Three big mistakes lead to where we are today. The choices we, both colloquially as every individual but also we as our governments and the world as a whole, repeatedly make choices that let COVID remain.
So, I’m not a epidemiologist nor I have any medical degree. That’s also not what I intend to talk about. I’m here to talk about how our dumb decisions have lead here and how our misunderstanding of the exponential function keeps us here.
The three big mistakes that IMO lead us here are 1) failure to completely lock down, 2) allowing BLM protests, and 3) allowing summer vacations.
The first mistake is understandable. We all looked a China’s brutal way of locking down: bolting apartment doors shut, raking (suspected) infected people up in sports arenas, and in general being all totalitarian about their lockdown. The west would not do this, we’d instead perform an “intelligent lockdown” and in a humane way “flatten the curve” and get past COVID in that way.
That sounds crazy today, but it made sense: maybe locking down a bit can halt the spread even if less efficient than the brutal Chinese way? After all, we had seen “pandemics” like ebola (several times), the original SARS, MERS, bird and pig flus being pumped all up only to turn out to be giant nothing burgers to the west. It’s completely fair to think so, and I did so too. If we shut down the country for 1-2 months every time there’s a new pandemic anywhere in the world, we’d get even less done than the French. In retrospect it is obvious we should have done differently, but at the time it was far from clear. Flatten the curve seemed like a good compromise. Slowly smother the disease with some effect on the economy, but without the complete effect of a total lockdown.
Turns out, we were wrong and the Chinese were right. The only way to get rid of the disease using lockdown is with a complete lockdown. That means everybody is completely isolated, and no infection is transmitted. It takes 2-4 weeks, after which people either a) have symptoms and can get treated/quantised as needed, b) have beat the disease without symptoms and are no longer infectious, or c) never got infected.
For this to work, it is necessary to have absolute no transmission. Not reduced transmission, but no transmission. Even if only 1% of the population breaks the quarantine, reducing transmission by a factor 100, that only sets the disease back by 1½ months assuming a propagation factor of 2 and an average transmission time of 1 week. On other words, if the lockdown is not 100%, not 90%, not 99%, not even 99.9%, but 100%, the disease will have a comeback as fast as the lockdown reduced it. After all, the virus started with just a few infected individuals and is still asymptomatically transmissible.
Of course, for the individual, it is easy to rationalise going out for just getting groceries or visiting family if they are ok with the risk. It’s barely against government recommendations at that. Surely, if everybody is quarantined it cannot do any harm if I, a totally uninfected person with just a bit of hay fever got out, right? Wrong! That’s what all the idiots in zombie films do, the ones we groan at and feel slightly relieved when they get a buckshot to cure them of stupidity. You know who else thought that their sniffles was just a hay fever? The super-spreader killing 100 people.
The lockdown needs to be 100% and individuals cannot be trusted with the decision because they cannot fathom they can be the next problem. So, the government has to enforce this. Ruthlessly. We saw it can work in China, but decided we were too good for that. That was our first mistake.
The first mistake is forgivable. It was a compelling attempt to resolve the problem, ignoring that China has experience with the first generation of the disease and claiming we knew better. Of course, large parts of that was also due to upcoming elections, pride, and just being too damn humane to make the right decisions, but that is neither here nor there.
The second mistake was allowing BLM protests to happen. They diverted attention from the very real crisis and likely facilitated a lot of spreading. At the time, the protests should have been forbidden and ruthlessly stopped. I’m not talking about putting in police and firing warning shots, but putting the military in, arresting anybody participating and locking them up for weeks (to deter, to prevent from taking part in the foreseeable future, as well as to quarantine). If that wouldn’t stop the protests, escalate until they could be stopped. A bunch of people sitting bored at home due to a half-assed lockdown could have done an online protest or if the M of BLM mattered, have taken up the issue a year later when it could get the focus it deserved. This mistake is easier to understand, but was already met with a lot of disbelief: how could the protesters be so dumb, the governments so naive? Of course, they could be explained away with elections, unrest, and freedom, but even so a lot of people could see it was a mistake.
The third mistake is even dumber: how was vacationing allowed? Why were borders not just closed and people forced to stay in their home country? It seems very obvious that summer vacations is what put us in the second wave. During the Spanish Flu, the second wave was caused by winter and ignorance (not in a negative way: they just had no idea to expect a second wave), but during COVID it was caused by the middle class being unable to just not go on summer vacations?
We made the choices that got us here. We made the choice not to shut down in the first place. Fewer people made the choices to allow protests to fuel infection in the US, and fewer people still made the dumb choices to go on summer vacation, likely igniting the second wave in Europe. Again, the problem is that these choices are not made by democracy; if even 1% of the population makes those dumb choices, they make the choice for everybody. And people cannot make the right choices themselves, because a) governments pretty much encouraged summer vacations and allowed BLM protests so what is people to do?, b) for the individual there is all reason to break or bend the rules and no reason not to, and c) people do not understand exponential growth so they cannot see when numbers are getting bad.
And we cannot just blame it on others. The toilet paper/disinfectant/flour/yeast shortage was not caused by stockpilers, but by all the normal people buying an extra pack of toilet paper because others were stockpiling. We are all guilty of this; I tried not to, but must admit I bought two packs of flour when one was all I needed because I also baked before the pandemic, so I have a pretty steady consumption of flour, and got used to seeing empty shelves. We call caused the shortages. Similarly, we all caused the first mistake: if we went out grocery shopping or for sport, we contributed to the first lockdown not working. Sure, we may not have infected anybody, but on the other hand we may have. And the people that did infect others probably didn’t do so out of malice.
The really unfair thing is that people are guilty even if they followed government guidelines. And people are punished regardless of how guilty they are. That’s a good way to cause depression in people honestly doing their best, because if they are punished it must not be good enough, and cause people with disregard for the rules for exacerbate that, because they get punished either way.
The “flatten the curve” story seemed very enticing. Don’t lock down, but instead keep infections down so hospitals don’t get overwhelmed. Seems good, right? Except I don’t think anybody did the math at the time. In the Netherlands, we have had 90000 confirmed infections from mid-February to now, mid-September. That’s 90000 infections in 7 months. We started lockdown in mid-March, and have had 88500 infections since then. That’s either 90000 infections in 7 months or 88500 infections in 6 months. That means that to get everybody sick (and thus “beat” COVID), we’d need 97-112 years (17,28 mio / 90k / (6 months / 12 months.year)). Even assuming the number of infections are under-estimated by a factor of 10, that’s still around a decade. If we assume that herd immunity sets in around 50% infection, that allows us to cut down the time by a factor of two, assuming against all odds the virus doesn’t mutate to allow reinfection. Of that time, we’ve spent 2½ months in “strict” lockdown and the remaining period in “mostly” lockdown, so that’s a 5 years worth of that. Even if we use the current peak infection numbers of 2000/day, that would mean it would take 2¼ years to get everybody infected (and just over a year to get half infected). This is obviously not sustainable – people are going nuts already after half a year and there’s no way people can handle 5-10 times that.
The hope is that we’ll need a lockdown until we get a vaccine. Problem is, a properly tested and widely available vaccine is probably not ready for another ½-1 year, despite what various presidents promise to make themselves look good internationally or in time for elections. If even that. AIDS never got a vaccine because people kept getting AIDS from the vaccine (note that AIDS has killed more people than COVID in 2020 so far). The original SARS never got a vaccine, because while patients got an immune reaction (like the Russian vaccine has shown), they in fact got so much of a reaction that it killed them. That’s not to say that no vaccine can be found, just that we need to be incredibly certain it is safe and efficient before administering it to 7,8 billion people, and this can take more time than the American election schedule can accomodate.
Half a year later, we still have no effective treatment against COVID and a vaccine it at least half a year away, so we need to consider an alternative to a vaccine. That doesn’t mean we should stop trying for a vaccine, but we need a contingency plan for the case that we’re not getting one. Quarantining our way out of the crisis would theoretically work, even today. A good complete lockdown would eradicate it after 4-6 weeks. But there’s no way people would go along with that now. And that’s ignoring that the crazies have gotten conspiracies spun up and spread, and that the entire world would need to coordinate on such a reaction (and good luck getting huge but poorer countries like India or Brazil on board). Our first mistake means quantising and containment is realistically off the table. Infecting people by alternating periods of lockdown and easing would take years.
Another alternative would be to slowly throttle the disease. If we can get the reproduction factor down below, we will eventually kill the disease. The problem is that then we run into the opposite problem of exponential growth, namely exponential decline. In the Netherlands, during the “strict” lockdown from mid-March to end of May, the reproduction number was around 0.75. If we start with 10000 infected (probably a very optimistic guess of the current infections), and assume that each reproduction takes 1 week, it would take 32 weeks (log(10000)/log(1/0,75)) until the disease is eradicated. Even though we’d get to 100 infections in just 16 weeks, getting to zero wold take more than half a year. That’s half a year of the strictest lockdown we had that lasted only 2½ months, or three times as long. If people get tired we’d just get another wave and another half a year or strict lockdown. Were we to open up a bit more and get reproduction down to 0,95 (note that we have not been that low since easing in June), it would take 3½ years, and at 0,9 it would take 1½ years. Those periods would all be under lockdown that are either as strict as in April/May or at least stricter than June. Oh, and all those numbers assume reproduction over 1 week, if it is instead 2 weeks, then all the periods need to be doubled.
Of course, we could say that at some point, we’ll be able to track infections and ease restrictions, so we only need to reduce the infected to, say, 50. That’s roughly the number of infected South Korea were tracking when one “got away.” Assuming we could handle that, we’d have to wait 18 weeks (log(10000/50)/log(1/0,75)) at April/May levels, or 1-2 years at “stricter than June” levels.
The numbers are of course based on Dutch numbers, but the funny thing about exponential growth is that the difference between 1000 and 10000 is the same as the difference between 10000 and 100000, so even scaling up to 10 times the infected or one tenth, would only add/subtract a couple of months (log(10)/log(1/0,75)) or less less if we assume a larger country with more infected can track and quarantine proportionally more people as well. Basically, the orders og magnitude are the same for all European countries: it would take from half a year to a decade to get rid of COVID with the current methods.
This shows that throttling the disease by keeping infection low also means we are really in it for years unless a vaccine becomes available. Take together, this shows that lockdown are not working. Sure, they reduce infection, but the moment we ease them, infections go right back up. Unless we are prepared to be in this situation for literally years, we need a plan for if a vaccine doesn’t manifest.
One way I’ve seen presented is the “infect the young” strategy. It takes the heard of the “flatten the curve” and tries to reduce number of severe cases that stress the health systems, not the total number of infections. We are not necessarily here to reduce the number of infections, but to reduce the number of fatalities. We understand the risks much better now, so maybe it’s time to isolate the over 50s (or over 55s or over 60s) and eliminate restrictions for others. That way we can get infection numbers up and build up herd immunity in a foreseeable timeframe. Sure, it will cause some casualties, but so will the other scenarios. The easing/tightening scenario will eventually lead to everybody being infected, everybody including the weak so at a higher death rate. The throttling approach would eventually lead to 4-20 times the initial infected (so 40000-200000 or 1/40 – 1% of the population), and whether that’s better or worse than the infect the young scenario depends on a lot of variables I’m not qualified to estimate.
Curiously, Sweden, the former dark horse of Europe is currently one of the only (the only one I found) without a second wave. They are still somewhat ahead of other European countries, but not the US. Whether this is due to less lockdown fatigue, due to actually closing down summer vacations (due to no other country allowing Swedes in due to high infection numbers), or simply due to random chance (though that is less likely comparing with literally all surrounding countries), it at least suggests that while letting the disease roam for a while may be bad here nd now, it may be better on the longer term.
Our next choice is which of these scenarios we want. We may not get the choice explicitly, just like we did not explicitly elect to have the pandemic spread everywhere (mistake number 1), nor to have a second wave (mistakes 2 and 3). Right now, we’re in the middle of mistake 4: pretending the second wave is NBD, numbers are going down and a vaccine is weeks away, so we can start going to the office and visiting friends and family again. Our actions around mistake 1 ruled out the quarantine and contain choice.
If we at least keep distancing, start wearing masks like civilised human beings, and go back to working from home and not traveling, even if not explicitly requires us to do so, we may at least have a choice which of the above scenarios we want. If we keep pretending we are past the pandemic, we may again indirectly make a choice we are not happy with; right now we are betting everything on a vaccine that may not come (and definitely won’t this side of the winter) with the only alternative being to infect everybody, but take 5 years to do so. And this is a real-life trolley problem: we only have a “bad” and “worse” option, there’s no “do nothing” or “good” option. The “do nothing” option is a trap that disables the “bad” button.
