In short:
- Outside China, 2% of cases have progressed to “severe”. But if the lag is eight days then progression to severe is more like 5%.
- In China about 1/5th of severe cases are “Critical”
- If that rate occurs in the West, hospitals will be overwhelmed if just 1% of the population gets infected.
- In Singapore, the doubling period for confirmed cases is about nine days.
- Currently the spread is not exponential outside China (most days)
News today: First victim in Europe — an 80 year old Chinese tourist.
Stats: Total cases outside China: 787 Deaths: 4 Severe cases: 18 (2%)
Early days of “outside China” data
The 2% rate of severe cases is an underestimate above. There is an eight day lag from diagnosis to “severe,” and then a longer lag to death. Total cases outside China on Feb 8th was 354. So a more realistic estimate is that about 5% of confirmed cases outside China have now progressed to “severe” (i.e. 18/354).
What does severe mean? It appears “severe” means hospitalized but not necessarily in ICU. In China, the rates issued in a Feb 7 press release were 82% mild, 15% severe, and 3% critical. From that, we might assume that only a fifth* of “severe” cases are critical and therefore in need ICU care, so outside China that might be 1% of all recorded infections. (Remember the Chinese statistics are all from hospitalized people, outside that in the whole population it probably is much lower, because many people apparently get a cold and stay home and aren’t included. Though there are some who stay home and die at home and they aren’t included either. Twitter shows vans visiting apartment complexes, and being loaded with bodies. How many? Who knows. This is why statistics outside China are the only ones that count.)
*Technically one sixth (3/15+3 — the accumulated severe plus critical total)
Estimating when trouble may really start in Singapore
So, with Western hospitals, perhaps only 5% of confirmed cases become “severe”, and only 1% of confirmed cases need an ICU. Current ICU bed availability in the West is typically about 1 bed per 12,000 of population (and there are even fewer of the proper “negative pressure ICU” rooms we need for best quarantine of an aerosolized disease**). Once the the infection has reached 1% of the total population in a Western nation, about 120 out of 12,000 people have a confirmed case, and about 1% of them — or 1.2 patients — will need an ICU bed. At that point, all the ICU beds are in use, and we have run out of ICU places — even if none are required for other uses. Not happy days in hospital-land.
Obviously we need to slow the spread of the virus urgently, aggressively, so that hospitals don’t have to send people home with an oxygen tank, a how-to-guide and good wishes. Even being optimistic and if the rate of progression to “severe” is only one fifth as common in the West as it is in China (there are reasons to take an optimistic stab), our current medical system stops being able to cope when about 5% of the population gets infected. All numbers are loose — the 5% rate of severe cases assumed in Singapore above might be too low — in Singapore the exact current rate is 8% in HK it is 13%. Sorry about all these numbers.
**Aerosol or not? There is a lot of disagreement over whether it is or isn’t?
The exponential curve we don’t want
The exponential growth of infections in China meant that hospital system was always going to get overwhelmed. It took just two months. The Lancet reported on January 24th that, of the first 41 patients admitted in Wuhan (by Jan 2nd), 32% ended up in the ICU and 15% died. In the Wang study the news was better: 26% of 138 patients needed ICU and only 4% died. But one month after the first study, by Feb 5th, Wuhan hospitals were overwhelmed and turning away all but the most severe cases.
As I said on my first post two long weeks ago, human brains don’t seem well adapted to planning for exponential curves. The doubling period inside China was six days in January. Outside China it is about eight days so far, though that is mostly dominated by the unfortunate cruise ship, which is in lock-down off Japan. Ominously, the doubling period in Singapore — which has dedicated advanced infection tracking — is about nine days (40 infections on Feb 8, and 72 today).
Singapore bad case: assume 10 day doubling, 5% progress to severe, 1% progress to ICU
To give some idea of how rapidly this might go, ponder that those 72 cases in Singapore could become 73,000 with ten doublings — which is only three months away. Of that, there may be around 3,500 severe cases and 700 ICU cases. There are probably around 12,000 total beds in Singapore hospitals. Occupancy rates already peak at 85% in March. That’s not a happy set of numbers.
To extrapolate (just to make a point), in less than six months the entire population of six million Singaporeans could theoretically have been exposed — except that sometime around four months the growth curve would slow, because a large section of the population will already be immune (we hope) and the most vulnerable will already have caught it. I expect things will be slower as we learn more how best to help those with it, and how to quarantine. But we can see why Singapore’s health officials are sweating and working so hard to track and trace and hunt down every last case (which they haven’t been fully successful at). What they are not announcing publicly is that without any effort to slow this, or any anti-viral or vaccine, and without entirely shutting schools, factories and enforcing a mass home quarantine, Singapore is only a few months away from hospitals reaching full capacity. We can all see why they don’t want to dwell on worst case possibilities in public. Beyond a few months, without a slowing, the unthinkable, potential pandemonium and mayhem unfolds. We hope that doesn’t even come close. But keep those worst case numbers in mind. Anyone who says “it’s like the flu” hasn’t run the numbers. This is nothing like the flu.
As hospitals fill, manufacturing systems and supply chains will decay. The system will be far beyond the normal epidemiological curves. It will be hard and then impossible to get enough masks, consumables, or even medicine (especially if it’s made in China). If things hit that point, it’s a “black swan”. China is deep inside that.
Best case: West contains the spread
As always, let’s repeat the optimistic caveats: Covid 19 will likely be less severe outside China due to cleaner air, healthier lungs, better diets, lower population density, possibly genes (ACE2 receptor), cultural habits, more sun, better nutrition, lower rates of smoking, and better medical systems. We also got a head-start and, if we are not totally stupid, we might use that to our advantage. We hope we can stay above all this and help the poor sods stuck in China, and probably Africa, and possibly Indonesia, India, etc. We won’t be much use to them if we lose control ourselves. The point of this post is to raise awareness that Singapore is walking on a ridge between control and a deep abyss, and it’s not out of the question that the West may follow. We really really don’t want to get on the wrong side of that exponential growth curve.
Perhaps we are seeing the awful result of malnutrition in China?
I’m astonished to note in the Wang et al study that two thirds of those in the ICU are listed as having “anorexia,” whereas that one third that didn’t need an ICU were listed the same way. I am baffled that there is not more discussion of this. Does it mean malnutrition? Is it a bad translation? (Anoxia is spelt like anorexia?)
Patients treated in the ICU (n = 36), compared with patients not treated in the ICU (n = 102), were older (median age, 66 years vs 51 years), were more likely to have underlying comorbidities (26 [72.2%] vs 38 [37.3%]), and were more likely to have dyspnea (23 [63.9%] vs 20 [19.6%]), and anorexia (24 [66.7%] vs 31 [30.4%]).
Perhaps some medico’s can help out — it simply makes no sense that most elderly Chinese would be anorexic in the same sense as the term is used in the West.
As for dyspnea — it means means “shortage of breath”.
The good news: Outside China things are not exponential
Not yet anyway.
Coronavirus Cases outside China | Worldometer
Obviously it could go exponential, and probably will if the virus takes hold in places like Africa (which has just reported it’s first case). But Africa hasn’t done much testing, and has a large fly-in-fly-out (FIFO) population of Chinese workers. If things take off in Africa, we will need a new category of graph “Cases in The West”.
….
Still noisy early data, and we can’t tell at this point whether the West will keep control or lose it.
As long as the growth factor is below one (in the growth of daily cases below), the growth is not exponential. However, be aware that at some point our ability to slow the linear growth and keep it under “1” will be overwhelmed. It is simply not possible to do exhaustive tracking of each new case, tracing back to find the source and isolating all the other contacts. At 50 cases Singapore had one unexplained source. Is that the point nations lose control? We don’t know. Singapore might get lucky. They will be tracking hundreds of people.
Cruise boats aside I hope this is a pattern that stays under 1. But there is that nagging concern about untested cases and the superspreader wildcard. Perhaps Singapore was just unlucky and got one, or perhaps even the strain of virus there may be different. Mutations are high in single stranded RNA viruses so there is possibly a cloud of different ones spreading right now. The most infectious strains will win that race, we just hope they are also less nasty.
Daily growth curve outside China. The red line is “1”. Above that is exponential growth. | Worldometer.
There are new hints today that the virus may weaken hearts and that those recovered may still not be out of the woods. I need to follow that. There were reports like that a few weeks ago, but those referred to the first week of infection. People were sent home, then had to come back.
I’ve heard that China has very stringent tests before people are declared recovered, which is why it takes a month to get on that list. But of course, if the experience of fighting Covid 19 leaves long term damage in some tissues there is probably no data on that yet.
A historic event. Let’s hope it gets boring soon.
REFERENCES
Huang et al (2020) Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China, Lancet.
Wang et al (2020): Clinical Characteristics of 138 Hospitalized Patients With 2019 Novel Coronavirus–Infected Pneumonia in Wuhan, China – JAMA, February 7, 2020