[chbot] Modelling virus spread...

Charles Manning cdhmanning at gmail.com
Sun May 10 05:02:48 BST 2020


Bevan

The one problem that is not addressed by "flattening the curve" is that the
impact is different for different countries.

In the Northern Hemisphere, there is a benefit in flattening the curve as
all those delayed cases are pushed out into summer (ie. out of the flu
season).

In the Southern Hemisphere it is the other way around. Flattening the curve
pushes cases out INTO the flu season. Maybe not such a good idea.

As you know with mathematical modelling, you get what you want to get. It
isn't reality. Still, some models are closer to reality than others and
some are more useful than others.

Ultimately if a model does not match reality, then reality is right and the
model is broken.

The modelling done by that NZ "centre for excellence" have been appalling
but, unfortunately, seem to be the ones the government used the most.

They used two different modelling methods and got some ridiculous numbers
(eg. 80,000 would die) and were built on some pretty ridiculous critera
(eg. 400 days - more than a year) of lockdowns and controls.

At least one of their models (the first set) were copied from overseas and
just a few parameters were tweaked. The second set of modelling (the
stochastic models) were slightly more credible but were clearly off by the
time they published their reports.

I believe, but have not verified, that the second set of models were copied
from the Uk modelling that is available on github. My reasoning  is that it
takes a lot of effort to develop a serious model and they would have just
copied one if it was available.

FYI the latest UK models are here:
https://github.com/mrc-ide/covid-sim

On Sun, May 10, 2020 at 2:57 PM Charles Manning <cdhmanning at gmail.com>
wrote:

> Volker I think I need to push back a bit on one point you made:
>
> "Whether the economic cost of trying to stop the spread is worth it
> heavily depends on whether you're in the risk group,"
>
> Those at-risk people are for the most part frail and have co-morbidities.
> Life expectancy would be 2 years or less. Of that, not all would be high
> quality life, but let's just say 2 years.
>
> Of those at risk people, less than 1% would die, but let's just say 1% for
> an easy number.
>
> So the expected extra life for an at risk person is 1% of two years = 1
> week. If you want to change it to 2% (2 weeks)  or 5% (5 weeks) I don't
> believe it matters much.
>
> Now to get that extra week, everyone, including that at-risk person, has
> had to give up
> * Six weeks (and counting) of all the things that make life worth living:
> ** Socialising with friends
> ** Going to the park
> ** Visiting their family members, grandchildren etc.
> ** Travelling and having fun.
> * Their children and grandchildren have lost businesses, jobs etc.
> ** This is a long-lasting impact that will go on for years.
> ** Many frail people are going to go to their graves seeing their families
> wrecked in the name of "saving" them.
>
> Getting an expected 1, or even 5, week of life, but throwing away 6 weeks
> (and counting) of quality of life is not a good deal for the at risk people
> - let alone everyone else.
>
>
>
>
>
> On Sun, May 10, 2020 at 1:16 PM Volker Kuhlmann <list57 at top.geek.nz>
> wrote:
>
>> > An interesting result, but it makes several assumptions that impact the
>> > conclusion significantly. One is that once people have recovered from
>> the
>> > virus, they are immune. This has not been shown.
>>
>> Don't forget that the absence of any long lasting or permanent
>> detrimental effects on health of the recovered has also not been shown.
>> Permanently damaged lungs have already been mentioned.
>>
>> Volker
>>
>> --
>> Volker Kuhlmann
>> http://volker.top.geek.nz/      Please do not CC list postings to me.
>>
>> _______________________________________________
>> Chchrobotics mailing list Chchrobotics at lists.ourshack.com
>> https://lists.ourshack.com/mailman/listinfo/chchrobotics
>> Mail Archives: http://lists.ourshack.com/pipermail/chchrobotics/
>> Meetings usually 3rd Monday each month. See http://kiwibots.org for
>> venue, directions and dates.
>> When replying, please edit your Subject line to reflect new subjects.
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ourshack.com/pipermail/chchrobotics/attachments/20200510/b444e996/attachment.html>


More information about the Chchrobotics mailing list