The San Marino Limit: A Covid19 Hypothesis
What if, whatever happened in San Marino, happens everywhere?
Banner Image: CC-By-4.0 license by copyright 2020 ISO-FORM LLC.
EDITORIAL NOTE: Subsequent events have proved that the speculations in this article are definitely wrong. It is kept here for future reference.
Note: I am not an epidemiologist, trained in public health, nor a medical doctor. You should believe anyone with those backgrounds, before you believe anything I say below.
The San Marino Limit.
The United States hit a covid19 marker for me this week. I wouldn't call it a positive marker, but it's one where I am very interested in, for what happens next. Because it may be that the US's cases will now peter out, and go to "very low" (10% of the current new cases rate), as early as November 1, or as late as February 1, with no further medical interventions (such as a vaccine).
I cannot tell you why it will do this. I can only tell you - it is an observational hypothesis, based on what I see, first from San Marino, and then, from elsewhere. Several elsewheres, in fact.
Back in May, I noticed that San Marino had seemingly "peaked out" on the number of deaths, and the number of cases. San Marino is a tiny country -- about 34K people, surrounded on all sides by Northern Italy, which, as you know, had covid19 rampant.
San Marino peaked out near 20,000 cases/1M population - only about 2% of the total population, but it was the highest rate of any documented country in the world back in May. If you adopt the hypothesis that San Marino had hit the worst-case scenario of herd immunity, then the difference between a typical herd immunity (thought to be 30-70% of the population) and the number of observed cases could be due to a number of factors, like asymptomatic carriers, and (we've learned since) that there is some evidence of pre-pandemic immunity, possibly due to a related, less virulent virus which circulated globally, previously. But the difference is still startling: 2% of total population as cases, vs. 70% for "herd immunity".
New Cases by day in San Marino. New Cases peaked in late April/early May, and then petered out, in spite of there being a substantial population — about 50 times as many those counted as confirmed cases — which had not been diagnosed with Covid19. Source: https://www.worldometers.info/coronavirus/country/san-marino/ Accessed September 12 2020.
That factor of 35 difference, between what is expected, and what is observed, is substantial.
And since May - these cases, and deaths, haven't climbed in San Marino. That, to me, is shocking. Imagine you have a leaky tub filled with water -- the water is people, and when water leaks out, those people have caught "cases". The tub has uncontrolled leaks, and leaks out 1/4 a cup, but then it stops. You know that the tub contains 2 more gallons, but no more water is leaking out. Why not? Why isn't the water leaking out any more? What has stopped the uncontrolled leak?
Where are the rest of the cases in San Marino? Why haven’t they happened? Why hasn’t some group slipped up, gone out to dinner, and spread an event which exponentiates like the first time? Some of this is certainly due to asymptomatic cases. But that rate is thought to be a factor of 5 difference - perhaps 10. Not a factor of 35.
So I don't know why San Marino’s cases largely stopped at 20K cases/1M population. One obvious possible reason is medical: that the virus ran out of possible hosts -- well before we would have thought due to "herd immunity". In which case, behavior in San Marino would no longer be a factor in future infections -- people could crowd into bars, and no (or, few) more infections would occur. But another possible reason is psycho-social: perhaps at 20K cases/1M population - and the deaths which accompany it - is when a community at long last says OKAY HOLD ON EVERYBODY STAY WHERE YOU ARE THIS SHIT IS REAL and real lockdowns and caution actually start occurring, without additional coercion of the population. In that case, if people crowd into bars, the pandemic - and disease, and deaths - would take off again.
I don't know which of these is playing out in San Marino (if I lived there, I could tell you - perhaps someone who lives near there could say in the comments). If it's due to the psycho-social factors - then nothing I say below here matters, and this San Marino Limit does not represent a true end to the pandemic in a population. In fact, our social assumption now, is that psycho-social factors are the reason the cases rates has dropped, anywhere, and everywhere. And that may well be the case. Perhaps at a 2% infection rate is when people begin to take the public health admonitions seriously, stay at home when they can, avoid others, avoid concentrations of people, avoid poorly ventilated interiors, avoid staying inside with others for long periods of time.
Nonetheless - San Marino's experience gave me something to look for: what happens to other countries as they approach the 2% mark in cases? Do any of them blow past it by a factor of a few, a factor of 5, a factor of 10?
Since then, this mark has been passed by several countries (see the Table below, from WorldInfo on September 10, 2 days ago), mostly not by much (a factor of 1.1-1.3), with one (Qatar) a factor of ~2 higher. But none of them by a factor of 3 or 5, 10, or anywhere near a factor of 35.
Countries with the highest number of confirmed Covid19 cases per 1 million population, on September 10, 2020 (source: https://www.worldometers.info/coronavirus/). The cyan lines indicate countries with fewer than 100K population. There appears to be a clustering, or wall, within +/-50% of 20,000 cases per 1 M population - about 2% of population - above which there are only a few countries (Qatar, Bahrain, and French Guiana). This wall is the “San Marino Limit” - the per-population case rate where new cases stopped (or rather, slowed considerably) occuring in San Marino. The coincidence of this happening in several countries, while the limit of “herd immunity” remains a factor perhaps 35 times higher (near 70%) is a curious mystery.
This suggests that, indeed, the San Marino Limit (as I have come to call it) may be a real (if approximate, plus or minus 50%) medical limit on the total integrated population case rate for Covid19. (It does not point toward a medical reason, over a psycho-social reason, however).
This is significant, because this week (September 10, 2020), the US has surpassed the 20,000 cases/1M population mark. If, indeed, the San Marino Limit is an important one - a medical one - then the pandemic in the US will now peter out, much as it seems to have done in San Marino, and has also done in other countries on the above list: Aruba, Panama, Chile, Peru, Kuwait, Brazil, Oman, Maldives, Israel, and others.
So here are my predictions, based on this (scientists, take note) *hypothesis*. I repeat: this is a *hypothesis*. This means, if the *hypothesis* is wrong, everything I predict below can turn out to be wrong.
The US has a daily case rate of 35K/day, and dropping. If the San Marino Limit is real and medical, then based on other countries' excess, we might expect the US to stop short 25K cases/1M population -- 1.65M more cases, on top of the 6.6M we have now. With only 25% more cases before the epidemic would be effectively over in the US, that would imply at most only 25% more deaths - or 50,000 more people dead. There's evidence, however, that the fatality rate has dropped by half in the last 2 months (due to a variety of factors - a younger case population, for example, and improvements in treatments); in which case, the US would suffer "only" 25,000 more people dead.
When would all this be "over" by? At 35K cases/day (today's rate), this 1.65M will be exhausted in 47 days (by November 1 2020). Or, if the case rate continues to drop exponentially in the US, as it has been since July 20, then the exponential timescale is about 75 days, and that means the case rate will be down to about 10% of its current rate no later than Feb 1 -- without any medical interventions (such as a vaccine). Of course, the cases and deaths won't go to zero, in this scenario - but they will be well below where they are now, and would certainly never shoot up exponentially in an uncontrolled outbreak -- as one would conclude they could, if there really is a larger population of people who could be infected.
New York City daily case rate. New cases peaked in early April, decreased thereafter, and today holds to a near constant rate. By June 1, when the major flux of cases had largely completed, 203K cases had been counted - 2.4% of the total population. The total confirmed cases today (September 10 2020) are 2.8% of the 8.4M population of NYC. The new cases rate has not gone to zero, and continues to increase. At the current new-cases rate (about 240/day), the doubling time is 2.7 years. (Source: https://www1.nyc.gov/site/doh/covid/covid-19-data.page accessed September 12 2020).
Additional data points, are NYC, Lombardy and Montreal- where the pandemic spread very early on, in a nearly uncontrolled manner. Today (September 12 2020) in NYC, there have been 233K cases, in a city of of 8.4M, which is 2.8% of the population. In Lombardy, there have been 103K cases, in a population of 10M, which is 1%. Montreal, where I live and which for a period had the highest new cases and deaths rate in North America, has 30.3K cases in a city of 1.8M, which is 1.6% of the population. I regard these as approximately consistent with the San Marino limit of 2%. (In the case of Lombardy, the spread was so substantial, that under-reporting due to a breakdown in medical records collection and communication may explain a factor of 2 discrepancy.)
Montreal daily case rate (7 day average). New cases peaked in late April, decreased thereafter. By July 1, when the major flux of cases had largely completed, 26724 cases had been counted - 1.5% of the total 1.8M population. The total confirmed cases today (September 10 2020) are 2.8% of the population of NYC. The new cases rate has not gone to zero, and continues to increase. At the current new-cases rate (about 240/day), the doubling time is 2.7 years. (Source: https://montrealgazette.com/news/local-news/coronavirus-live-updates-most-quebecers-approve-of-in-person-schooling-poll-finds/wcm/4a1e4050-b7ba-44e3-9cb2-ca200d9bfecf/ accessed September 10 2020).
To summarize: if this hypothesis of the San Marino Limit is correct -- that ~2% of a given population becomes a covid19 case, and then there is a *medical* reason for contagion to end -- then the Covid19 pandemic will be effectively done in the US sometime between November 1 and February 1, with no further medical intervention (such as a vaccine). During that time, between approximately 25K and 50K more people will die.
Similarly for other large countries which are today, at or above a total cases rate of 20,000 per 1M population — Panama, Chile, Peru — if we hold to a maximum of 25,000 cases per 1M population, these countries would not have any greater outbreaks than they have had in the past; their cases should not double, although they may grow by 25% or so, and the new cases rate should slowly decrease in time.
There is a thing which could happen, which would falsify the hypothesis that the ~2% limit is a medical one: if, in some jurisdiction, the total new cases should skyrocket, to 6%, or even 10% of the population — still well below where “herd immunity” is thought to be important — that would demonstrate convincingly that this ~2% limit is not a medical one, but more likely is due to the psycho-social factor of a population respecting public health directives. This has not happened yet.
——————————————
Banner Image: CC-By-4.0 license by copyright 2020 ISO-FORM LLC.