What happens after lockdowns diminish Covid-19? What next?

Empty shelves during the Covid-19 panic

We're all sitting in lockdown in much of the world. Lockdown should reduce the famous R0 number for the virus to less than one -- which means that each patient who gets it infects less than one person. An R0 of more than 1 means a virus spreads. More than 2 means it spreads like wildfire. Less than one means it dies out. Much less than one means it dies out fast. That's what they did in Wuhan.

There's a problem not being spoken of as much, namely what happens then. Unless you wipe out the disease completely in a region, once you end the lockdown measures, it will come back, and you need to lockdown again. This cycle of waves has been called "The hammer and the dance" and it's better than the horror of unchecked growth, but it's a very long slog.

If you can wipe it out completely in a town, you can open up again, but not to the outside world. You need to build a border around your virus-free-zone and don't allow anybody in without highly accurate testing and isolation. If your region is an island or country, this may be doable. Even so, air travel becomes highly minimal until the vaccine.

Completely wiping it out is hard, though. Tests have false negatives (the test says you are clear when you're not.) People hide and cheat if there are economic consequences for testing positive.

A treatment

One possible savior would be a treatment. If some treatment arises that cuts the rate of hospitalization and death way down, the virus can be tolerated like flu (which is pretty nasty) or other diseases with lower death rates. Lots of research is going on into antivirals of various sorts.

A working treatment could bring the total death rate (as percentage of the population, not just the infected) below that of flu or other diseases we tolerate.

Donald Trump's advisors did not just pull Chloroquine out of thin air for him to promote. It was being tested in China back in January and many other tests are underway. The doses used are risky, so that's why more testing is needed before wide deployment. There are also those attempting to find out why Chloroquine inhibits the virus. In one model, it acts in a fairly simple way by allowing zinc ions to enter cells and block RNA replication. If this is true, fortunately there are many, much safer, substances which do that.

Of course, the world will also drive like crazy for a vaccine. Many are searching for ways to reduce the vaccine timeline, including testing existing vaccines for similar substances, and considering manufacturing a promising vaccine before it is proven, at the risk that if it fails later tests this money is wasted, but if it passes, the value is immense.

Better, faster testing. Everybody. Every day or two.

Another saviour is better, cheaper, faster testing. Better means both fewer false negatives, and detection very quickly after infection, and in particular not long into (or before) the infectious asymptomatic period.

If the virus is at low levels, say 1 in 10,000, testing can be made cheap using Group Testing, a technique where you mix up swabs from a group of say, 256 people, and if you get a positive, breaking them up into 2 halves and repeating the test, and so on until you know who's positive and who's clear. To do this you need a test with a low false negative rate, though in some cases you can lower that rate by just repeating the test multiple times. That may mean taking lots of swabs (which is cheap) but the test still is done far less than once per person.

Due to false negatives, tests would be taken multiple times until the accuracy level was sufficient.

Tests would be available every day. Carts would roam the streets collecting test samples (multiple swabs from each person) and issuing/updating "clear" cards for people who passed earlier tests. Cards would be good for only a couple of days, and would be needed if you want to go places where you will be in proximity with other people, like work, or flights.

Some places, like international flights, might require multiple tests plus a short isolation period. Services would crop up to document your isolation period.

Due to false negatives and incubation periods, some people with infection would pass the test. Some virus would spread. But the spread would be much smaller, and with contact tracing, detected quickly and stamped out quickly.

You would get a wallet card with your photo and an electronic tag regularly updated with the date of your last test and to show it's authentic. This may not be needed if card fraud is not a problem. Card fraud (giving a clean test card to an untested or infected person) would have harsh punishments. The auth tags would not be remembered by any readers, under penalty of imprisonment.

Your name or other information would not be on the card, just your face and the auth tag.

Testing of the recovered/immune.

Tests for antibodies -- which reveal you had the virus but now are immune, can also play a central role. Those who pass such tests should (unless against all expectations that recovery does not give immunity) be able to fully travel, and participate in society, and keep things moving. You would get a permanent card.

Privacy-protecting contact tracing

Android and iOS would be updated to support an emergency contact tracing mode, which would stay on until June, and then be disabled unless renewed by emergency law. The protocol would be designed to be privacy protecting. Security experts are working on such protocols. An example of a possible protocol would be to have each phone transmit a different seemingly random token by broadcast Bluetooth every so often. Other phones would pick up and record these tokens. There would be no way to correlate them.

If you were found to be infected, you would publish your list of tokens to a database. Other phones could check this database from time to time to see if they have stored any of them, indicating they were close to the infected person. (Signal strength would be recorded with the token.) Having several tokens would indicate you were in proximity for more than a brief time. (You don't want to trigger if you just passed somebody on the sidewalk.) If so, you would isolate and make sure to do testing. If you had symptoms you would publish provisionally, then fully (or revoking) after getting results.

The OS update would be applied to as many old phones as possible. Those old phones (sitting in closets mostly) would be updated and given to those who don't carry a smartphone. Even if you never use the phone, you would carry it when going outside because it could save your life, by helping you know as fast as possible that somebody you had close contact with was infectious.

(It should be noted that GPS tracklogs, which most cell phone users already keep in things like Google location history, are not adequate. They only tell if you were in the same rough area which would cause way too many false positives and panic, and they have big privacy concerns. They also only sample at long intervals. You need to know whether you just walked in the same building as somebody or if you sat down for a chat.)

Isolation of only the at-risk

This proposal is disturbing, and not just because I'm at-risk. As it becomes clear that the fatality rate is low (but not zero) for people who are young and have none of the comorbidities (including, but not limited to diabetes, hypertension, heart disease, lung problems, etc.) it is not out of the question to end isolation for the healthy under-60s, but continue it for the at-risk. This would require that the non-isolated stay away from the at-risk. If they live together, they must live in isolation, or move out. Almost all children would be low-risk, but the parents of older children might not be, creating a problem.

This would restart the economy but cause effectively massive age discrimination in the world and workplace. Laws would need to be put in place to protect those who could not work because of isolation, assuring that they can't lose their jobs or be passed up for promotion without extremely well documented and extreme reasons. Of course, the isolated would continue to work from home, and also guide and supervise the more junior people via video and augmented reality "remote assist" functions.

The law would need to be extreme. It would be almost impossible to fire or lay off people who isolated for a period of time, to assure it never happens that somebody ends a career due to the virus. As we know, there are laws to protect people who go on family leave, and they aren't strong enough.

Research into immunity through small doses

It is currently being investigated how much the size of your dose affects the severity of the disease. It is speculated that it may be possible to receive a small dose, through the skin (ie, not in the lungs) and take low risk, but gain immunity. Volunteers could participate in trials to attempt to learn if it's true and what the small dose is. Volunteers would be young and healthy with no known health problems. If they get immunity, they would be rewarded with a card. If they get sick, they will be promised top grade medical care and ventilator priority, but if people get sick the trials would get stopped. Also tested would be the use of small doses and possible treatment candidates to see if they can make the risk of complications from the inoculation dose extremely small.

This sort of research is the sort the medical community will not want to do, under the "first, do no harm" doctrine. On the other hand, many healthy people (who already have a very low chance of complications) might well be quite willing to do it for the chance at immunity.

Economist Robin Hanson has been proposing deliberate infection to flatten the curve for some time, though it is unlikely anybody in the high risks groups would be willing to take that risk (even if the trade-off is "get it early" vs. "maybe get it or not later but when they are out of ICU beds." However, he has switched to promoting low dose "variolation" which was used a a precursor to vaccines.


There are tens of thousands of tests already taking place. Effectively everyone that gets COVID is a sample based on their risk factors and the drugs they are already taking. Hopefully, enough records are being kept that this data can be analyzed to look for correlations.

For example, how many people with Lupus or Sever Rheumatism, taking Cholorquine get COVID compared to people with similar risks factors.

But medical research is not used to working at this speed. It's learning. For example, it is known that hypertension is a risk factor. Most hypertensives take a variety of blood pressure medicines. Is it the medicines that are the cause of the risk? Is it the hypertension even though the medications are keeping the pressure normal? Data would tell us that immediately if you had a database of all patients and their conditions and drugs.

I'm pretty sure we're going to do a bare minimum of all of this, but mostly none of this. Treatment will get better, but probably not much in the near term. Testing will get better, but not good enough to test everybody (and there's no way the governments here in the USA are going to issue cards). Maybe you'll get a test before going to a football stadium, or going into a theme park, but I'm not sure if rapid tests will be that plentiful. Antibody testing will get better, but it'll mainly be used for contact tracing and development of treatments (there aren't enough people who have antibodies to restrict those without them). Contact tracing will get better, but there's not going to be an app for that. Many of those most at-risk will choose to stay away from high-traffic locations, but many won't, and most governments won't mandate it.

Things will open back up, though they won't be the same as they were before. Not until we have a widely deployed vaccine, and maybe not even then. It's hard to say how much change people are going to be willing to accept, though.

In a year or so, we'll have a vaccine that is ready for mass-deployment, most people in wealthy countries will get vaccinated, we'll get herd immunity in those wealthy countries, and the risks will drop significantly.

If 300,000 people in the USA die in a year, and it'll probably be less than that, that's 0.01% of the population. Less than heart disease. Less than cancer. As long as we can keep the virus from getting out of control again, which we should be able to do once we have good testing (for the virus and for the antibodies), it's going to be an acceptable risk.

But the card as I propose it would not be an ID card. It would not have your name or address, just your picture, and a certification you tested clean on a given date. The certification would not be tied to you in databases, though people reading cards could start doing it if they know who you are (or if they store the photo.)

Alternately, instead of a card, there could just be a database that says, "Person of this name and address is clean" and it's referring to the name and address (or if you prefer ID number) on an ID card you already have. That means the people reading the cards (to let you in to places) need to be able to read that database, though.

Sure, 300,000 people is 0.1% of the population (not 0.01% -- off by an order of magnitude there.) But that's how we are. There is not much we as individuals can do to prevent those larger numbers of deaths from cancer and heart attack. If you said, "Hey, if everybody locks down for 2 months, we can prevent cancer" you can bet people would do it.

Your card idea might be a good one. Except for the last paragraph, my message above was a prediction about what I think is going to happen, not a discussion of what should happen. If the USA were a dictatorship run by technologists, maybe it would happen. Maybe something like it will happen in other countries. I can't imagine it happening in the USA. Ditto with the database. Maybe these are good ideas, but they're not going to happen. Not in the next year, anyway.

Many of the risks we take with respect to cancer and heart disease are preventable. The CDC says that 34% of deaths from heart disease and 33% of deaths from stroke could have been prevented or delayed through changes in health habits. Many deaths from cancer can be prevented, yet people still smoke, and people still delay or ignore screening. And that's just rational things that people can do to reduce their risks of dying. If were going to do the irrational, and never go out of our houses except when absolutely necessary, we could prevent many more deaths (at least in the short term; in the long term the horrible effect on people's mental health would probably result in lower productivity which would, very indirectly, cause deaths). There were 38,000 deaths in 2019 from car crashes, and countless others (many of them cancer and heart disease) caused by car pollution, but people keep driving.

No, I don't think everybody would lock down for 2 months to prevent cancer. I think that'd be a losing bet. More to the point, I don't think we'd lock things down to the extent that things are locked down right now, for a year and a half, to prevent cancer for a year and a half. Nor should we.

We locked things down to the extent that things are locked down right now to prevent millions of deaths. But once we can bring the numbers down to more reasonable ones, we're going to open the country back up, and we should open the country back up.

You're right that I messed up my percentage. Of course, I also overestimated the risk of a person dying from COVID-19 in the next year, if they haven't already gotten it, by about an order of magnitude. My conclusion was still right. People are not going to (and should not) lock themselves in their houses until a vaccine is made. There's going to be some risk until that happens. That's how life works. If you want to lock yourself in your house until a vaccine is made, that is your choice (assuming you can find work from home or afford to live on your savings, anyway). I'm not going to, and the vast majority of the country is not going to. My mental health is important to me just like my physical health is important to me, and this life of social isolation is not at all good for it.

So far it looks like I was mostly right. I thought antibody rates would be somewhat lower, but I did correctly predict that there wouldn’t be any special privileges given to people with immunities. Treatment has gotten better, but not tremendously better. There’s more testing, but we’re nowhere near testing anyone, and have no plans to ever test everyone. There are no cards. There are apps, but I don’t think many people use them. There were lockdowns, but now things are fairly open. 300,000 deaths may or may not be close to the final death toll. IHME is projecting 220,000 through October, but that’s not the projected end of the virus, it’s just where the projection ends.

In hindsight, it looks like there never would have been millions of deaths. Sweden never locked down at all, and they saw 555 deaths per million. So I was wrong there, I guess.

If I lived in China, and I had a chance to move to the USA, but I’d have to risk a 0.1% chance of dying, I’d take it. Maybe some others wouldn’t, but unlike China, you’re allowed to leave the USA, so if you live here and don’t love freedom, don’t let the door hit you in the ass on your way out.

Yup, though people keep proposing them. And debating how long immunity lasts.

Sweden does not have an open society. They just don't have legal lockdowns on some things that have legal lockdowns in other places. And I sometimes wonder if the lockdowns in the USA are being followed any more than the swedish voluntary restrictions are.

As to what was possible? Well, Bergamo, which had 0.58% population fatality after locking down late, suggests that a total "do nothing" approach would indeed result in 1-2 million dead.

What is missing from that analysis is that you don't need laws or government action to reduce spread. Fear does quite a lot, though fear needs scary stories in the media. Bergamo resulted from people having no fear. So while it is I think, still true to say, "If we do nothing, about 0.7% death will hapen" it is not possible to do nothing.

I think your last sentence explains it all about "doing nothing." By the time I wrote what I wrote, it was clear that we, the public, weren't going to "do nothing."

With that said, I don't think we would have seen anywhere near the number of deaths in Bergamo even if it were possible to "do nothing." The population density of Bergamo alone greatly exceeds that of most of the USA, and I hope you agree that population density matters a lot.

My recent comment was that I don't think we would have seen millions of deaths in the USA had we not locked down. In fact, I think we ultimately might have seen fewer deaths had we gone that route, though time will tell (we haven't even reached herd immunity yet in many states). As you point out, Sweden's approach may have resulted in better compliance than that of the USA.

So of the things I said back in the beginning of April, I think the most wrong one was that I suggested that not locking down would have resulted in millions of deaths. I think I was wrong to have bought into that.

Still looks like my predictions have mostly held up. Here's something I got wrong:

most people in wealthy countries will get vaccinated, we'll get herd immunity in those wealthy countries,

The first half was mostly right, though I thought it'd be a bigger majority.

The second half was wrong. We haven't reached herd immunity here in the USA, and we might never reach it. It might even be impossible, in that even if you vaccinated everyone, the effectiveness against asymptomatic spread (which might be in the low 70% range) would still not be high enough to achieve herd immunity.

In hindsight, I probably should have realized that.

Well, I don't know that we won't reach it. After all, there is no dangerous disease, by definition, that doesn't either stay isolated (like Ebola) or reach herd immunity. Otherwise we would be dying from it all the time.

You don't need a lot. If the "base" replication factor is 2, then 50% immunity drops it below 1, and outbreaks self-extinguish. So 70% is just fine up to a factor of 3.

We might never reach it.

Do you consider the flu to be a dangerous disease? The flu doesn't either stay isolated (like Ebola) or reach herd immunity.

COVID is much less deadly than Ebola, especially among the non-elderly. After more mutations it might get even less deadly. And it looks like vaccinated people can, especially after some period of time, still get infected and spread that infection, while being even less likely to die.

Immunity is not 100%, and immunity fades. If you need 70% vaccination to get herd immunity, and vaccines are only 80% effective against contracting even an asymptomatic case of the disease, and the vaccine drops to say 50% effectiveness against asymptomatic cases after one year, then you need 88% of people to get vaccinated in single year to achieve herd immunity. And unless it's 88% of people in the entire world, you'll need to get those 88% vaccinated every single year forever.

Now look, I made a lot of assumptions there. By no means is it a guarantee that we'll never achieve herd immunity. But it's definitely possible, and it's a possibility I failed to account for in my analysis from a year ago that is above.

It's the case with all of the other coronaviruses in existence.

Actually, the Flu is a much more disparate set of diseases, and we do get herd immunity on the individual ones. We have been questing a generic flu vaccine for some time, and it may come eventually.

Immunity may fade, and so boosters are likely.

Yes, the rest of the world is an issue. Some countries may reach herd immunity the hard way, without vaccination, and lose 0.5% to 0.8% of their population (mostly over 50) in the process. But they do reach it. As soon as natural or vaccine immunity goes high enough, spread keeps slowing down, more and more, until outbreaks no longer grow. Before that, they grow slowly and can often be managed by other means, but after a while, you hit R0 < 1 in your area, and outbreaks tend to die out rather than grow.

TLDR: Google "endemic virus"

Actually, the Flu is a much more disparate set of diseases, and we do get herd immunity on the individual ones.

That's not at all accurate.

Immunity may fade, and so boosters are likely.

Immunity does fade. The only question is how rapidly.

Just getting enough people to get their first shot (or two) is hard enough. I'm not sure 85% of the population of the world is going to be willing to get a booster shot every year for the rest of their lives.

Some countries may reach herd immunity the hard way, without vaccination

That's probably not possible, for the reasons I outlined above. Immunity likely fades quicker than you can reach herd immunity. Keep in mind that the virus is seasonal.

after a while, you hit R0 < 1 in your area, and outbreaks tend to die out rather than grow

First of all, it's R that goes to less than 1 when you reach herd immunity, not R0.

But aside from that nitpick, then immunity fades and the next coronavirus season starts, and R is once again above 1.

Again, google what the term "endemic" means. You'll probably see lots of stories saying some form of SARS-CoV-2 is likely going to be with us forever.

Here's one: https://www.nature.com/articles/d41586-021-00396-2

Looking at the flowchart, I think the answer to the question of whether or not immunity fades has been answered: Yes. It fades significantly in about six months to a year. Maybe even sooner when it comes to immunity against mild disease that isn't likely to cause people to get tested. And we also know that vaccines can prevent (or at least greatly reduce the risk of) severe illness. So fortunately that really bad scenario in red seems to be off the table, and there are two possibilities, really, and given how rapidly immunity seems to fade, and the fact that the vaccines don't completely stop transmission, I think the most likely scenario is "the virus will continue to circulate, but once people are protected from severe illness (through vaccination or prior infection), it might cause only mild to moderate disease."

What I mean is that while having a particular strain of flu does give you immunity for that strain for some time, you get no or limited immunity to other strains. At present, the Covid vaccines are providing some protection against all the variants, though that's not assured. Recovery from Covid provides immunity for some of the variants as well, though new data suggests not so much with Delta.

Well yes, it is a question of rapidity but nonetheless there are diseases you only get once in your life. If you didn't age, you would get it again. Sars-Cov2 is not one of these class, probably.

I know there are endemic diseases. What I said was that some diseases do fade away. Whether Sars-cov2 will be with us forever is not yet settled. And the link shows those surveyed as having no consensus, though they leaned towards it being likely to become endemic, though eradicated in some regions.

If R<1, then imported viruses from endemic regions die out. This is the case for a variety of diseases today.

If R is near 1 but now, spread of outbreaks can be often handled. In fact, it can be handled with much nastier diseases like sars-1 and MERS.

There are various techniques which could exist, including group testing of flight passengers from regions where the virus is endemic and required recent boosters.

But I don't deny it's tough. I just came from an outdoor event that had an outbreak regardless and am quarantining. All those involved were vaccinated or recovered, though there were unvaccinated at the the event, as there are everywhere it seems. Delta is going to be a big deal. I am thinking of going up to Canada, where for some reason there is no Delta surge. Yet. With negative test of course.

No argument that the flu is different from COVID. But you made the bold, and utterly incorrect, assertion that "there is no dangerous disease, by definition, that doesn't either stay isolated (like Ebola) or reach herd immunity," and that was what I responding to with regard to the flu.

there are diseases you only get once in your life

Yes, there are. And there are lots that you don't. The diseases caused by all known coronaviruses are an example.

What I said was that some diseases do fade away.

You said that all "dangerous" diseases either fade away or stay isolated. I was questioning what contorted definition of "dangerous" you were using to try to claim that that's true. Although maybe it's also a contorted definition of "disease," as the flu is definitely a disease.

If R<1

If R is near 1

R is constantly changing, and varies across the globe. Thus we get seasonality.

Don't oversimplify things so much by assuming a constant R/R0.

There are various techniques which could exist, including group testing of flight passengers from regions where the virus is endemic and required recent boosters.

I doubt it, but maybe it's possible. But it's not going to happen. For vaccinated individuals, this virus is much less deadly than many others that we've learned to live with.

Delta is going to be a big deal. I am thinking of going up to Canada, where for some reason there is no Delta surge.

Delta is being blamed for much more than it deserves to be blamed for. The virus is surging in the USA primarily because we stopped wearing masks and resumed attending large events.

Delta did contribute. It's maybe 1.5 times as contagious as the variants we had here a few months ago. But the situations people are putting themselves in, even the ones who are unvaccinated, are a much much bigger factor.

Really sucks for those who can't get vaccinated.

I know it's not constant. When I write that if an outbreak occurs, and R is less than 1, I obviously mean at the time of the outbreak. It must however be kept that low almost all the time, because outbreaks will happen at any time. This usually means you need it to average sufficiently below 1 that it rarely creeps above that.

The flu is our one nearly annual pandemic of dangerous disease, but it is not the same one as before (though some last 2 years.) Even the worst of it, the 1918, faded away at immense cost. Each specific flu fades away. Perhaps someday we can get them all (and the coronaviruses as well.) Some coronoviruses, including of course the closest cousin to this one, Sars-1, were contained.

I wish I understood why Texas did not surge until now, even though it stopped all closures and masking in early March. It's having a bad surge now, and so is the highly vaccinated and still masking SFBA. Must of Canada has a bad surge in January even with strong lockdown orders in most counties. We don't have a unified model.

This usually means you need it to average sufficiently below 1 that it rarely creeps above that.

The average is fairly irrelevant, as long as the season is long enough to ramp up the exponential growth.

The flu is our one nearly annual pandemic of dangerous disease

Flu pandemics are rare. https://www.cdc.gov/flu/resource-center/freeresources/graphics/seasonal-vs-pandemic-flu-infographic.htm

Even the worst of it, the 1918, faded away at immense cost.

The 1918 flu is still here. https://www.history.com/news/1918-flu-pandemic-never-ended

Each specific flu fades away.

That's not really accurate. There are a few different flu viruses, but what "fades away" is not the virus, but a particular mutation of the virus.

With that said, the flu is fairly different from COVID-19. SARS-CoV-2, the virus that causes COVID, mutates much more slowly than the flu viruses (like H1N1, the one that caused the 1918 pandemic).

The problem with SARS-CoV-2 is that, like other coronaviruses (most notably HCoV-NL63, HCoV-229E, HCoV-OC43 and HCoV-HKU1), our immunity from it fades rather quickly.

So we can get infected by it year after year. It may be a mild infection, but it's still transmissible, so the virus sticks around, basically forever.

And in fact, there's a theory that HCoV-OC43, one of those four "common cold" coronaviruses above, is actually a descendant of the virus that caused the 1889-1890 "Russian Flu" pandemic. As the theory goes, because humans had never encountered HCoV-OC43 before, we had zero pre-existing immunity. So the virus spread far and wide, causing many deaths. And never went away. It spreads so widely that young kids are highly likely to get it some time growing up, when their immune systems are strong and they can fight it off without a problem. And then we get it over and over again throughout our lives, but because we have partial immunity now it doesn't affect us as much as it would have if we got it for the first time when we're 85 years old.

In all likelihood, this is where SARS-CoV-2 is headed. That's very bad news for people with compromised immune systems, as they probably won't make it many years or decades without eventually contracting it for the first time. But for those of us who can be vaccinated (and choose to do so), and for those who are young and are either vaccinated or acquire immunity through infection, it's probably not going to be too bad.

Some coronoviruses, including of course the closest cousin to this one, Sars-1, were contained.

As was MERS. But those are much less contagious. Which is good because they're also much more deadly.

The four coronaviruses that are now endemic are the ones I listed above. I would speculate that they all caused a pandemic when they first arrived in humans.

I wish I understood why Texas did not surge until now, even though it stopped all closures and masking in early March.

Could be a combination of weather and summer vacation. Were they masking in schools in early March?

In 2020, the south got hit harder than the north in the summer, and the north got hit harder than the south in the winter. It's not clear if that's weather or behavior, but it's probably a combination of both. Also keep in mind that weather affects behavior.

Must of Canada has a bad surge in January even with strong lockdown orders in most counties. We don't have a unified model.

No, we don't. But Canada surging in January fits in with the north/south divide.

It's hard to predict the ebbs and flows.

But I think it's fairly easy to predict that this virus is probably going to be around for many decades, at the least.

Add new comment