Covid 19 virus

[metrologist]

Another day, another data point as the virus steadily marches up the exponential curve here in the US.  As of yesterday fear has suddenly prompted policy makers into action.  There are new international travel restrictions.  Universities are closing.  Grade schools are winding down activities and telling parents to prepare for closures.  Finally we have policy makers responding.  Unfortunately, it may be too late.  There are still lots of misleading instructions and advice being put out by community, press, and government leaders.  Still many are saying the risk is low.  Clearly few understand what 'exponential' means.  If they did, they would see that on the current growth track the spread could result in about a million deaths in the US by about 30 days from now.

Your chart is interesting. I think we are using the same data.

[ebastler]

Your chart is interesting. I think we are using the same data.

I would recommend against combining a linear and a log scale in the same chart. And I certainly recommend against showing a spline interpolation through set of discrete datapoints, even more so when done without showing the underlying datapoints themselves. That (a) suggests a resolution which is not there, and (b) suggests spurious dips and peaks, which are artefacts from the interpolation.

Electronics content: Rigol is apparently guilty of the latter sin in their scopes. 

[whalphen]

Your chart is interesting. I think we are using the same data.

I'm using the data from the worldometer website.  We have to take the data with a grain of salt because we know it's understated due to limited testing.  But to get a glimpse of what's to come, it does the job.  (Though I sometimes wish I hadn't looked and seen such a frightening picture.) 

Those who criticize the chart are missing the point.  This data is publicly available and anyone can use it to make their own projection in whatever format they prefer.  It doesn't change the fact that spread is exponential and many people may suffer and/or die.

We all need to recognize what's coming and do what we can to protect our families, communities, and ourselves.  Our only control over this plague is to slow the spread -- which will reduce the strain on health services and, thus, save lives.  Each and every one of us can take actions to slow the spread.  Trying to assign blame will do nothing to slow the spread.  Cooperation, improved hygienic practices and social isolation are our greatest tools at this point.

[Simon]

Sadly governments are doing very badly. In the UK it has been announced that we are moving from the "containment" stage to the "delay" stage. Well fuck me if someone can tell me what he "containment" plan was other than to do nothing I'd be very grateful! So far as i can tell it was do nothing but ask people to stop panic buying loo roll. Public Health England, the people telling the government how to stop this are having a mass gathering of employees from more than one location for training. I mean are we stupid or something? apparently yes!

[ebastler]

Those who criticize the chart are missing the point.  This data is publicly available and anyone can use it to make their own projection in whatever format they prefer.  It doesn't change the fact that spread is exponential and many people may suffer and/or die.

I was only criticizing metrologist's way of plotting the data. Didn't mean to question your point, and I don't think I said anything to that effect.

I have been looking at the German data in much the same way, and have pointed out the (so far) unmitigated exponential growth in a couple of posts here. In Germany, the rate has even been slightly higher -- doubling pretty exactly every 2.0 days. No idea whether that difference is due to a higher rate of tested and detected cases, a significant influx of infected travellers from Italy, or other reasons.

The conclusion seems clear: Unless we manage to fundamentally change people's behaviors, and hence slow down the spread, we are in for a collapse of the health systems.

[Cerebus]

Sadly governments are doing very badly.

It takes special "government grade idiocy" to handle this as badly as some governments have been doing. I think so far, of those I have heard about, the US takes the biscuit:

The Donald Trump White House told the federal Department of Health and Human Services to classify all information from top-level coronavirus meetings.

The decision to keep all coronavirus deliberations classified made it hard for public health and medical professionals to access vital information. The restricted flow hampered the U.S. government’s response to the contagion, Reuters reports, citing four Trump administration officials.

“The officials said that dozens of classified discussions about such topics as the scope of infections, quarantines and travel restrictions have been held since mid-January in a high-security meeting room at the Department of Health & Human Services (HHS),” report Wednesday Aram Roston and Marisa Taylor at Reuters:

    Staffers without security clearances, including government experts, were excluded from the interagency meetings, which included video conference calls, the sources said.

    “We had some very critical people who did not have security clearances who could not go,” one official said. “These should not be classified meetings. It was unnecessary.”

    The sources said the National Security Council (NSC), which advises the president on security issues, ordered the classification.”This came directly from the White House,” one official said.

    The White House insistence on secrecy at the nation’s premier public health organization, which has not been previously disclosed, has put a lid on certain information - and potentially delayed the response to the crisis.

In the UK it has been announced that we are moving from the "containment" stage to the "delay" stage. Well fuck me if someone can tell me what he "containment" plan was other than to do nothing I'd be very grateful!

Sorry, can't t help there. As far as I can tell the government didn't have a plan. Fortunately the NHS was a bit more on the ball. I think the "containment" state is just a bit of back-naming for "We'll contain ourselves until someone tells us what to do".

[Kjelt]

My place of work they shifted to two teams work, one team may work at the office on even weeks the other on odd weeks. Temperature check at the entrance, cafeteria closed, not allowed to use the keyboard and mouse on the shared office space, no meetings >5 people, and work as much as possible from home.
Never had this in 25 yrs on the job.

[Simon]

Well I think Trump pretty much declared himself an expert in covid-19 so, sorted!

[Simon]

My place of work they shifted to two teams work, one team may work at the office on even weeks the other on odd weeks. Temperature check at the entrance, cafeteria closed, not allowed to use the keyboard and mouse on the shared office space, no meetings >5 people, and work as much as possible from home.
Never had this in 25 yrs on the job.

I doubt my employer even has a backup plan. We struggle to get our IT to work as it is with the IT staff located at our parent company. I am one of few with a company laptop with emails already on it and i could borrow out the only 3D CAD floating licence the company has and go home but i mostly use KiCad at work and the libraries live on my portable hard drive anyway. The mere fact that I can in 30 secands take out the 3D CAD licence is only because I made them put the floating licence on the laptop for me.

Electronics wise I am better set up an home than at work.

[Kjelt]

Yes well issues do occurr, VPN service overcrowded so not everyone can get in.
People that can't reach colleagues anymore.
Ever tried a skype call with more than ten colleagues? 

[not1xor1]

https://www.reddit.com/r/medicine/comments/ff8hns/testimony_of_a_surgeon_working_in_bergamo_in_the/?utm_source=share&utm_medium=ios_app&utm_name=iossmf

We now know that the virus had already made a few victims there (Bergamo and Codogno areas) before anybody noticed that.
Just too many people do not make the due vaccination against ordinary flue so at first they thought it was just that, ordinary flue.

In the meantime the virus had already taken foot spreading to many unknowingly people.
In Italy, we now have more than 15000 cases.
Since they are testing just people who shows symptoms, that likely means that there are already 50k-200k cases who got the virus and have never shown any symptom or are yet at a very early stage of the disease.

Average incubation time is 5.1 days, while 14 days of quarantine covers just 99% of cases (as recently reported by John Hopkins University).
So we have to wait a few days to see if the measures taken so far are working.

While COVID-19 is not as bad as SARS or MERS, it will kill many people and will cause severe problems to the economy.
It will be fine if we can learn from that (at a regional level, e.g. EU or better at the whole world one) and get ready for the next wave.

[whalphen]

The conclusion seems clear: Unless we manage to fundamentally change people's behaviors, and hence slow down the spread, we are in for a collapse of the health systems.

It's interesting that I've seen very few projections on the internet and none in the press.  I think humans, in general, have a hard time appreciating the implication of exponential spread.  The more we can give people a glimpse of what's to come, the more willing they may be to change behaviors.  And there's really only two things they need to do, or even can do to slow the spread: 1) wash their hands frequently, 2) go home and stay there.

[not1xor1]

by May it the virus should be much weaker.

Snopes rates that as "unproven". https://www.snopes.com/fact-check/coronavirus-warm-weather/

I'm not Trump, but you do behave like him. I never wrote that the virus will magically disappear in May.
It is high likely that COVID-19 is here to stay, together with the other coronaviruses which cause flue-like symptoms since centuries:
- Human Coronavirus 229E (HCoV-229E)
- Human Coronavirus OC43 (HCoV-OC43)
- Human Coronavirus NL63 (HCoV-NL63)
- Human Coronavirus HKU1 (HCoV-HFU1)

By May the virus should be weaker, as shown in studies regarding the SARS virus, but I'm afraid dr Zhong Nanshan might be wrong pretending it will completely disappear (like SARS) in June.
It is a pandemy. I suspect it will just spread in the other hemisphere and then come back here next winter.

[not1xor1]

"The virus should be much weaker" and 'the virus is less likely to be transmitted in warmer weather' - the latter being what Snopes considers - are not the same thing.

The integrity of the capsid in the air is strongly related to both temperature and RH. So in the Death Valley, although the temperature is quite high it might survive a bit longer than near the beach during later spring in a temperate climate. 

There's a real possibility that the virus in circulation may be less virulent by May by a combination of two reasons: (1) Coronaviruses are highly mutable, there's a very high probability of new variants emerging during a pandemic, (2) viruses/bacteria/parasites that kill their hosts are less likely to spread than less lethal variants. This was actually seen, and confirmed, during one of the earlier flu pandemics (I don't offhand remember which one) where later in the pandemic the original flu variant died off and a mutated, less virulent, variant continued to spread - the death rate fell while infection rates continued at the same level. Although this is a real possibility, it's not a effect that any plan to control the disease ought to rely on.

You're probably referring to H1N1 AKA 2009 pig flue. Yes, when a virus is too good at killing its host it has less chances to spread around.

[whalphen]

At the current rate of spread in the US, herd immunity will slow the growth before the summer arrives.  Experts are advising that at 60%-80% infection rate, there is enough immunity to slow the spread in the population.  The current growth pattern in the US will put us at that point in late April.

[not1xor1]

The virus is more prone to a damage with higher temperature, lower humidity and more ultraviolet exposure.
Therefore we have flu epidemics in the northern hemisphere usually in winter (and also the human immune system is at lowest level in winter/spring).

the protein shell (capsid) lasts longer in COLD (i.e. lower temperature) and DRY weather. Besides that, cold weather causes vasoconstriction of peripheral veins and so the immune system is less efficient at reacting to the virus in the nose (or in the eyes if you transfer the virus from a contaminated surface).
I think that a good mask (FFP3) while only partially stops the virus (it is around 0.1µm), creates a hot and wet micro-climate where the virus shell is more prone to degradation and the nose blood circulation works better.

[not1xor1]

Recent figures from China are laughable, and they cannot be rationally explained.

Given the harsh measures the Chinese government took (difficult to implement in other countries) those figures are believable like are those of South Corea.

[iMo]

The conclusion seems clear: Unless we manage to fundamentally change people's behaviors, and hence slow down the spread, we are in for a collapse of the health systems.

It's interesting that I've seen very few projections on the internet and none in the press.  I think humans, in general, have a hard time appreciating the implication of exponential spread.  The more we can give people a glimpse of what's to come, the more willing they may be to change behaviors.  And there's really only two things they need to do, or even can do to slow the spread: 1) wash their hands frequently, 2) go home and stay there.

The major problem we see is the vast majority of people are absolutely irresponsible by default (the people anywhere in the World). They simply do not care, unless forced by pretty draconian measures (or by totalitarian dictatorships). And they will not change their behavior, I am afraid.
I think only fast advances in vaccines manufacturing may save future generations from disasters. They have to be able to produce a new vaccine within a month - a nice challenge for talented scientists and high-tech companies, imho.

[metrologist]

I was only criticizing metrologist's way of plotting the data. Didn't mean to question your point, and I don't think I said anything to that effect.

Here is the way I had the chart before I saw the log scale, so all I did was change my scale to log. It was interesting because we use the same data, but my chart did not fit a 2.4 day doubling time. I'm sure there are a dozen other things you can pick on about my chart. This one is the exact same data, just changed the scale.

Of course I expect the case curve to deviate from the extrapolated curve, and of course the data is very limited and not truly representative. I was reading that the US has a capacity of about 1000 tests per day max (lab analysis), and the test methods are unproven. They can give all the millions of tests they have, but it cannot be processed that fast.

Some fun reading that crossed my path..

Interesting panel discussion:
#coronavirus info:
[bracketed text is by panel attendee] otherwise pretty much quotes from panelists

“University of California, San Francisco BioHub Panel on COVID-19
March 10, 2020

• Panelists
• Joe DeRisi: UCSF’s top infectious disease researcher. Co-president of ChanZuckerberg BioHub (a JV involving UCSF / Berkeley / Stanford). Co-inventor of the chip used in SARS epidemic.
• Emily Crawford: COVID task force director. Focused on diagnostics
• Cristina Tato: Rapid Response Director. Immunologist.
• Patrick Ayescue: Leading outbreak response and surveillance. Epidemiologist.
• Chaz Langelier: UCSF Infectious Disease doc

What’s below are essentially direct quotes from the panelists. I bracketed the few things that are not quotes.
• Top takeaways
• At this point, we are past containment. Containment is basically futile. Our containment efforts won’t reduce the number who get infected in the US.
• Now we’re just trying to slow the spread, to help healthcare providers deal with the demand peak. In other words, the goal of containment is to "flatten the curve", to lower the peak of the surge of demand that will hit healthcare providers. And to buy time, in hopes a drug can be developed.
• How many in the community already have the virus? No one knows.
• We are moving from containment to care.
• We in the US are currently where Italy was a week ago. We see nothing to say we will be substantially different.
• 40-70% of the US population will be infected over the next 12-18 months. After that level you can start to get herd immunity. Unlike flu this is entirely novel to humans, so there is no latent immunity in the global population.
• [We used their numbers to work out a guesstimate of deaths— indicating about 1.5 million Americans may die. The panelists did not disagree with our estimate. This compares to seasonal flu’s average of 50K Americans per year. Assume 50% of US population, that’s 160M people infected. With 1% mortality rate that's 1.6M Americans die over the next 12-18 months.]
• The fatality rate is in the range of 10X flu.
• This assumes no drug is found effective and made available.
• The death rate varies hugely by age. Over age 80 the mortality rate could be 10-15%. [See chart by age Signe found online, attached at bottom.]
• Don’t know whether COVID-19 is seasonal but if is and subsides over the summer, it is likely to roar back in fall as the 1918 flu did
• I can only tell you two things definitively. Definitively it’s going to get worse before it gets better. And we'll be dealing with this for the next year at least. Our lives are going to look different for the next year.

• What should we do now? What are you doing for your family?
• Appears one can be infectious before being symptomatic. We don’t know how infectious before symptomatic, but know that highest level of virus prevalence coincides with symptoms. We currently think folks are infectious 2 days before through 14 days after onset of symptoms (T-2 to T+14 onset).
• How long does the virus last?
• On surfaces, best guess is 4-20 hours depending on surface type (maybe a few days) but still no consensus on this
• The virus is very susceptible to common anti-bacterial cleaning agents: bleach, hydrogen peroxide, alcohol-based.
• Avoid concerts, movies, crowded places.
• We have cancelled business travel.
• Do the basic hygiene, eg hand washing and avoiding touching face.
• Stockpile your critical prescription medications. Many pharma supply chains run through China. Pharma companies usually hold 2-3 months of raw materials, so may run out given the disruption in China’s manufacturing.
• Pneumonia shot might be helpful. Not preventative of COVID-19, but reduces your chance of being weakened, which makes COVID-19 more dangerous.
• Get a flu shot next fall. Not preventative of COVID-19, but reduces your chance of being weakened, which makes COVID-19 more dangerous.
• We would say “Anyone over 60 stay at home unless it’s critical”. CDC toyed with idea of saying anyone over 60 not travel on commercial airlines.
• We at UCSF are moving our “at-risk” parents back from nursing homes, etc. to their own homes. Then are not letting them out of the house. The other members of the family are washing hands the moment they come in.
• Three routes of infection
• Hand to mouth / face
• Aerosol transmission
• Fecal oral route

• What if someone is sick?
• If someone gets sick, have them stay home and socially isolate. There is very little you can do at a hospital that you couldn’t do at home. Most cases are mild. But if they are old or have lung or cardio-vascular problems, read on.
• If someone gets quite sick who is old (70+) or with lung or cardio-vascular problems, take them to the ER.
• There is no accepted treatment for COVID-19. The hospital will give supportive care (eg IV fluids, oxygen) to help you stay alive while your body fights the disease. ie to prevent sepsis.
• If someone gets sick who is high risk (eg is both old and has lung/cardio-vascular problems), you can try to get them enrolled for “compassionate use" of Remdesivir, a drug that is in clinical trial at San Francisco General and UCSF, and in China. Need to find a doc there in order to ask to enroll. Remdesivir is an anti-viral from Gilead that showed effectiveness against MERS in primates and is being tried against COVID-19. If the trials succeed it might be available for next winter as production scales up far faster for drugs than for vaccines. [More I found online.]
• Why is the fatality rate much higher for older adults?
• Your immune system declines past age 50
• Fatality rate tracks closely with “co-morbidity”, ie the presence of other conditions that compromise the patient’s hearth, especially respiratory or cardio-vascular illness. These conditions are higher in older adults.
• Risk of pneumonia is higher in older adults.

• What about testing to know if someone has COVID-19?
• Bottom line, there is not enough testing capacity to be broadly useful. Here’s why.
• Currently, there is no way to determine what a person has other than a PCR test. No other test can yet distinguish "COVID-19 from flu or from the other dozen respiratory bugs that are circulating”.
• A Polymerase Chain Reaction (PCR) test can detect COVID-19’s RNA. However they still don’t have confidence in the test’s specificity, ie they don’t know the rate of false negatives.
• The PCR test requires kits with reagents and requires clinical labs to process the kits.
• While the kits are becoming available, the lab capacity is not growing.
• The leading clinical lab firms, Quest and Labcore have capacity to process 1000 kits per day. For the nation.
• Expanding processing capacity takes “time, space, and equipment.” And certification. ie it won’t happen soon.
• UCSF and UCBerkeley have donated their research labs to process kits. But each has capacity to process only 20-40 kits per day. And are not clinically certified.
• Novel test methods are on the horizon, but not here now and won’t be at any scale to be useful for the present danger.

• How well is society preparing for the impact?
• Local hospitals are adding capacity as we speak. UCSF’s Parnassus campus has erected “triage tents” in a parking lot. They have converted a ward to “negative pressure” which is needed to contain the virus. They are considering re-opening the shuttered Mt Zion facility.
• If COVID-19 affected children then we would be seeing mass departures of families from cities. But thankfully now we know that kids are not affected.
• School closures are one the biggest societal impacts. We need to be thoughtful before we close schools, especially elementary schools because of the knock-on effects. If elementary kids are not in school then some hospital staff can’t come to work, which decreases hospital capacity at a time of surging demand for hospital services.
• Public Health systems are prepared to deal with short-term outbreaks that last for weeks, like an outbreak of meningitis. They do not have the capacity to sustain for outbreaks that last for months. Other solutions will have to be found.
• What will we do to handle behavior changes that can last for months?
• Many employees will need to make accommodations for elderly parents and those with underlying conditions and immune-suppressed.
• Kids home due to school closures
• [Dr. DeRisi had to leave the meeting for a call with the governor’s office. When he returned we asked what the call covered.] The epidemiological models the state is using to track and trigger action. The state is planning at what point they will take certain actions. ie what will trigger an order to cease any gatherings of over 1000 people.

• Where do you find reliable news?
• The John Hopkins Center for Health Security site. Which posts daily updates. The site says you can sign up to receive a daily newsletter on COVID-19 by email. [I tried and the page times out due to high demand. After three more tries I was successful in registering for the newsletter.]
• The New York Times is good on scientific accuracy.

• Observations on China
• Unlike during SARS, China’s scientists are publishing openly and accurately on COVID-19.
• While China’s early reports on incidence were clearly low, that seems to trace to their data management systems being overwhelmed, not to any bad intent.
• Wuhan has 4.3 beds per thousand while US has 2.8 beds per thousand. Wuhan built 2 additional hospitals in 2 weeks. Even so, most patients were sent to gymnasiums to sleep on cots.
• Early on no one had info on COVID-19. So China reacted in a way unique modern history, except in wartime.

• Every few years there seems another: SARS, Ebola, MERS, H1N1, COVID-19. Growing strains of antibiotic resistant bacteria. Are we in the twilight of a century of medicine’s great triumph over infectious disease?
• "We’ve been in a back and forth battle against viruses for a million years."
• But it would sure help if every country would shut down their wet markets.
• As with many things, the worst impact of COVID-19 will likely be in the countries with the least resources, eg Africa. See article on Wired magazine on sequencing of virus from Cambodia.”

[whalphen]

I think it would be interesting to try to project when we would run out of hospital beds based on this growth pattern.  A number of factors would come into play, such as number of beds, % in use, turnover of beds, location of beds vs. hotspots, etc.  Just eyeballing the numbers without careful calculations and by making broad assumptions, it looks to me like the US health system could be overwhelmed early next month.

[paulca]

Total cases is a bit pointless.  80% of the total cases have recovered.  5% or so are dead, leaving only 15% actually infected and infectous.

The figure you want to look at is "Active Cases".  These are the number of people actually with the virus.  It's important as if that number is rising, it's getting worse, if it's falling, it's getting better.

Total cases will ALWAYS rise, this is why the media are using it to scare people into the zombie apocalypse.

Just been for my weekly shopping.  No loo roll, because of mass insanity.  No pasta, rice, bread and the tinned food and frozen isles completely cleared out!

[iMo]

The number of infected is XX times higher than those reported by WHO, imho.
Those undocumented are the people with none symptoms, with symptoms similar to light flu (especially when vaccinated against flu), and people who hide the symptoms.
Thus the 40-70% infected the US or Germany indicate as the possible result could be easily achievable.
What is reported are cases which were tested. And they test only a really small sample of population (those who show symptoms, or those who are suspect). Most states in EU do not test people who stay home with mild symptoms "similar to flu".
After this pandemic they will find antibodies for the covid19 in 70% of the World population, I bet.

[Simon]

Yes the bigger problem is the idiots panic buying. I have several weeks of food already. I will buy no more.

[ebastler]

Total cases is a bit pointless.  80% of the total cases have recovered.  5% or so are dead, leaving only 15% actually infected and infectous.

The figure you want to look at is "Active Cases".  These are the number of people actually with the virus.  It's important as if that number is rising, it's getting worse, if it's falling, it's getting better.

In the exponential growth phase, with the number of total cases doubling every 2 to 3 days as seen in most affected countries these days, the vast majority of the total cases will be "active". At any time, 80% of the total cases will be cases only detected within the last week.

Exponential growth is a bitch, and can be counter-intuitive. 

[VK3DRB]

Anecdotally, the virus appears to be rampant in France despite the statistics. I just found out a relative in France has got the corona virus. Also, my daughter lives in France (she does not have the virus) and said some client walked into her work and told the judge she has the virus and will spread it on purpose because she has contempt for the French court system. The woman was arrested and tested as positive. Seems like the virus can be used as a weapon.

By the way, is it "coronavirus" or "corona virus"? If seems both terms are used, with the former being more popular. It seems the bad spelling is spreading faster than the virus itself. Why would Anglophones join the two distinct words together when there is no reason to do so? Itseemsratheroddtome.