sokolow

anyway, today is the first day I have caught all the way up on the ‘rona news and not been grimly depressed.  UW-IHME sees evidence that social distancing actually works,

https://twitter.com/IHME_UW/status/1247237645945704455?s=20

And the incorporation of new data into the IHME model indicates the plausible possibility that IL might not go up against the wall on its medical resource capacity

I thought you all might find this comparison interesting.  a biostatistician asks this:

And a meteorologist from EUMETSAT replies like yea:

1 hour ago, Baum said:

I was referring more to the fact it's being politicized. I have no doubt there is a lot that goes into it on a professional level.

It’s been political since the industrial revolution, or for a more specific starting point since the radium girls and since  SoL Frances Perkins hulked out hardcore about silicosis in wake of the protracted horror of the Gauley Bridge miner’s disaster. Because in many cases it invokes a proper regulatory question.  I suppose the best way to look at it is, “it is automatically political, how can we make it productively political”

44 minutes ago, Baum said:

What a world. We debate how someone should be classified in death

I mean, its not just something that only became relevant for the ‘rona, mortality and injury data is also fundamental to public health policy nationally and internationally, for doing demographic and actuarial research, for IDing patterns of domestic abuse & child abuse, for understanding social inequality, for workplace safety & health, to understand emerging health concerns, and so on. my wife tells me its also not so simple to do a good job assessing a cause of death & that its an art and a science worth studying

edit also, sad to say, its important to law enforcement generally and also for tracking elder abuse

@Hoosier

WaPo had an article about this just today

https://www.washingtonpost.com/investigations/coronavirus-death-toll-americans-are-almost-certainly-dying-of-covid-19-but-being-left-out-of-the-official-count/2020/04/05/71d67982-747e-11ea-87da-77a8136c1a6d_story.html

Quote

Coronavirus death toll: Americans are almost certainly dying of covid-19 but being left out of the official count

Chicago crew has best Stern Mayor Quarantine Memes

 

 

15 hours ago, Stebo said:

I would bet their cases are in the millions and deaths in the 10s of thousands easily.

Probably, and I’ve seen people who aren’t running an anti-chinese paranoia line treat 30k deceased but not and/or not yet identified as c19 cases as plausible.

It makes sense that given that Vincent Racaniello was willing to speculate actual infections in the USA were likely over 1M by end of last week, and we’re going to have the same issue as other countries of having to wait and do post-event analyses of excess deaths against baseline to sort out what was the potential number of fatalities that might have been the ‘rona that just weren’t lab-confirmed

I try to keep reminding myself the number on the tracker is not the Real True Number of C19 deaths, its a incomplete and provisional number given the constraints of data collection and reporting during a dangerous, politically charged, potentially historic pandemic.  

1 hour ago, Baum said:

But wouldn't the same concerns exist in the H1N1 pandemic?  Also, is there a site that shows actual hospital bed availability as this virus continues to ramp up? 

In what sense? My recollection of the debrief on 2009 is public health officials in the USA and at WHO saw it as a reassortment of H1N1 that was relatively novel, they were concerned about its apparent lethality when it first appeared in Mexico, they were concerned about its apparent rapid international spread. The response in the USA was to ramp up public and inter/intragovernmental, communication about it in the sense of “dust off your pandemic plan,” to raise risk measures, to issue advisories about testing, monitoring, reporting, and school closings, and to push the production pipeline on strain appropriate vaccine and antiretrovirals

And in the end the lab confirmed death counts were low, estimated worldwide deaths on par with seasonal flu, and it was a bit clinically weird in that younger people had worse outcomes. Like a weather high risk day that blows low

friend of mine opened her weather app and discovered a strong entry for ’nonoptimal advisory messaging’

Quote

 

SUNDAY 
The Sun will turn into a giant red angry virus ball. 

Deceased: N/A 

 

Veterans have spoken out against the decision to relieve the captain of the U.S.S. Theodore Roosevelt after he sent a letter to the Navy pleading for help after his ship was stricken with the coronavirus.

https://www.newsweek.com/coronavirus-navy-ship-captain-brett-crozier-1495923

About the rollout of antibody AKA “was that nasty flu I had in early March the ‘rona and I didn’t know it” test, I thought this was informative.

1 hour ago, hoosierwx said:

Good point. Those areas east of the Rockies are pretty socially isolated anyway and usually it's a pretty good drive to get anything. The SE is a bit worrying however.

Sent from my SAMSUNG-SM-G935A using Tapatalk
 

I can’t immediately find to credit the person who did the quick overlay. i will try to.

but she pointed out the NYT movement map and the map of “limited transport / no market within a mile” standard about what social geographers & food security people call food deserts are strikingly similar.  no consideration in the travel headline of whats trying to be measured, same with the cell carrier data about driving graphic from last week

10 hours ago, Toro99 said:

Food for thought, not sure how reliable the data is.

 

 

this turns out to be an excellent map of far it is to a grocery store

If you are interested in reporting lag and how it affects data presentation of e.g. things like the “epidemic curve” and whether the curve is flattening or not, here is an interesting thread using Ontario as a case study

who all do you think has the best data vis site? i actually think some of the weirdo projects have more intuitive portals than JHU or STAT or the FT (https://www.ft.com/coronavirus-latest) and the worldometer is fastest glance

https://coronavirus.1point3acres.com/en

its the county level data and testing tabs that i think are real helpful, tho i dunno, maybe the name brand institutional sites have better QA? 

2 hours ago, Snownado said:

With everyone on lockdown, where are these new cases coming from ? Maybe from people who were infected weeks ago before measures were put in place ?

1 hour ago, beavis1729 said:

Not sure if this perspective has been discussed much on here...but if you can get past his angry and egotistical style of writing, it's an interesting hypothesis.  Most importantly - if this hypothesis is really true, what is the best solution going forward?

https://market-ticker.org/akcs-www?post=238738

well its an embittered, polemical perspective that swings past issues which are actually relevant such as: 

its wise to have provisions for health care workers to have separate residences available so they don’t infect their families or general public if they get exposed. spouse’s institution has a plan like that for our family. 

hospitals are huge vectors, nosocomial infections are a Big Deal not just in pandemics but in everyday life

and we’ve seen legit instances where the hospital or the medical supply chain was part of the problem in a pandemic, in the last 50 years. 

but like. these are things that are understood to exist and be worthy of serious study & serious policy. but apparently the author of that post is a crank who, idk, hates an ex lover who was a nurse or doctor, or thinks clinical staff should live like monks in weird health cloisters, and who also has a paranoid conspiracy theory of disease

update now laughing uncontrollably at vision of gamestop manager hulking the f out on his geocities era design covid truther blog b/c they shuttered his storefront then sheriff rousted his SCA sword enthusiasts club from the park, frothing @ paragraph 15: meanwhile my b!tch exwife tina is swanning around in her nurses scrubs buying groceries 

This is what I’m reading r/n,

via this guy,

https://bedford.io/team/trevor-bedford/

ime if you like following along & getting a glimpse into how expert teams of science professionals do research on & create frameworks for understanding emergent phenomena with high potential for community impact, then its a must-click

5 hours ago, madwx said:

Now that the decade of March is over with it's time to look forward to the eon of April.  Looks to be starting off on a warming trend for the region with storminess becoming more common from Friday through much of next week.

Last year it seemed like March was three months long, and then the last week of November seemed to continue for endless months as well. Subjectively for me it was unrelenting gloom, felt like vacationing in Scotland. Constant freeze thaw. Never wet enough to justify the full on overboots, but damp enough that with three pairs of workboots in rotation the first would dry in time for the third to come home soaked. Maybe just bad luck that this winter and an extensive below-grade excavation coincided, so there was no getting out of it.

Should have bought the GI moon boots.

2 hours ago, sokolow said:

There was a really good roundup in that guys circle as it were, of epidemiological modeling successes and partial successes (as seen so far) in what has verified, and been useful prior to verification, lemme see if i can find it

here is is

https://twitter.com/AdamJKucharski/status/1244549516352720897?s=20

Before I do that though this podcast episode

https://www.microbe.tv/twiv/twiv-special-lipkin/

was fantastic, covering what its like to have mild/moderate disease, examining public health responses & outcomes/potential outcomes in China and the USA, putting SARS-COV2 in context of other human viruses, talking vaccine / testing / treatment timelines

the title says ‘interview with a covid patient’ but the patient is this guy

W. Ian Lipkin

John Snow Professor, Epidemiology, Professor of Neurology and Pathology and Cell Biology, Director, Center for Infection and Immunity

perhaps best known to us plebs as

Quote

... the chief scientific consultant for the Soderbergh film "Contagion" ...

There was a really good roundup in that guys circle as it were, of epidemiological modeling successes and partial successes (as seen so far) in what has verified, and been useful prior to verification, lemme see if i can find it

14 hours ago, Jackstraw said:

  I've seen some models as low as 20k total and some as high as 500k total with current restrictions in place.  While I agree that modeling this pandemic is a great tool, the accuracy of this modeled data has yet to be seen.  More important than forecasting when this thing will peak is modeling the data of critical care needs to available resources in specific areas as we move forward.  That's the serious data in the short term that these models can provide to help get emergency supplies, personnel and equipment where it's needed before those areas get overwhelmed.  

  Kinda surreal 30 days ago we were throwing around weather model projections for snow and now were moving into Covid19 model projections.  If anybody knows how to nit pick computer modeling it's weather nerds lol.

This guy is a bioinformatics professor at Uni Washington and he did a snap review of the UW above study and a followup thread

Quote

Carl T. Bergstrom (@CT_Bergstrom) Tweeted:
1. On Friday night I posted a rapid post-publication peer-review of the @IHME_UW model and associated white paper. Here's the IHME website and projections: https://t.co/rveXDBFUjc

https://twitter.com/CT_Bergstrom/status/1244815009303023616?s=20

Where he highlights potential strengths & weaknesses of the modeling, as well as the public & policy impact pitfalls inherent to the messaging on something that’s deeply contingent on how policymakers, professionals, community leaders, and laypeople then act

This forecasting exercise will come up, so I figured I would post it embedded as this guy works through the inbuilt assumptions. Click thru show thread, 20 odd tweets in chain, no 1-page collated version because thread aggregator was janky for me.

Must read for model lovers, cause for cautious optimism and serious concern

if your employer has a closet full of PPE sitting around

https://www.projectn95.org

if you yourself have an open box of n95s or what have you, check with all the cops, EMTs, fire rescue, RNs and doctors you personally know & make sure they got a stash. they’re already rationing PPE at my wifes workplace and i have been told it makes a huge difference to know you have a new clean mask, bunny suit, whatever waiting if the one you’ve got gets soiled beyond use; an emergency airway w patient in distress eg is not a clean affair

same with niosh rated half / full mask painting & remediation gear if they’re in good condition and you have spare clean filter sets