in the next week or so.

This raises a lot of questions: Why should we go into lock down? How long will it last? Why will it work (if it does)? How will we know when we can emerge?

The answers are not too hard … but they aren’t fun to contemplate. The key elements are captured in the following conceptual graph. To understand the dynamics, we need to examine what happens at the numbered points in more detail.

Before we start, recognize that COVID will have been circulating in our community for some time. It takes one to two weeks for a person to move from infection to getting confirmed results from a reliable test. This lag time is a hugely important factor in the spread of the virus. The exact length of time will vary, depending on how quickly the virus emerges and how long it takes for an infected individual to get the test.

We can shorten the lag time by tracking people who might have had contact. We don’t need to wait for them to show symptoms before administering a test.

When the spread has begun, society agrees to a lockdown (1). This should start to slow the rate of infection, but the numbers of confirmed cases will continue to rise because people are already infected (2) and for a week or two, those are the cases that will show up in tests.

After a couple of weeks (3), we should start to see a slight slowdown in the rise of new confirmed cases (4) to reflect the lockdown’s impact on infections a couple of weeks earlier. Ultimately, the rate of new confirmed cases will begin to slow dramatically compared to what might have happened (5) … and we will have “flattened the curve”.

The key factor driving the timing is the lag between infection and test confirmation. With the current lag time, I can’t see us emerging from lockdown in much less than 5 to 6 weeks. If someone told me it might be 12 weeks or more, I wouldn’t care to argue.

How will it end?

It’s not at all clear how the lockdowns will end. If we stop when the new case rate drops, contagion will resume and we may have to repeat the whole exercise. This article gives the most complete summary I’ve seen … and it is sobering.

China, Japan and South Korea have faced down the COVID monster. They did lockdowns, but they backed them up with aggressive quarantining and massive public testing. Yet, even they can’t rest on their laurels. From what I have read, there are only three ways to put COVID to bed:

  • Defeat it with a vaccine
  • Suffer until we have enough recovered individuals that we gain herd immunity.
  • Develop therapeutics that are strong enough to protect everyone from the serious complications … then transition to herd immunity.

To me, the most plausible exit scenario is a sequence that starts with a) a strong lockdown, followed by b) therapeutics that permit some relaxation, followed by c) vaccine(s) that allow us to move on.

Unfortunately, that looks like it might take a while.

The Psychological Challenge

For many people, the development pattern in the diagram will make it harder to weather the lockdown. I predict they’ll be OK in the first week. Then, the fact that confirmed cases continue to rise will be dispiriting. At the end of two weeks, I expect a lot of people will yell “are we there yet?”. The confirmed case increases will just barely start to slow and, for some people, the temptation to break quarantine will be become irresistable.

More testing will help … a lot

The challenge for lockdown psychology is the time between feeling the pain (lockdown) and seeing the gain (a slowdown in confirmed cases). Given the biological timeline of the virus, we probably can’t do much about that in absolute terms. We might, however, help people believe sooner that the process is truly working. The key is to massively increase our rate of testing.

The dominant protocol for US COVID testing is to pre-screen the patient and only do a COVID test if the patient shows a plausible link to the disease: if they travelled recently, if they had contact with someone who was infected, or if they have hospital-worthy symptoms. By the time a patient qualifies for the test, the destructive lag time is baked in.

Suppose we did massive testing: hundreds of thousands per week … even millions per week. We could give the test to a general cross-section of America, not just suspected victims. We could focus on individuals in jobs with more personal contact … the canaries in the mine. We might catch some people much closer to the time they were infected. The results wouldn’t be complete, but they would constitute a trackable picture of progress.

We could report the statistics on the “estimated current infection rate” instead of the confirmed case rate. That should show movement much sooner after the start of a lockdown. That would encourage citizens to believe that their sacrifice was working. That, in turn, might help them maintain the discipline to hold on a bit longer and bit tighter.

Bottom line: If we don’t start mass testing immediately, America will have to live through the lockdowns purely on guts, faith and grit. It remains to be seen whether we can do it.

PS – On March 13, 2020, Jack Ma (China’s richest man and owner of Ali Baba) announced that he will send 1 million face masks and 500,000 COVID-19 test kits to the US as a gift. Will America have to brains to accept his offer and put those test kits to use immediately. We don’t have to use them as our go-to medical evaluation, but they would be invaluable for the mass testing that we should be doing to map the spread of the disease.

PSS – A strong comment on US testing failures by WHO Director General.