Why we should continue taking actions that are proven to work.

I like the change to the thing.

If tomorrow no one interacted with anyone and we did that every other day it would be gone.

Have your models any capacity to throw light on how many infections we actually have here? Maybe basing off SK data.

My big suspicion is that our positive test numbers in no way reflect reality (although really it's hospitalised, ICU and death numbers that really matter atm). The labs are clearly at capacity and the new case definition hasn't fed through to the results yet. So if 94% are negative and we can't increase the numbers of tests getting through the labs, then you'd have to reasonably assume that for the next few days we'll continue to see new cases in the low 200s..

DeVore wrote: »

I dont know why you think that. What would happen is that every other day the virus would spread. You might think that that would mean it would grow half as fast but it wont.
Every lockdown-day, the people who housed with some one who picked it up on a previous free-day, would almost certainly get it. So, it grows by more than half for sure.

Total lockdown you cant leave your house for 24 hrs.

Then the next day the rules are the same as now you can go out but not more than 2 k etc etc.

Signore Fancy Pants wrote: »

I would be interested to hear your assessment of the Pandemic, specifically the characteristics which have contributed to the global spread of the virus, the value of mitigation measures and your 2 month projection for the course of the Pandemic.

Who said I could do any of that? I'm pointing out that this model, like so many others presented here, addresses only the benefits of our actions, not the costs.

The only attempt I have seen to quantify them is this paper, which more or less points out that we're already beyond the point of costing more lives than we save:
http://jvalue.co.uk/papers/J-value-assessment-of-combating-Covid-19-Thomas-23.3.2020.pdf

DeVore wrote: »

We have to do this without blowing through the vent limits because as soon as we exceed vents, mortality rates go sky high.

Cite. In Wuhan, ventilators made no difference. If you were put on one, you died anyway, just later:
https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)30566-3/fulltext

sydthebeat wrote: »

if the numbers arent dropping by the end of next week then its continue as we are right now

Why would we continue as we are if the numbers are not dropping?

If the numbers are not dropping we need to do more.

DeVore wrote: »

Ok so this is a good question.

So far most models predict "NEW CASES!!" and work from there (because we know the hospitalisation rates and the mortality rates for cases, from places like Italy etc).

What if we worked that backwards... what if we said "we know we have had 46 deaths, how many cases would be needed to cause that".

So, you can do that and the answer is today we should see about 4000 rather than the 2,615 we have reported.

Why dont we do this? Its because the system is very very sensitive to that number of deaths. Bluntly put if a few more grannies (or less) popped their clogs, well that 4000 number goes flying up and down. In technical terms its "chaotic" ie: highly sensitive to starting conditions.

We DO DEFINITELY need large scale wide spread testing. Right now we are blind and we need someone to turn on a light of data. Do you think they havent thought of that? They know that.
1. They need test kits. Those are in kinda short supply world wide now...
2. Its a hellova job to mobilise the logistics needed to test 4M people.

As you imply, models are only as good as the data they get. The issue is that much of the data is heterogeneous and requires several models to provide information.


I have worked in the area of diagnostic testing for the past ten years and have been loathe to get involved on these threads, but I thought this thread was at least serious in its intent so I wrote the below as An attempt to add to this thread with thoughts on vaccination
The virus is what is vaccinating us at present, those who have been sick and survived are “vaccinated” (assumption: this virus is like other corona viruses and re-infection cases are rare).


It is (very likely) correct that there are many more people with the virus than the testing numbers are telling us, but for now, that is not critically important because of those most will be asymptomatic, mildly symptomatic, and moderately symptomatic and not require hospitalisation (read “Healthcare resources”).


What is critical is protection of the available health system resource because of the highly contagious nature of this virus.
Social distancing and work from home measures may be considered (modelled as) a way of artificially applying “herd immunity” as herd immunity and these measures both result in the same outcome, i.e., apply an effective barrier between infected people to reduce risk of contamination.


So, we need the population to SLOWLY get to the 60% estimate that achieves herd immunity before we can think of re-opening contact channels.


So, while we are all focused on the headline news of new cases and unfortunate deaths, meanwhile the general population is moving toward herd immunity. The implication of this is that the “peak” in numbers of new cases is NOT the only measure that will be considered important when making the decision to remove isolation measures – it will be soon important to get a handle on how many have been infected to enable us to understand how close we are to herd immunity.


So, the strategy is effectively trying to achieve a twofold outcome:

• Flatten the curve to reduce health system burden
• Control the rate at which the wider population achieves herd immunity

Unfortunately, even when we achieve herd immunity there will be new cases and vulnerable people will still be vulnerable and will remain so until we have an effective vaccine. So it will still be very important to practice safe contact such as handwashing and cough etiquette and not avoiding handshaking etc.. But when we do achieve the level to be at Herd Immunity, the risk to overwhelming the health system will be largely gone.


Points on testing:

• The test being carried out on people who are showing symptoms are tests to directly detect the virus through its genetic material – the presence of symptoms indicates that there is enough virus present in the person to be within the limits of detection of the test.

• Antibody tests require a person’s immune system to have reacted to produce the antibodies as a result of infection so cannot be used too early, even when initial symptoms appear. It may take several days AFTER immune system reaction before there is a high enough concentration to be detectable by an antibody test.

In fact, there is a window in which a person would test positive for test 1 and negative for test 2 so its important to understand when and what testing should be done.

DeVore wrote: »

Ok so this is a good question.

So far most models predict "NEW CASES!!" and work from there (because we know the hospitalisation rates and the mortality rates for cases, from places like Italy etc).

What if we worked that backwards... what if we said "we know we have had 46 deaths, how many cases would be needed to cause that".

So, you can do that and the answer is today we should see about 4000 rather than the 2,615 we have reported.

Why dont we do this? Its because the system is very very sensitive to that number of deaths. Bluntly put if a few more grannies (or less) popped their clogs, well that 4000 number goes flying up and down. In technical terms its "chaotic" ie: highly sensitive to starting conditions.

We DO DEFINITELY need large scale wide spread testing. Right now we are blind and we need someone to turn on a light of data. Do you think they havent thought of that? They know that.
1. They need test kits. Those are in kinda short supply world wide now...
2. Its a hellova job to mobilise the logistics needed to test 4M people.

Also, another factor to consider, won't the current death rate change when the capacity of icu beds become full? Afaik we're not at full capacity yet, so when we hit that, shouldn't we see an increase in deaths??