Covid-19 FAQs

[David Jones]

Covid-19 FAQs

The RSS has just published this webpage "Covid-19 FAQs". This mainly
relates to statistical aspects of Covid in the UK, but may be of
interest elsewhere.

https://rss.org.uk/policy-campaigns/policy/covid-19-task-force/covid-19-faqs/

[Rich Ulrich]

BTW, thanks for the link.

Question: Does anyone know of modeling for how quickly
"death" follows infection, or symptoms, or hospitalization?

Months ago, I picked up "about 5 days" for symptoms, 10 days
for hospitalization, and 20 days for death. But those were
estimates of the means, the scanty start of modeling, without
distributions attached.

What I have noticed here (U.S.) is that the number of
"confirmed cases" fell in the second half of January (from
values previously over 200 thousand per day) to the present
level that is slightly under 100 thousand per day. Hospitalizations
are also down, from 125 thousand beds (total) to 80 thousand.

But the rate of deaths persists at over 3000 per day, from the
numbers I collect from Johns Hopkins. I figured that true death
rates could have been increased due to hospital crowding in
January, but those conditions have eased.

So, I'm wondering, what is going on. Maybe the lag is a
little more than I expected? Models?

- I wonder if the terrible data collection that owed to the T****
administration is being upgraded.

--
Rich Ulrich

[David Jones]

Rich Ulrich wrote:

> On Thu, 4 Feb 2021 17:47:58 +0000 (UTC), "David Jones"
> <dajh...@nowherel.com> wrote:
>
> > Covid-19 FAQs
> >
> > The RSS has just published this webpage "Covid-19 FAQs". This mainly
> > relates to statistical aspects of Covid in the UK, but may be of
> > interest elsewhere.
> >
> >
https://rss.org.uk/policy-campaigns/policy/covid-19-task-force/covid-19-faqs/
>
>
> BTW, thanks for the link.
>
> Question: Does anyone know of modeling for how quickly
> "death" follows infection, or symptoms, or hospitalization?
>
> Months ago, I picked up "about 5 days" for symptoms, 10 days
> for hospitalization, and 20 days for death. But those were
> estimates of the means, the scanty start of modeling, without
> distributions attached.

This is a fairly recent outline of progression


https://www.businessinsider.com/coronavirus-covid19-day-by-day-symptoms-patients-2020-2?op=1&r=US&IR=T

THere are older versions, such as these
https://heavy.com/news/2020/04/covid-19-symptoms-day-by-day-chart-of-coronavirus-signs/

https://www.drugs.com/medical-answers/covid-19-symptoms-progress-death-3536264/

>
> What I have noticed here (U.S.) is that the number of
> "confirmed cases" fell in the second half of January (from
> values previously over 200 thousand per day) to the present
> level that is slightly under 100 thousand per day. Hospitalizations
> are also down, from 125 thousand beds (total) to 80 thousand.
>
> But the rate of deaths persists at over 3000 per day, from the
> numbers I collect from Johns Hopkins. I figured that true death
> rates could have been increased due to hospital crowding in
> January, but those conditions have eased.
>
> So, I'm wondering, what is going on. Maybe the lag is a
> little more than I expected? Models?

You may need to take into account the improved treatments for Covid
that have been developed (non-vaccine), leading to fewer deaths.

https://www.pharmaceutical-journal.com/news-and-analysis/features/everything-you-need-to-know-about-the-covid-19-therapy-trials/20208126.article?firstPass=false

https://www.pharmaceutical-journal.com/news-and-analysis/news/dexamethasone-is-first-drug-to-be-shown-to-improve-survival-in-covid-19/20208074.article

[Rich Ulrich]

On Mon, 15 Feb 2021 21:13:18 +0000 (UTC), "David Jones"
<dajhawk18xx@@nowhere.com> wrote:

>Rich Ulrich wrote:
>
>> On Thu, 4 Feb 2021 17:47:58 +0000 (UTC), "David Jones"
>> <dajh...@nowherel.com> wrote:
>>
>> > Covid-19 FAQs
>> >
>> > The RSS has just published this webpage "Covid-19 FAQs". This mainly
>> > relates to statistical aspects of Covid in the UK, but may be of
>> > interest elsewhere.
>> >
>> >
>https://rss.org.uk/policy-campaigns/policy/covid-19-task-force/covid-19-faqs/
>>
>>
>> BTW, thanks for the link.
>>
>> Question: Does anyone know of modeling for how quickly
>> "death" follows infection, or symptoms, or hospitalization?
>>
>> Months ago, I picked up "about 5 days" for symptoms, 10 days
>> for hospitalization, and 20 days for death. But those were
>> estimates of the means, the scanty start of modeling, without
>> distributions attached.
>
>
>This is a fairly recent outline of progression
>
>
>https://www.businessinsider.com/coronavirus-covid19-day-by-day-symptoms-patients-2020-2?op=1&r=US&IR=T

Wuhan was before anyone knew much at all.

>
>THere are older versions, such as these
>https://heavy.com/news/2020/04/covid-19-symptoms-day-by-day-chart-of-coronavirus-signs/
>
>https://www.drugs.com/medical-answers/covid-19-symptoms-progress-death-3536264/
>
>>
>> What I have noticed here (U.S.) is that the number of
>> "confirmed cases" fell in the second half of January (from
>> values previously over 200 thousand per day) to the present
>> level that is slightly under 100 thousand per day. Hospitalizations
>> are also down, from 125 thousand beds (total) to 80 thousand.
>>
>> But the rate of deaths persists at over 3000 per day, from the
>> numbers I collect from Johns Hopkins. I figured that true death
>> rates could have been increased due to hospital crowding in
>> January, but those conditions have eased.
>>
>> So, I'm wondering, what is going on. Maybe the lag is a
>> little more than I expected? Models?
>
>You may need to take into account the improved treatments for Covid
>that have been developed (non-vaccine), leading to fewer deaths.
>
>https://www.pharmaceutical-journal.com/news-and-analysis/features/everything-you-need-to-know-about-the-covid-19-therapy-trials/20208126.article?firstPass=false
>
>https://www.pharmaceutical-journal.com/news-and-analysis/news/dexamethasone-is-first-drug-to-be-shown-to-improve-survival-in-covid-19/20208074.article
>

That last one was way back in June. But these numbers surprised
me, in both directions. This is for the control group in a
dexamethasone trial.

"Among the patients in the control group, mortality after 28 days
was found to be highest in those who required ventilation (41%),
intermediate in those patients who required oxygen only (25%), and
lowest among those who did not require any respiratory intervention
(13%)."

I thought that back in June, it was still like 85% on ventilation who
died. On the other side, I did not know that so many as 13% of
the non-respiratory cases died. And I don't know what symptoms
they do represent.

>
>>
>> - I wonder if the terrible data collection that owed to the T****
>> administration is being upgraded.

--
Rich Ulrich

[David Jones]

No doubt there are differences in protocols about when patients should
be placed on the different types of ventilator.


But you did ask about modelling. Have you looked at

https://covidsim.org/

You can change the country if it doesn't start with your location. It
seems that this automatically makes use of the most recent data
available. The various tabs on the top bar lead to details of the
models, including times of transition between stages of the disease.

[Rich Ulrich]

On Tue, 16 Feb 2021 12:14:23 +0000 (UTC), "David Jones"

Thanks.

That modeling is interesting. The default page came up with
a projection of severe peaks in at the beginning of 2022 and
again for 2023. Hmm. I don't know what makes that likely.

One relevant parameter is whether vaccine "prevents disease"
or "prevents infection." I assume that the former describes
a non-symptomatic spreader. Is that likely or possible for
vaccines in general? - for covid-19 in particular?

The unlikely screen encouraged me to fiddle with a couple of
the parameters, to try other projections. Having another peak
in a year (or so) came up again when I did not expect it.

Next, I will have to click on the tab for "Manuall" to see what
I ought to be varying.

--
Rich Ulrich