Required ICU Capacity for COVID-19

[s243a]

I was looking up some numbers related to COVID-19 and the ability of the medical system to cope:

I would like to address the "treating aspect" because I'm not show how effect the world will be at "preventing" COVID-19. Alberta has 7.9 ICU beds per 100 000 people.
https://www.theglobeandmail.com/canada/ ... -outbreak/

There are more than 4.371 million people in Alberta so there are about (4300000/100000*7.9~340 ICU beds).
https://www.theglobeandmail.com/canada/ ... -outbreak/
...
although this source says that there is 207 ICU beds in Alberta:
https://edmonton.ctvnews.ca/how-many-ve ... -1.4859181
so my population numbers must be wrong.


Currently in Alberta there are 10 patients in ICU beds.

https://edmontonjournal.com/news/local- ... 19-update/

5% of patients will need an ICU bed.
https://jamanetwork.com/journals/jama/f ... le/2763188

I'm not sure how many ICU beds are needed in the absence of COVID-19 but I don't think it would take too many cases to reach ICU capacity.

However, with linear growth rates new beds should open up as patients recover. If we knew the spare ICU bed capacity then we would know the maximum linear growth rate. Near the peak of the virus the growth rate will change from a positive exponential to linear and on the decline it switches to a negative exponential.

Anyway, the number of ICU beds required would be equal to the number of new cases per day multiplied by the time needed in the ICU. I can't find any numbers about the time needed in the ICU but we could probably estimate it by the number of people in the ICU for places that are not near ICU capacity. My guess is it isn't that much greater than the 5% of people that will need ICU treatment. For instance in Alberta with 542 cases 10 are in intensive care (about 5.42% of patents).

Currently in Alberta there are about 56 new cases per day. In severe cases the median recovery time is 3-6 weeks:
https://www.therolladailynews.com/news/ ... -epicenter

Let's say 3weeks of that is in the ICU.

(56 new cases per day)*(0.05 of peple requiring ICU cases)*(21 days)=58.8

From those numbers 3 weeks would likely be too long in the ICU given current capacity. I'm not sure how long of the above median recovery time is actually in the ICU.

Anyway, from the above numbers I see why the current measures are quite extensive. I do need to research this more though.

What I'm unsure of is how long is typically needed in the ICU for the 5% of people that will get COVID-19 and need ICU care. From the above numbers 3 weeks in the ICU would be too long for Alberta if the number of new cases per day grew too much more. I'm not sure who well prepared other regions are but I can see why such extensive measures are being taken. Perhaps it will be more like 1 week in the ICU and if that is all that is required than maybe the numbers won't be so bad.

[s243a]

I found some more info:

Here's the number that I was looking for. On average people are spending 15 days in the ICU (for COVID-19 ICU cases)
https://www.esahq.org/esa-news/analysis ... ology-esa/

and this is in Italy, where there is likely rationing of ICU care. so more than 15 days may be required.

Anyway, if we use the 15 day number. then we have in Alberta:
(56 cases per day) * (5 % of cases in the ICU ) * (15 days in the ICU)

56*0.05*15=42 beds required for COVID-19 at current rates of cases per day.

Alberta only has 207 ICU beds and with only linear growth we will required 20% of those for COVID-19. I now really see why such extensive measures are being taken.

[s243a]

Maybe it's not as bad as I think.

"Province-wide, Williamson said Alberta has 207 intensive care unit (ICU) beds. But when special units are counted—those for burn units, coronary care units, NICU, etc—the total tally of ICU beds in Alberta is 704 beds."

https://edmonton.ctvnews.ca/how-many-ve ... jyY7KPKyCA

I was using the 207 number but maybe we can use the 704 number for the amount of ICU beds? Either way though I don't know what our spare capacity is (In Alberta) in the absence of COVID-19.

[musher0]

Hi, s243a.

Next time, please use the URL button to highlight the URLs in the text?
Highlighted URLs make any article easier to read. TIA.

As to your subject, I'm not sure, but I think ICUs for patients with serious burns
do not have ventilators.

Also, not all COVID-19 patients develop a severe case. Hopefully, some Intensive Care
beds will remain vacant.

Regards..

[s243a]

Hi, s243a.

Next time, please use the URL button to highlight the URLs in the text?
Highlighted URLs make any article easier to read. TIA.

Fixed. Apologies. I was posting late at night.

As to your subject, I'm not sure, but I think ICUs for patients with serious burns
do not have ventilators.

If true that could be a problem and might make the numbers worse.

Also, not all COVID-19 patients develop a severe case. Hopefully, some Intensive Care
beds will remain vacant.

Regards..

Multiple sources say that around 5% of people that get COVID-19 require an ICU bed.

[Moose On The Loose]

Multiple sources say that around 5% of people that get COVID-19 require an ICU bed.

That number is likely higher than the actual. We really have no idea how many cases there are in the US or most likely Canada. Without testing of the general public, you have a strong selection bias in the numbers. It is more likely that many cases go uncounted than cases that need care. Some people who get a milder case may think it isn't COVID or not even know they are infected at all. They may have as little as feeling slight off for a day.

The US is lagging so badly on testing that here it is just a pure guess how many cases there are. The US's medical services had almost no slack going into this mess so the system is far from robust against it. Urban areas that are places where people come and go got their first cases a while back. More rural places had a bit of a delay. This creates the illusion of "hot spots" were it is likely that the difference is only one of timing.

[s243a]

Multiple sources say that around 5% of people that get COVID-19 require an ICU bed.

That number is likely higher than the actual. We really have no idea how many cases there are in the US or most likely Canada. Without testing of the general public, you have a strong selection bias in the numbers. It is more likely that many cases go uncounted than cases that need care.

Numbers from Germany suggest that you could be correct:
"According to the country's federal health agency, The Robert Koch Institute (RKI), the mortality rate is currently at less than 0.5%. It's remarkable when compared to the grim numbers in Italy or Spain. Italy has reported more than 86,000 confirmed cases and over 9,000 deaths, which would seem to work out to a mortality rate of roughly 10%. Meanwhile Germany has reported over 49,000 cases but only 342 deaths, according to data compiled by Johns Hopkins University."

https://www.cbsnews.com/news/coronaviru ... -covid-19/

This would mean that there would be many more COVID-19 cases than reported. It will also mean will build up herd immunity faster.

Some people who get a milder case may think it isn't COVID or not even know they are infected at all. They may have as little as feeling slight off for a day.

There is a WHO study that contradicts your claim here:
"Asymptomatic infection has been reported, but the majority of the relatively rare cases who are asymptomatic on the date of identification/report went on to develop disease. The proportion of truly asymptomatic infections is unclear but appears to be relatively rare and does not appear to be a major driver of transmission."
https://www.who.int/docs/default-source ... report.pdf

However, I'm not sure how well defined "develop the disease" is and perhaps this might include symptoms minor enough for people to not know if they are sick.

The US is lagging so badly on testing that here it is just a pure guess how many cases there are. The US's medical services had almost no slack going into this mess so the system is far from robust against it. Urban areas that are places where people come and go got their first cases a while back. More rural places had a bit of a delay. This creates the illusion of "hot spots" were it is likely that the difference is only one of timing.

In lieu of good testing you can estimate the number of people that have COVID-19 based on the number of people that die from it:
https://medium.com/@tomaspueyo/coronavi ... NoF-jV_3rs

[musher0]

Hi s243a.

Fantastic article you found, Coronavirus: Why You Must Act Now, by this
Tomas Puyeo, and not just for the case rates.

It's got the clearest explanation of the relationship between spread and confinement
that I've read.

There is a bug in my SeaMonkey, I can't see the tables / graphs, but the man explains
their contents well, so they're not absolutely necessary for the understanding.

Plus his article has been translated in a wide range of languages.

Many thanks for sharing.