Welcome to another stream of consciousness edition. I have been doing a lot of thinking this week about the COVID-19 data that I have seen to this point, what it means, and what the future might bring.
For those not familiar with who I am, I make a living doing data analysis, and am especially interested in trying to get a sense of where we are heading with this Corona virus pandemic.
For those hoping to get the third installment of “The Consummate Cockburn Collection”, all I know is that my co-writer said, and I paraphrase, “blah, blah, blah, Financials for Annual General Meeting, blah, blah, blah”. Or in other words, “You ain’t getting it ’till next week.”
The good news is that today is a holiday, so I won’t have to share my overtime pay with Peter! I don’t think Peter is too concerned. He has his Ph.D. in Mathematics and so knows that zero multiplied by anything is still zero! This is a labor of love, so be kind to me in the comments.
Here are some of my observations from the COVID-19 data that I have seen so far.
Deaths has always been a more accurate measure than cases.
There was evidence very early on that COVID-19 could possibly be transmitted asymptomatically, or that a significant number of cases would go undetected.
There is a possibility (the jury is still out on this) that people may be infecting others before they start to show symptoms. If so it will be very difficult to contain. Lancet warns that “Independent self-sustaining outbreaks in major cities globally could become inevitable because of substantial exportation of presymptomatic cases and in the absence of large-scale public health interventions.” – InternetMonk – February 7th.
This means that…
The number of cases has been significantly under reported.
How much so? In cases that have been pro-active with their testing and were testing before the virus got well established, deaths (the more accurate measure) are about one percent of cases. Russia for example at time of writing, had 281,000 cases and 2,600 deaths. The U.A.E. had 23,000 cases and 220 deaths. So we know from these numbers that the ration of deaths to cases is somewhere around 1%. (Note this is different to the fatality rate which is the current rate of deaths divided by the total of deaths plus recovered) Contrast this to the U.S.A., Canada, and the World whose number are 5.9%, 7.5%, and 6.6% respectively.
But… deaths have also been under counted.
The New York Times has been doing a good job of tracking this. When comparing death rates over the last two months with death rates from previous years, the death rate currently has been significantly higher than would be expected from the reported COVID-19 deaths. This is likely to be some combination of unreported/untested deaths due to COVID-19 in combination with increased deaths that have resulted from factors related to decreased access to hospitals, delayed surgeries, increased domestic abuse, suicides, etc. It is hard for me to determine whether the former is greater than the latter, so for arguments sake let us assume they have relatively equal impacts. The numbers are all over the place, but my best estimate is that roughly 20% of deaths caused by COVID-19 are being missed. This is partly borne out by…
Apparently we can’t count on weekends…
When you look at the graph of the daily number of cases, or the daily numbers of deaths, the reported numbers are always lower on Saturday and Sunday (as reported on Sunday and Monday). This is true both in the USA, Canada, and also seen in the World totals. There have now been several times where I have heard late on a Monday or on a Tuesday morning, “We have good news to report… We have seen the number of cases [or deaths] decline over the last couple of days.” It would be funny if it weren’t so tragic.
The real number of cases.
World wide then, it can be estimated that the real number of cases is approximately 6.6 times the reported number of cases. Or 100 times the number of deaths. We can also factor in an additional 20% as mentioned above. So, instead of the reported 4,800,000 cases world wide, the real number of cases is likely around 38,000,000.
Here is the scary part.
That 38,000,000 just mentioned is only half of one percent of the World’s population. If that number is correct, then the virus still has 99.5 percent of the population to infect. I can’t get over the number of people who I have heard say recently that “I hope/think the worst is behind us” usually accompanied by “It’s time to get back to work.” Even after multiplying the known number of cases by eight, we still don’t get to one percent of the population. The number of deaths that we have experience so far is just a drop in the bucket compared to what could be coming. (Note that I did qualify my remark with the word “could” which I will explain later.”
Negative results from testing confirms we have only just begun.
Consider this. The people most likely to be tested are those who are experiencing symptoms, or are at a higher risk of contracting the virus. Yet in spite of this the huge majority of tests are coming back negative. The U.A.E. has tested 16.2 percent of their population. 98.5% of their tests have come back negative. Canada and the U.S.A. by contrast have only tested 3.5% of their populations, and so the negative results have been significantly lower. In Canada 94% of the tests have been negative, in the U.S.A. that number has been 87%. Why are those numbers so low compared to the U.A.E? Simply because we have tested such a small percentage of our population. When you consider that the U.S.A. leads the world in the number of cases, and yet less than half of one percent in the U.S.A. have tested positive, we might have a penetration of the virus into the the U.S.A. of maybe 3 to 5%. (Taking the .5 percent of positive tests and multiplying it by 6 to 10 as referenced above.) Taking the number of deaths and multiplying that by 120 gives us a penetration of 3.3%, so let’s go with that number for our further analysis.
What is likely to happen – A worst case scenario along with some possible numbers
Let us use a U.S.A. example here. People with more experience with me in this area are thinking that we need to get to between 60% and 70% exposure to approach herd immunity. Exposure can happen one of two ways. You get the virus, or you get the yet non existent vaccine. For those thinking that the vaccine will come soon, remember we have just started human trials for the first vaccine. Stage 1 may take as long as six months and there are several stages to go through. So, worst case scenario is that the number dead in the U.S.A. is 20 times what it currently is. That is, approximately 1,800,000 people. Remember this is just the U.S.A. example. Canadians can divide by ten.
In the U.S.A. there have been 84,500 deaths over the last six weeks. That has been under close to lock down conditions. Currently we are averaging about 1,500 deaths a day. If we project that forward for a year we end up with another 550,000 deaths. So if we keep things relatively controlled and we get an available vaccine in a year, we are looking at ten times the number of deaths than we have now. While that sounds absolutely horrible, it is a lot better than the maximum number of 1,800,000. I initially called this section good news, but 634,000 dead can not be conveyed as good news.
The more things are opened up the more that that number will rise above 634,000.
You will note from the graph above, that the number of daily deaths has been declining. I expect that with opening up the economy we will see that decline stall over the next two weeks, and then start to rise again. By the end of June we are likely to be back up to 2,000 deaths a day, likely to lead to another shut down.
What about a second wave?
There have been warnings about a second wave, and how it could be worse than the first. The pandemic of 1918 (which killed my great-grandfather), and the SARS virus, had two waves, the second more deadly than the first. I don’t think it is that simple. If you look at the graphs of Canadian provinces, you will see that each pattern is quite distinct. Some have two peaks, some have one, some have a higher then lower peak, some a lower then higher. I think what we see in waves is really going to vary from place to place, and from region to region. Remember the first real wave was China, and now their numbers barely register in the big picture.
There is also a large part of the world that hasn’t even experienced this virus yet. We are still in the first phase. There are many many more countries who are still going to be devastated by this.
Please continue to pray. For those who are able, seek to help those who are in need. Be generous. Be kind. And be forgiving and gracious to those who may see this differently to you.
As usual your thoughts and comments are welcome.
Update: I have deleted a number of comments that were over the top in terms of rhetoric. I am invoking my updated Godwin rule here.