Covid19, to use its "proper" name, has spread around the world, with an unknown death toll so far, and an unknown eventual death toll. In the developed world, the death toll is fairly well understood, and predictable. For example, I predict that by the 3rd of May, just under two weeks from the date of writing this, there will be:
80,000 deaths from Covid-19 in the United States, and
29,000 deaths from Covid-19 in the UK
How can I make that prediction? I look at the number of cases, and extrapolate the number of coming deaths. It really is not that difficult, and having done so over a week ago, I was able to confirm that the methodology, as blunt a tool as this is, results in projections that tracked fairly closely to actual death toll through the following week.
For countries that have adequate testing programmes (adequate, not good) that enable them to identify cases presenting at hospitals or doctors, these simple extrapolations work.
This does not work for the entire world, and if anything serves to confirm that the global death toll will be far higher than any numbers that we can estimate from reported cases. There are too many parts of the world where there is inadequate counting of actual cases. This is especially true for underdeveloped or as we like to call them these days, developing countries.
Ecuador, a lovely country of warm and welcoming people, is suffering terribly. Early on there was the hope that warm weather countries would be spared the pandemic. This was not the case. Unfortunately, as a developing country, Ecuador has a limited healthcare system; effective and good quality for the rich, but completely inadequate for handling something approaching Covid-19. Earlier, in late March there were reports of the first case in Ecuador, and by early April the news was simply terrible.
As of today (April 21, 2020) Ecuador officially reports a total of 10,000 cases, with 500 deaths. But these are the confirmed numbers only. On the 13th of April, France24 was reporting that close to 800 bodies had been collected from people's homes.
Eventually, Guayaquil city reported that the first two weeks of April saw an almost unimaginable spike in death, with bodies literally lying in the streets, and there had been almost 6000 additional deaths. The BBC reported:
How does this relate to the US and the UK?
The single biggest difference is the ability of the "developed" countries to test. And yet, both the US and the UK have been "late to the party" when it comes to testing, and because of that, the virus has spread much further through their populations then they are willing to admit or prove (or prove otherwise).
So returning to the original purpose of this post, the projected death tolls in the US and the UK on the 3rd of May, just under two weeks from today.
How did I get to those estimates?
It was observed some time ago that deaths tend to be delayed by two weeks from the date of confirmation of the diagnosis. That means that the news cases today will result in an estimated number of deaths in two weeks time. By taking the historical data from this pandemic, we can calculate the percentage of people reported as confirmed cases that may die. For example, if 10,000 cases are confirmed on a given day, and the mortality rate was 3%, then we can predict that in two weeks were will be 300 deaths.
Now, this is an extremely blunt instrument of a calculation of future deaths.
But for the past week, I have been watching that blunt instrument, and it has held directionally correct, with daily peaks and troughs suggesting it is either too aggressive or too meek.
Therefore I updated my charts, and fed in last weeks numbers through the 19th, and this is what it tells me:
The US numbers will continue to grow to reach 80,000 on the 3rd of May. These deaths are already "baked in" to the existing numbers of cases. And as there will most certainly be many more cases, it is safe (if that is a word that I want to use right now) that the total will be much higher.
Remember that these expected deaths are based on cases already reported and confirmed across the United States.
Looking at the UK, the situation is no better. My prediction for the UK is 29,900 deaths by the 3rd or May. This uses the same blunt instrument. Again, tracking the totals over the past week have convinced me that this methodology is directionally correct. Certain I hope that it is out significantly on this high side.
Based on the fact that so many deaths are already "baked in", my expectation is that the total UK death toll by the end of May, with the hope that the peak is in or is very close, will still be around 40,000 deaths.
These numbers or terrible, and each is a lost family member, a friend, a grandparent, parent sibling. There will be time later to look back and learn from the responses of government both countries, and certainly for all of us to look at what we need to do to help developing countries avoid this in the future.
But to be clear, it does not take significant computing or mental power to run projections based on what has already happened, We are not talking about what will happen, because we don't want to, not because we cannot predict what will happen.
Disclaimer: I am not a mathematician, I do not play one on TV, and if fact I do not even know how to spell sadistics.
80,000 deaths from Covid-19 in the United States, and
29,000 deaths from Covid-19 in the UK
How can I make that prediction? I look at the number of cases, and extrapolate the number of coming deaths. It really is not that difficult, and having done so over a week ago, I was able to confirm that the methodology, as blunt a tool as this is, results in projections that tracked fairly closely to actual death toll through the following week.
For countries that have adequate testing programmes (adequate, not good) that enable them to identify cases presenting at hospitals or doctors, these simple extrapolations work.
This does not work for the entire world, and if anything serves to confirm that the global death toll will be far higher than any numbers that we can estimate from reported cases. There are too many parts of the world where there is inadequate counting of actual cases. This is especially true for underdeveloped or as we like to call them these days, developing countries.
Ecuador, a lovely country of warm and welcoming people, is suffering terribly. Early on there was the hope that warm weather countries would be spared the pandemic. This was not the case. Unfortunately, as a developing country, Ecuador has a limited healthcare system; effective and good quality for the rich, but completely inadequate for handling something approaching Covid-19. Earlier, in late March there were reports of the first case in Ecuador, and by early April the news was simply terrible.
As of today (April 21, 2020) Ecuador officially reports a total of 10,000 cases, with 500 deaths. But these are the confirmed numbers only. On the 13th of April, France24 was reporting that close to 800 bodies had been collected from people's homes.
"The number we have collected with the task force from people's homes exceeded 700 people," said Jorge Wated, who leads a team of police and military personnel created by the government to help with the chaos unleashed by COVID-19. He later said Sunday on Twitter that the joint task force, in operation for the past three weeks, had retrieved 771 bodies from homes and another 631 from hospitals, whose morgues are full.
Eventually, Guayaquil city reported that the first two weeks of April saw an almost unimaginable spike in death, with bodies literally lying in the streets, and there had been almost 6000 additional deaths. The BBC reported:
The government said 6,700 people died in Guayas province in the first two weeks of April, far more than the usual 1,000 deaths there in the same period.
How does this relate to the US and the UK?
The single biggest difference is the ability of the "developed" countries to test. And yet, both the US and the UK have been "late to the party" when it comes to testing, and because of that, the virus has spread much further through their populations then they are willing to admit or prove (or prove otherwise).
So returning to the original purpose of this post, the projected death tolls in the US and the UK on the 3rd of May, just under two weeks from today.
How did I get to those estimates?
It was observed some time ago that deaths tend to be delayed by two weeks from the date of confirmation of the diagnosis. That means that the news cases today will result in an estimated number of deaths in two weeks time. By taking the historical data from this pandemic, we can calculate the percentage of people reported as confirmed cases that may die. For example, if 10,000 cases are confirmed on a given day, and the mortality rate was 3%, then we can predict that in two weeks were will be 300 deaths.
Now, this is an extremely blunt instrument of a calculation of future deaths.
But for the past week, I have been watching that blunt instrument, and it has held directionally correct, with daily peaks and troughs suggesting it is either too aggressive or too meek.
Therefore I updated my charts, and fed in last weeks numbers through the 19th, and this is what it tells me:
The US numbers will continue to grow to reach 80,000 on the 3rd of May. These deaths are already "baked in" to the existing numbers of cases. And as there will most certainly be many more cases, it is safe (if that is a word that I want to use right now) that the total will be much higher.
Remember that these expected deaths are based on cases already reported and confirmed across the United States.
Looking at the UK, the situation is no better. My prediction for the UK is 29,900 deaths by the 3rd or May. This uses the same blunt instrument. Again, tracking the totals over the past week have convinced me that this methodology is directionally correct. Certain I hope that it is out significantly on this high side.
Based on the fact that so many deaths are already "baked in", my expectation is that the total UK death toll by the end of May, with the hope that the peak is in or is very close, will still be around 40,000 deaths.
These numbers or terrible, and each is a lost family member, a friend, a grandparent, parent sibling. There will be time later to look back and learn from the responses of government both countries, and certainly for all of us to look at what we need to do to help developing countries avoid this in the future.
But to be clear, it does not take significant computing or mental power to run projections based on what has already happened, We are not talking about what will happen, because we don't want to, not because we cannot predict what will happen.
Disclaimer: I am not a mathematician, I do not play one on TV, and if fact I do not even know how to spell sadistics.
No comments:
Post a Comment