Coronavirus

[pescetom]

From The Coronavirus, by the Numbers by James Gorman at NYT:

 

A lot of good sense here, in particular "We’ve found for things like Covid-19, it’s close-knit interactions that seem to be most important." is relevant for anyone still playing face to face bridge in a club with elderly people, even if it is permitted.

[Winstonm]

The other side of this story:

 

Here’s a tweet thread which runs the numbers based on our current understanding of COVID-19.

 

 

Liz Specht

@LizSpecht

I think most people aren’t aware of the risk of systemic healthcare failure due to #COVID19 because they simply haven’t run the numbers yet. Let’s talk math. 1/n

 

87.2K

8:03 PM - Mar 6, 2020

Twitter Ads info and privacy

37.8K people are talking about this

 

The follow-on gut punch: if the states and federal government do not develop and implement a comprehensive plan to mitigate contagion, the U.S. will run out of hospital beds in early May.

 

That’s in a little over eight weeks.

 

If we don’t have adequate beds let alone mechanical ventilators and intubation equipment, the mortality rate will jump from an estimated 2-2.3% to at least 5%.

 

If you don’t read the twitter feed, understand that the writer is an engineer with a Chemical & Biomolecular Engineering degree from Johns Hopkins and her PhD from UCSD - no lightweight assessment.

 

The risk of coronavirus is not to any one individual but whether or not it systemically overwhelms the healthcare systems, which in turn could create higher mortality for non-virus cases, not to mention the economic damage done.

[barmar]

If you don’t read the twitter feed, understand that the writer is an engineer with a Chemical & Biomolecular Engineering degree from Johns Hopkins and her PhD from UCSD - no lightweight assessment.

But is she a "stable genius" with a "hunch"?

[pescetom]

But is she a "stable genius" with a "hunch"?

No need to be a stable genius, just look at Lombardy, wealthiest region of Italy, population 10 million.

Only 16 days after outbreak, 2217 infected in hospital and 399 in intensive care (many in beds liberated by the 227 already dead).

The region is already exporting intensive care cases to surrounding regions and the possibility to increase ICU facilities in short term is limited.

10% of doctors and nurses are quarantined for possible infection, the rest are at the limit and doctors in pension are being sought.

The epidemic numbers have been growing at a steady logarithmic rate which shows no sign of declining despite containment measures.

Even without a hunch it is evident that very soon a large proportion of those who require intensive care (not just from virus) will not receive it.

 

Today's Italian statistics: infected 7375 (+45%), deaths 366 (+57%), no longer infected 622 (+6%). Ratio of deaths/cases 4.1%.

That's a bigger rise than usual even on a logarithmic scale, no change in criteria or death rate either.

[barmar]

What's surprising me when I hear numbers like that is how slow the increase has been here in the US, especially considering the limited resources that have been applied to the outbreak (e.g. we don't have enough testing kits). How is it that our total cases is still under 200?

 

This may be why Trump thinks this thing is blown out of proportion.

[awm]

What's surprising me when I hear numbers like that is how slow the increase has been here in the US, especially considering the limited resources that have been applied to the outbreak (e.g. we don't have enough testing kits). How is it that our total cases is still under 200?

 

This may be why Trump thinks this thing is blown out of proportion.

 

If you want to minimize the number of diagnosed cases, it helps not to test anyone, which is basically what is going on in the US. Given the overall mortality rates and 22 confirmed deaths in the US it’s a good guess there are >2000+ cases and not the 500 reported.

[kenberg]

If you want to minimize the number of diagnosed cases, it helps not to test anyone, which is basically what is going on in the US. Given the overall mortality rates and 22 confirmed deaths in the US it's a good guess there are >2000+ cases and not the 500 reported.

 

For me, this is fundamental. And we do not need to assume any sinister reasons for non-testing, we simply don't have the capacity to do what is needed so we prioritize. Mathematics could be of some use in estimating the under count but I strongly suspect that the main mathematical conclusion would be that we do not yet know. It's apt to be one of these butterfly things where a small flap of the data wings will produce a large change in the estimates.

 

We have a president who says anything he wishes and thinks anything to the contrary is fake news. This can destroy sensible caution in the interpretation of data.

 

There was a cartoon a few days ago that captured my views pretty well. It showed a guy at his computer saying something like "It's amazing. All of my FB friends that six weeks ago were constitutional law experts have now become experts in viral disease transmission".

[pescetom]

If you want to minimize the number of diagnosed cases, it helps not to test anyone, which is basically what is going on in the US. Given the overall mortality rates and 22 confirmed deaths in the US it’s a good guess there are >2000+ cases and not the 500 reported.

 

Another way to minimize is to ascribe all the deaths to other illnesses, which is presumably what Germany is doing (1151 reported cases, 0 deaths).

 

Yesterday's unusual jump in Italy - infected 7375 (+45%) - was silently corrected later to 6387 infected (but still 366 dead +57%).

Today's statistics in Italy: infected 7985 (+25%), deaths 463 (+25%), no longer infected 724 (+16%). Ratio of deaths/cases 5.3%.

So we're back to the usual logarithmic growth, in the first day of a massive clamp-down on most of northern Italy.

Yesterday we had people panicking to get out of Milan to the south while they still could, today revolts in prisons with shootings and a mass escape.

[cherdano]

If you want to minimize the number of diagnosed cases, it helps not to test anyone, which is basically what is going on in the US. Given the overall mortality rates and 22 confirmed deaths in the US it’s a good guess there are >2000+ cases and not the 500 reported.

Small but important correction.

It seems most deaths occur 10-14 after infection. So if there are 20 confirmed deaths in the US, this means there were probably already 2000+ cases in the US 10-14 days ago.

 

Similarly, Italy must have a huge number of undiagnosed cases (unless Italians have a much much higher fatality rate than e.g. South Koreans).

[cherdano]

Another way to minimize is to ascribe all the deaths to other illnesses, which is presumably what Germany is doing (1151 reported cases, 0 deaths).

There are 2 reported deaths now. And that is completely consistent with the number of cases reported 10-14 days ago, see https://en.wikipedia.org/wiki/2020_coronavirus_outbreak_in_Germany#Statistics.

[cherdano]

And we do not need to assume any sinister reasons for non-testing,

We don't need to assume. But when you have a president saying clearly out loud "I like the numbers where they are", and his economic idiot advisor repeatedly proclaiming that the virus has been "contained", it becomes just a little bit harder to make that assumption.

 

In fact, as some friendly advice for our American citizens on this board: in a normal government, it is health experts not economic experts commenting on the status of an epidemic. You might try getting yourself one. (A normal government.) Though to be fair to Larry Kudlow, it's probably best not to consider him an economic expert, either.

[Winstonm]

A few years ago, China rode out their H5N1 flu epidemic with hardly a ripple; with Covid-19, the Chinese government has sacrificed at least a month of GDP to try to halt its spread.

 

This ain't the flu, Matey. Pretending otherwise does not address the reality.

[y66]

From Sandi Doughton at the Seattle Times (March 8)

 

Testing for the novel coronavirus in the Seattle area will get a huge boost in the coming weeks as a project funded by Bill Gates and his foundation begins offering home-testing kits that will allow people who fear they may be infected to swab their noses and send the samples back for analysis.

 

Results, which should be available in one to two days, will be shared with local health officials who will notify those who test positive. Via online forms, infected people can answer questions about their movements and contacts, making it easier for health officials to locate others who may need to be tested or quarantined, as well as to track the virus’ spread and identify possible hot spots.

 

The goal is to eventually be able to process thousands of tests a day, said Scott Dowell, leader of coronavirus response at the Bill & Melinda Gates Foundation. The project is ramping up as quickly as possible, but it’s not clear exactly when it will launch, he added. Among other things, software needs to be upgraded to handle the expected crush of requests, and a detailed questionnaire finalized for people who request tests.

 

“Although there’s a lot to be worked out, this has enormous potential to turn the tide of the epidemic,” Dowell said.

 

While Public Health – Seattle & King County has confirmed 71 cases and 15 deaths as of Saturday, modeling by Trevor Bedford, a computational biologist at Fred Hutchinson Cancer Research Center, estimated on Wednesday that the actual number of cases in the Seattle area was about 600. Unchecked, that could theoretically increase to 12,000 cases – and possibly as many as 30,000 – by the end of March, according to projections from Mike Famulare at the Institute for Disease Modeling in Bellevue. But steps to slow transmission can significantly reduce the number of new infections, underscoring the importance of acting quickly to protect people from the virus.

 

The new effort aims to leverage the formidable resources and expertise of the Gates Foundation, known for fighting disease and epidemics around the globe, to assist local health agencies struggling to keep up with a fast-moving outbreak. The Seattle area has emerged as an epicenter of the new disease, with far more cases and deaths than any other U.S. city.

 

“One of the most important things from our perspective, having watched and worked on this in other parts of the world, is the identification of people who are positive for the virus, so they can be safely isolated and cared for, and the identification of their contacts, who can then be quarantined,” Dowell said.

 

But testing has been limited until now, leaving many people frustrated — and frightened. Last week, a laboratory at UW Medicine got approval to begin processing specimens collected by physicians and other health care providers. The Gates-funded project will reduce the need for sick people to visit a doctor’s office or clinic, lowering the chance of exposing others.

 

The initiative grew out of the Seattle Flu Study, a 2-year-old research project based at the University of Washington to track the spread of infectious diseases like influenza. Funded with $20 million from Bill Gates’ private office, the project recruited thousands of volunteers and sent them self-test kits. The focus has now shifted to the new coronavirus, using similar methods to aid the public-health response.

 

The expanded testing program to deliver kits to the public will be supported by a separate group within the Gates Foundation. When the system is up and running, people in the Seattle area who think they might be infected with SARS-CoV-2, the scientific name for the new coronavirus, can fill out a questionnaire online. If their symptoms are consistent, they can request a test kit, which will be delivered to their home within two hours. The swabs will be collected and delivered to a lab for processing.

 

In the meantime, the Seattle Flu Study itself will be increasing its testing capacity to help public health staff investigating the outbreak understand how coronavirus is moving through the community.

 

The Gates Foundation recently announced it’s committing $5 million for coronavirus response in the Seattle area, and much of that will go for expanded testing and analysis. While the initial focus of the test-kit effort will be on the Seattle area, it could eventually expand statewide – and beyond, Dowell said.

 

Outside of King County, one person has died and more than 30 infections have been confirmed as of Saturday.

 

A major goal of the project is to collect as much information as possible online, which will ease the burden on health officials who are stretched thin and hard-pressed to investigate every new case. Local resources have been focused on Life Care Center, the Kirkland nursing home that accounts for the majority of deaths.

 

“They simply don’t have enough epidemiologists to do the shoe-leather epidemiology, the house-to-house case identification,” Dowell said.

 

The Seattle Flu Study already has contributed greatly to the understanding of COVID-19, the respiratory disease caused by the new coronavirus. As the outbreak started in China, the scientific team, co-led by Dr. Helen Chu, an infectious-disease specialist at UW Medicine, quickly developed a genetic test for the virus, similar to one they used for flu.

 

A physician who knew about the work sent in a sample from a teenage patient suspected of having the disease, and the lab was able to identify what was only the second case in the state at that time.

 

The flu-project scientists also did the first genetic analyses of new coronavirus cases in Washington, and will continue that work. Bedford, the computational biologist, used those first genomes to analyze changes in the virus over time and concluded that it had probably started circulating in the state earlier than anyone realized.

 

The Seattle Flu Study has also already been collecting nasal swabs from volunteers for a research study on the new coronavirus. People can still sign up for that study, but they cannot get their individual results yet and home-testing kits are not yet available from the new Gates-funded program.

 

The Seattle Flu Study is led by the Brotman Baty Institute in collaboration with UW Medicine, Fred Hutch and Seattle Children’s hospital.

 

The Gates Foundation has also committed $100 million to the global coronavirus response, with an emphasis on vaccine and drug development and improved testing, treatment and control in vulnerable parts of Africa and South Asia.

 

Note: This story has been updated to reflect additional information provided by The Bill & Melinda Gates Foundation, distinguishing between the Seattle Flu Study and a new project offering home test kits to the public for the new coronavirus, which will be supported by a separate group in the foundation.

 

The story has also been updated to clarify that projections of a possible 30,000 infections in the Seattle area by the end of March represent the upper end of modeling results for unchecked spread, with 12,000 cases as a more likely number. The researcher who did the modeling is also identified.

[Zelandakh]

What I find tragic is that this stupid virus might take Trump down when children in cages didn't..

I would humbly suggest that grabbing women's pussies without their consent is worse and still cannot believe that Americans actually elected a President who admits to doing so.

[Zelandakh]

(unless Italians have a much much higher fatality rate than e.g. South Koreans).

Actually they do. There are (at least) 2 different strains of coronavirus (S and L) going around and the one prevalent in Italy (L) is stronger, more deadly and more aggressively transmitted than the one that initially dominated in East Asia (S). The strange thing is that according to Italian authorities, the strain there in Italy is the same as that that arrived in Bavaria (where I live) and set the ball rolling in Germany and yet, as pescetom mentioned, the death rate in Germany is unusually low. The good news is that the 2 strains are genetically so similar that they should not require different vaccines and a person should not get infected with one after having already suffered the other.

[pescetom]

There are 2 reported deaths now. And that is completely consistent with the number of cases reported 10-14 days ago, see https://en.wikipedia.org/wiki/2020_coronavirus_outbreak_in_Germany#Statistics.

 

As that page confirms, in late January Bavaria was home to the first outbreak in Europe which then spread to southern Lombardy, so the timescales for Germany are basically comparable to Italy. In the last few weeks Germany has been more or less keeping pace with France and Spain (now doing a sprint) in terms of growth in number of cases, we currently have:

Germany 1281 cases, 2 dead

France 1412 cases, 30 dead

Spain 1500 cases, 35 dead

 

Italy 9172 cases, 463 dead

 

Clearly there are different criteria for case reporting in different countries and even in different moments (both China and Italy changed their criteria multiple times). But the difference in reported deaths between Germany and the rest of Europe is evident and not explained by any difference of timing that I can see.

[Winstonm]

This is not peer reviewed but seems valid:

 

COVID-19 can be spread before it causes symptoms, when it produces symptoms like those of the common cold, and as many as 12 days after recovery, according to a virologic analysis of nine infected patients published today on the preprint server medRxiv.

 

The findings contrasted starkly with those from the 2003 outbreak of SARS in terms of viral load. "In SARS, it took 7 to 10 days after onset until peak RNA concentrations (of up to 5x105 copies per swab) were reached," the researchers wrote. "In the present study, peak concentrations were reached before day 5, and were more than 1,000 times higher."

 

"Social distancing is the most effective tool we have right now."

[pescetom]

Similarly, Italy must have a huge number of undiagnosed cases (unless Italians have a much much higher fatality rate than e.g. South Koreans).

That's for sure. Angelo Borrelli, the head of the Civil Protection Department, said yesterday that they are now only testing people who seek assistance and present evident symptoms, which whatever it means is quite different to how they proceeded a couple of weeks ago (testing anyone under suspicion of contagion, even entire villages). And of course the fatality rate would be shocking without the explanation of undiagnosed cases.

 

Today's statistics in Italy: positive 8514 (+7%), deaths 631 (+36%), no longer infected 1004 (+39%). Fatality rate 6.3%.

So a dramatic daily drop in new positives but an equally dramatic jump in deaths. Looks more like a hiccup in the regional reporting system than a first sign of deceleration, but we will see in the next few days. 877 people in intensive care.

[pescetom]

Today's statistics in Italy: positive 10590 (+24%), dead 827 (+31%), no longer infected 1045 (+4%). Fatality rate 7.1%.

As I imagined, there was a hickup yesterday with incomplete data from Lombardy. Over the last two days the increase in positives is "only" 32% which is a significant logarithmic drop from the usual 50+%, indicating that the nationwide measures and a change in awareness are at least having some effect.

The dramatic jump in fatality continues however with almost 200 dead in one day compared to just 41 declared cured.

That represents 13% of deaths from all causes, just below coronary heart disease and 21 times road accidents.

[shyams]

The global statistics for the day are 6,136 new cases recorded yesterday.

[Winstonm]

To all my right-wing friends and family:

 

Now that the WHO has announced a pandemic, you guys who have been told not to worry, it's nothing, the flu is worse, etc. - well, you might want to consider other sources of information.

[Zelandakh]

Now that the WHO has announced a pandemic,

Pandemic is a scary word but the criteria for it are really not. Pan means "across" and "-demic" is disease". I suspect even the most bucket-headed conservative has noticed that we have a disease spanning multiple countries! Yes I know the WHO definitions includes some additional criteria - new strain, sustained transmission etc - but using the word itself to scare people is not sensible. Take precautions and avoid unnecessary risks sure, but let's not be part of spreading any misinformation or stoking up panic!

[cherdano]

As that page confirms, in late January Bavaria was home to the first outbreak in Europe which then spread to southern Lombardy, so the timescales for Germany are basically comparable to Italy.

But initially the outbreak was contained in Germany, it took until late February for the numbers to rise.

[Winstonm]

Pandemic is a scary word but the criteria for it are really not. Pan means "across" and "-demic" is disease". I suspect even the most bucket-headed conservative has noticed that we have a disease spanning multiple countries! Yes I know the WHO definitions includes some additional criteria - new strain, sustained transmission etc - but using the word itself to scare people is not sensible. Take precautions and avoid unnecessary risks sure, but let's not be part of spreading any misinformation or stoking up panic!

 

For sake of brevity, I excluded many details why caution is warranted rather than the "Oh, it's nothing" coming from some.

[cherdano]

Actually they do. There are (at least) 2 different strains of coronavirus (S and L) going around and the one prevalent in Italy (L) is stronger, more deadly and more aggressively transmitted than the one that initially dominated in East Asia (S). The strange thing is that according to Italian authorities, the strain there in Italy is the same as that that arrived in Bavaria (where I live) and set the ball rolling in Germany and yet, as pescetom mentioned, the death rate in Germany is unusually low. The good news is that the 2 strains are genetically so similar that they should not require different vaccines and a person should not get infected with one after having already suffered the other.

What is the evidence for the claim that (L) is stronger? I.e., I am saying [citation needed]