[FRIAM] Fwd: [NICAR-L] AP/Johns Hopkins COVID-19 data

Wed Apr 8 17:21:41 EDT 2020

Friammers,

Jim Girard and I have been having a conversation about the period between the bent curve and the vaccine.  His responses have been fascinating, but in some particulars, are more technical than this old rejected English major can parse.  I therefore wanted you all to get the benefit of them and, perhaps, explain them to me.  Jim gave me permission to forward the correspondence to you. 

So right now I have contact with 5 other people, my “pod”.  (Think whales)   I walk a mile or so a day in the neighborhood, designating one “clean” hand (for adjusting my glasses, etc.) and one “dirty” hand for touching anything outside my own person, on those rare occasions that I have to.  I scrub both hands when I get back.  I go once to the store a week (at most).  I put on a parka, hood, glasses, face mask, wear gloves, and I carry alcohol with me for rubbing down any surface I touch and sterilizing the gloves afterwards.  I back off from any encounter with a human being closer from 6 feet.  Assume that all members of my pod follow these same rules.  Which of these rules do you see us relaxing after the bent curve?  

I went to the SouthWest airlines website to see what comfort they might offer me for flying back to MA for the summer.  Lots of talk of bottles of purel on every counter and in every waiting area.  Apparently, some procedures have been altered concerning the verification of documents at check in to minimize actual exchange of physical objects.  No  snack service. They make a big deal about how they hose down the airplanes every night, but there is no attempt to screen passengers either at checkin or at boarding.  Is this sufficient?  Are flight attendants getting sick?  Does anybody know? 

Any way.  Sorry to run on.  Look at what Jim has to say, and help me to understand it. 

NIck

Nicholas Thompson

Emeritus Professor of Ethology and Psychology

Clark University

 <mailto:ThompNickSon2 at gmail.com> ThompNickSon2 at gmail.com

 <https://wordpress.clarku.edu/nthompson/> https://wordpress.clarku.edu/nthompson/

From: jpgirard at thinkingmetal.com <jpgirard at thinkingmetal.com> 
Sent: Wednesday, April 8, 2020 1:44 PM
To: thompnickson2 at gmail.com
Subject: RE: Fwd: [NICAR-L] AP/Johns Hopkins COVID-19 data

Nick,

Can't find the harvard study now, it was brought up at Friam 2(?) weeks ago.  The upshot I think was more than just avoiding overloading the medical system - it was amplifying herd immunity via quite selective quarantining/distancing and reducing the overall number of cases, not just spreading them out.

Consider: Let's say we develop an antibody test that is widely available.  And we determine the contagion is dependent on asymptomatic carriers, and we can roughtly guess based on priors (age, health, etc.).  A selective drawdown and reimposal of restrictions at different times & places could effectively amplify the herd immunity such that the R-effective for the virus is below 1, so it dies out and total cases over time is reduced.  I'm not even sure the antibody test is necessary.

Separately, thinking about it more, the prior system I described would result in a true sin/cosine wave.  There is the added effect of the delay between the forcing function (quarantining/distancing) and the resulting cases.

I'm thinking that any kind of feedback of the forcing function being monotonically increasing and negative to the resulting cases will cause some kind of oscillation, FWIW.

In engineering systems theory, this would be an underdamped system.  An overdamped system would result in an asymptote.  Not sure what in the real contagion world would correspond to overdamped, but some kind of asymptotic behavior is what seems to happen with most contagion.

Thinking more, it really it should be a complex/agent-based analysis, with the tension in the model similarly between the quarantining/distancing and the delayed results of cases/deaths.  Various levels and accuracy of testing could be programmed in as well.  Who needs a thesis?

And anyway, where is episims/transims or their progeny in all this?

Feel free to share this.  My new position will overwhelm me for a while - unfortunately no real time to spend congitating on this.

be well Nick.

-Jim

-------- Original Message --------
Subject: RE: Fwd: [NICAR-L] AP/Johns Hopkins COVID-19 data
From: <thompnickson2 at gmail.com <mailto:thompnickson2 at gmail.com> >
Date: Tue, April 07, 2020 9:03 pm
To: <jpgirard at thinkingmetal.com <mailto:jpgirard at thinkingmetal.com> >

Jim

Well, I think the state will be unsteady in any case, rather like Steve G;s spot fire analogy.  Every once in a while a spark will land from TX of CA.  We have to be able jump on it and surround it  As long as the capacity to identify and contain is robust, most people in nm can go about their business most of the time.  But do we even have such a robust strategy and would our society tolerate it.  

Is the Harvard strategy what I think of as the “leak” strategy.  You leak new cases into the community at a steady low rate until heard immunity is established?  People die, but relatively few because the hospitals are ready for them.  I think such a strategy might make sense in a world in which no vaccine is coming.  But let’s say a million people die of this thing which means, presumably, that a hundred million caught it.  That still leaves 200 million for the virus to eat.  Think how few people it took to sustain the recent measles epidemic. 

Jim, can I forward this correspondence to the list?  I think it’s fruitful and I would like to know that others think.

N

Nicholas Thompson

Emeritus Professor of Ethology and Psychology

Clark University

 <mailto:ThompNickSon2 at gmail.com> ThompNickSon2 at gmail.com

 <https://wordpress.clarku.edu/nthompson/> https://wordpress.clarku.edu/nthompson/

From: jpgirard at thinkingmetal.com <mailto:jpgirard at thinkingmetal.com>  <jpgirard at thinkingmetal.com <mailto:jpgirard at thinkingmetal.com> > 
Sent: Tuesday, April 7, 2020 8:45 PM
To: thompnickson2 at gmail.com <mailto:thompnickson2 at gmail.com> 
Cc: in.walkd.bud at gmail.com <mailto:in.walkd.bud at gmail.com> 
Subject: RE: Fwd: [NICAR-L] AP/Johns Hopkins COVID-19 data

All good questions Nick

I tend to believe there's usually more than one way to handle these things (well, there are at least two: shutdown and tracing/quarantine)  The harvard study to manage herd immunity is another, but that may qualify as your "unsteady state"

>From an engineering systems standpoint, the unsteady state could be caused by a system that accelerates proportional to the negative of position.  If position is the number of positive cases being found, and acceleration would be the second derivitive of that.  Does lockdown correspond to the 2nd derivitive of cases?  Have to think about that.  

-Jim

-------- Original Message --------
Subject: RE: Fwd: [NICAR-L] AP/Johns Hopkins COVID-19 data
From: < <mailto:thompnickson2 at gmail.com> thompnickson2 at gmail.com>
Date: Tue, April 07, 2020 6:06 pm
To: < <mailto:jpgirard at thinkingmetal.com> jpgirard at thinkingmetal.com>
Cc: < <mailto:in.walkd.bud at gmail.com> in.walkd.bud at gmail.com>

Jim, 

Can I get you to ponder a bit on the unsteady state that will obtain between curve flattening and vaccine/herd immunity?  Do agree that that state cannot be reached until we can get the number of new cases a day down to that number that can be contact-traced and contact isolated.  So using santa fe as an example, how many contact tracings and contact isolations do you think the City public health system can handle in a day, week, etc.  So let’s say I get the disease tomorrow but don’t get tested for three day.  So in that time I am within six feet of three people and I so tell the public health department.  Do the police go and “arrest” those three people and stick them in a motel somewhere?  And where do they take me?  

I really don’t see the conditions under which social distancing could be eased UNTIL there is a vaccine.

Those two new cases in China, today.  What happened to the patients and what happened to the n people they happened to have run into since they themselves became infectious?

Nick 

Nicholas Thompson

Emeritus Professor of Ethology and Psychology

Clark University

 <mailto:ThompNickSon2 at gmail.com> ThompNickSon2 at gmail.com

 <https://wordpress.clarku.edu/nthompson/> https://wordpress.clarku.edu/nthompson/

From: jpgirard at thinkingmetal.com <mailto:jpgirard at thinkingmetal.com>  <jpgirard at thinkingmetal.com <mailto:jpgirard at thinkingmetal.com> > 
Sent: Tuesday, April 7, 2020 5:44 PM
To: Tom Johnson <tom at jtjohnson.com <mailto:tom at jtjohnson.com> >; Steve Smith <sas at lava3d.com <mailto:sas at lava3d.com> >; Nick Thompson <thompnickson2 at gmail.com <mailto:thompnickson2 at gmail.com> >
Subject: RE: Fwd: [NICAR-L] AP/Johns Hopkins COVID-19 data

In the 20 days or so I've been using it, I've noticed a single problem, which was fixed within a couple of hours.

Of course, I'm only looking at a handful of countries in the daily reports.

-Jim

-------- Original Message --------
Subject: Fwd: [NICAR-L] AP/Johns Hopkins COVID-19 data
From: Tom Johnson <tom at jtjohnson.com <mailto:tom at jtjohnson.com> >
Date: Tue, April 07, 2020 4:34 pm
To: Jim Girard <jpgirard at thinkingmetal.com <mailto:jpgirard at thinkingmetal.com> >, Steve Smith
<sas at lava3d.com <mailto:sas at lava3d.com> >, Nick Thompson <thompnickson2 at gmail.com <mailto:thompnickson2 at gmail.com> >

============================================
Tom Johnson - tom at jtjohnson.com <mailto:tom at jtjohnson.com> 
Institute for Analytic Journalism   --     Santa Fe, NM USA
505.577.6482(c)                                    505.473.9646(h)
 <http://nmfog.org> NM Foundation for Open Government
Check out It's The People's Data <https://www.facebook.com/pages/Its-The-Peoples-Data/1599854626919671>                  

============================================

---------- Forwarded message ---------
From: Thomas Wilburn <tbone at thomaswilburn.net <mailto:tbone at thomaswilburn.net> >
Date: Tue, Apr 7, 2020 at 4:25 PM
Subject: Re: [NICAR-L] AP/Johns Hopkins COVID-19 data
To: <NICAR-L at po.missouri.edu <mailto:NICAR-L at po.missouri.edu> >

Yeah, I have to say that we've been working with the JHU data for a few weeks now and it has just been a total nightmare in terms of consistency and organization. All due respect to that team but it's clear that they were not really expecting the attention and have been scrambling to keep up. If you can use a dataset that has been better normalized and organized, you should.

Thomas

-----Original Message-----
From: Dan Nguyen <dan.cao.nguyen at GMAIL.COM <mailto:dan.cao.nguyen at GMAIL.COM> >
To: NICAR-L at PO.MISSOURI.EDU <mailto:NICAR-L at PO.MISSOURI.EDU> 
Sent: Tue, 07 Apr 2020 4:57 PM
Subject: Re: [NICAR-L] AP/Johns Hopkins COVID-19 data

You mean the timeseries/daily counts found here?
https://github.com/CSSEGISandData/COVID-19/tree/master/csse_covid_19_data

I think every covid-19 dataset is subject to the same real-world
limitations of testing availability (nevermind accuracy/timeliness of
bureaucratic reporting). That said, for my own U.S. analysis, I've switched
to the NYT repo for now, which has daily counts at the state and county
level: https://github.com/nytimes/covid-19-data/

I did do a spotcheck of a couple NYC counts, and in at least one instance,
NYT did a better job of reflecting a completed day's count, i.e. recording
the afternoon city report for a day, rather than an early morning. But
overall, I wouldn't be surprised if both datasets were accurate enough,
when averaged over several weeks of data points. For me, the thing I like
about the NYT's data is that it's already normalized and tidied by standard
conventions.

-Dan

On Tue, Apr 7, 2020 at 4:23 PM Anthony Cave <Anthony.Cave at kxan.com <mailto:Anthony.Cave at kxan.com> > wrote:

> Just curious, how are y'all using the AP Johns Hopkins COVID-19 data, if
> at all?
>
> It seems like you'd need an editor's note/margin of error, given that it's
> noted that the data relies on testing availability as opposed to true
> infection rates.
>
> Was just curious what others thought about it.
>
> Thanks,
>
> Anthony Cave
> Investigative Producer/Reporter, KXAN News
> 305-467-7121 (cell)
> ==================================================================== To
> unsubscribe from NICAR-L, please send "unsubscribe NICAR-L" in the body of
> an e-mail message to "listserv at lists.missouri.edu <mailto:listserv at lists.missouri.edu> ". Please e-mail
> listmaster at ire.org <mailto:listmaster at ire.org>  if you need help or have questions.
> ====================================================================

====================================================================
To unsubscribe from NICAR-L, please send "unsubscribe NICAR-L" in the
body of an e-mail message to "listserv at lists.missouri.edu <mailto:listserv at lists.missouri.edu> ".  Please
e-mail listmaster at ire.org <mailto:listmaster at ire.org>  if you need help or have questions.
====================================================================

 <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=icon> 

Virus-free.  <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=link> www.avast.com 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://redfish.com/pipermail/friam_redfish.com/attachments/20200408/1cd2e92a/attachment.html>