Mischievous Ramblings II

Epidemic Math

Now 150 dead

14 days at least 1000

I suggest you read this with a drink in hand

[This is an email I sent a couple of friends who requested my take on things]

Lots of links down below.

 

If you’re going to read one, read this one (it’s also down below):

https://medium.com/@tomaspueyo/coronavirus-act-today-or-people-will-die-f4d3d9cd99ca

 

If two, add this one:

https://www.axios.com/coronavirus-report-us-uk-strategies-e45bc5d4-d2f1-40e2-825e-429b2b7c1b50.html

 

 

Here’s the short version:

COVID-19 is more infectious and more deadly than a typical (or even a serious) seasonal flu.  Estimates of total infections in the US range from 65 to 230 million without attempts at mitigation, with accompanying death tolls of 600K to 9 million.

 

The big worry is overwhelming our hospital capacity so that serious cases cannot be supported until recovery.

 

Social distancing, infection avoidance (hand washing, sanitizing, avoiding touching the face, etc.) are our only current weapons to mitigate spread of the disease.

 

Forgot to add: when you look at numbers of deaths, those infections happened at least 3 weeks ago (possibly 5).  So we are looking into the past. Depending upon doubling rate, the current expected deaths from infections happening down will be between 8-64 times larger if the doubling rate is 6 days and 16-1000 times larger if the doubling rate is 3 days.

Here’s the long version:

 

The Disease:

While numbers continue to come in, there appears to be no question that SARS-CoV-2 (the virus which causes the disease COVID-19) is both more infectious and more deadly than a typical seasonal flu.  Depending upon the research the doubling rate of COVID-19 appears to be somewhere in the 3-6 day range, and the death rate looks to be a minimum of about 1% (with full treatment options available) and a maximum of maybe 4% (if the healthcare system is overloaded and treatment options become more limited).

 

Incubation period can be up to 14 days and a full course of disease is 20-30 days.

 

The Spread:

Up to 80% of cases are thought to be mild, many of them completely asymptomatic.  There is no symptom based diagnostic that can differentiate between COVID-19 and a bad cold or the flu.  Up to 15% are “serious” requiring at least medical monitoring if not hospitalization, and around 5% are “critical” requiring significant medical intervention (at least oxygen, possibly ventilation) meaning the use of ICU or CCU beds.

 

Some research indicates that people infected with SARS-CoV-2 shed virus days before they become symptomatic (if they ever do) and for some time after they “recover” if their cases are mild.

 

Transmission is not yet known to be aerosolized, but appears primarily droplet-based.  Basically, people with the disease shed virus through natural secretions including mucous and feces.  A sneeze projects infected droplets for at least several feet, and those droplets stay on surfaces for at least hours and possibly days.  When a sick person touches their face they are likely to get traces of mucous on their hands which they then deposit on surfaces they touch, including tables, door handles, keyboards and screens, and so on.

 

Estimates of infection rates in the general populace over the course of the outbreak range from 20%-70%.  The time frame is not usually specified but we are most likely talking about under a year, depending upon mitigation strategies.

 

In the US, that means between 65 million and 230 million people will be infected.  65 million is roughly the number of infections in the US during a bad seasonal flu year.

 

At 1% mortality, that’s 650K - 2.3 million deaths in the US.  At 4%, it’s 2.5 million to 9 million.

 

Treatments and Vaccines:

A vaccine will take at least a year to get through safety testing and trials.  Initial testing has already begun on at least one vaccine but first attempts don’t always work.  A safe assumption is that a vaccine is at least a year and a half out, possibly two years.

 

For now, treatment is strictly supportive, meaning that serious and critical cases get put to bed and watched.  As their breathing is reduced, first oxygen and later mechanical ventilation are used to keep the patient alive while the disease runs its course.  In some cases, secondary infections will be treated with antibiotics.

 

Mitigation:

Without a vaccine, the only mitigation available is limiting the spread of the disease.  Epidemiologists use R0 (“R-nought”) as a measure of replication of a disease in a population.  If R0 is 1, each infected person gives the disease to one other individual (on average) and the number of cases in the population is stable over the long term.  If R0 is below 1, the number of cases falls over time.  If R0 is above 1, the number of cases increases over time.

 

The only weapon we have against this disease spreading is reducing R0.  The easiest way to do that is to reduce the number of contacts between infected people and uninfected people.  Unfortunately, infected people don’t have a giant sign on them saying “COVID-19!  Stay Away”.  Since many cases of COVID-19 are asymptomatic, and because patients shed virus before being symptomatic, we cannot isolate patients nor can we selectively avoid them.

 

The solution is to reduct contacts with _everyone_.  “Social distancing” means drastically reducing the number of direct in-person interactions you have with people.  Staying home.  Staying at least six feet away from people when you do go out.

 

Alongside reducing contacts goes reducing the chance of infection.  Since SARS-CoV-2 must enter the body to infect you and you generally pick it up on your hands, hand-washing often (for at least 20 seconds) is a useful mechanism to reduce infection rates.  Where hand washing is not possible, hand sanitizer (at least 60% alcohol) is better than nothing.  Avoiding touching your face with possibly infected hands reduces infections through the mouth, nose, and eyes.

 

Effects of Mitigation:

Modeling shows (and the experience of Hubei Province in China confirms) that radical contact reduction (“lockdown”) is effective in slowing or even stopping the spread of the virus.

 

But the level of contact reduction has to be big.  At least 75% on an individual basis.  That’s why all the recommendations to stay at home (“shelter in place”), close business, avoid crowds, etc.

 

 

 

Sources

Here’s where I start every day.  During the early stages, these various surveillance sites all agreed on the numbers.  Now they don’t, but they generally end up syncing up at least once or twice a day as they troll for numbers and update:

 

https://ncov2019.live/data

https://www.arcgis.com/apps/opsdashboard/index.html#/bda7594740fd40299423467b48e9ecf6

 

This is automated graphing of various statistics about the outbreak:

 

https://www.worldometers.info/coronavirus/

https://www.worldometers.info/coronavirus/#countries has a sortable table which lists information by country including new cases and new deaths in the last day

 

Here’s an article about the modeling which apparently changed the White House direction from “meh” to “OMG”: https://www.axios.com/coronavirus-report-us-uk-strategies-e45bc5d4-d2f1-40e2-825e-429b2b7c1b50.html

 

And here’s the report itself: https://www.imperial.ac.uk/media/imperial-college/medicine/sph/ide/gida-fellowships/Imperial-College-COVID19-NPI-modelling-16-03-2020.pdf

 

Editorial on how this economic hit is different and what we should do about it: https://www.marketwatch.com/story/theres-a-second-curve-that-needs-flattening-the-recession-curve-2020-03-17

 

Like any big story, Twitter is the place for the latest, if you can sort through the crap.  I follow a guy by the monicker epsilon: https://mobile.twitter.com/epsilon3141

 

Analysis of hospital bed capacity on a state-by-state basis and when COVID-19 cases will fill the system: https://www.healthleadersmedia.com/clinical-care/see-when-states-will-face-hospital-bed-capacity-shortages-during-covid-19-outbreak

 

Outbreak math: https://medium.com/@tomaspueyo/coronavirus-act-today-or-people-will-die-f4d3d9cd99ca

https://finance.yahoo.com/news/marty-makary-on-coronavirus-in-the-us-183558545.html

Pulled from a comment thread....

Pulled from a comment thread. This is in response to anyone saying "it's just a cold", "it's not a big deal", "compare it to the flu", "2% isn't so bad"...

This is the scary estimate:

ALL OF THE BELOW IS SPECULATIVE

If it spreads like the seasonal flu and has a 2% death CFR, we can expect hundreds of thousands of deaths in the US (high six figures), which would make it the largest single cause of death in the country.

If it spreads to 50% of the population (I've seen estimates 20%-70%) and has a 2% CFR we are talking 3 million deaths. That would be 50% more deaths per million than the flu of 1918.

End of the world? No. But the worst mortality event in US history.

Oh, and that 2% CFR? That presumes treatment of some 20% of cases which are "serious". There are fewer 1 million hospital beds in the US. Most of them are not ICU or CCU beds and do not have ventilators. If there are a million serious cases (which implies only 5 million infections based upon what we know now compared to the 40 million seasonal flu infections annually), the healthcare system will break down, and I'd bet that the notional 2% CFR rises under those conditions.

Not a certainty. Possibly a likelihood.

Choosing an Energy Source

(Visualization: Evan Robinson; Data Sources: Ecologist Magazine: The Future of Energy is Renewable,
Energy & the US Economy: A Biophysical Perspective, Renewables in Scotland: Carbon Dioxide)

I was reading The Transition Handbook in preparation for a workshop next weekend, and there's a couple of great charts showing the carbon intensity (grams of CO2 equivalent produced per kilowatt-hour) and EROEI (Energy Returned on Energy Invested) of various energy sources (although oil is prominent by its absence, I added some in later).  Those charts are online at The Ecologist magazine.  Although I found them interesting and useful, I thought they would be much more useful if they were combined.

The result is above: a scatter chart showing the relationship between EROEI and Carbon Intensity for various forms of energy.  Because a high EROEI is good, we would prefer to use energy sources far to the right.  Because low Carbon Intensity is good, we would prefer to use energy sources far down on the chart.

Several energy sources were "outliers" -- they didn't fit nicely on the chart as scaled, so they are marked by half cut-off blue dots at the very edge of the plot.  Their label includes the actual value, so although the "Tar Sands (465)" dot is at about 260 gCO2e/kwh, the actual value is 465 gCO2e/kwh as indicated by the (465).

For any given application, this chart provides a (roughly) optimal ordering of energy sources.  Wherever possible, Tidal Range power is the highest EROEI and the lowest (or very very close to the lowest) Carbon Intensity.  Next is a choice between Offshore Wind (if you prefer to reduce Carbon Intensity a bit more in exchange for a bit less EROEI) or Sewage / Landfill Gas.  And so on.

You'll notice that the included petroleum and petroleum like fuels (US oil, tar sands, Coal) are very dirty, except for possibly "Clean" Coal, which doesn't really exist yet.  And there are alternatives to each of them which have higher EROEIs.  But none of them have the physical energy density that these carbon based fuels like oil have.

A simple conclusion is obvious: carbon based fuels should not be used for electrical generation (that's what most of these energy sources are, with the exception of Solar Thermal and the generalized cogeneration of electrical power and heat intrinsic to Sewage / Landfill Gas, Concentrating Solar Power, Solid Biomass, and so on) if any of these other alternatives are available and can supply a significant fraction of the required power.

GIANTmicrobes!

GIANTmicrobes! is a site that sells plush toy versions of some of the most "popular" microbes: Cold (Rhinovirus), Flu (Orthomyoxvirus), Stomach Ache (Shigella), Sore Throat (Streptococcus), Earache (Streptococcus pneumonia), Cough (Bordetella pertussis), and others, including Plague and Ebola.

1918 killer flu ‘came from birds’

1918 killer flu 'came from birds'

The possibility of an H5N1 flu pandemic is getting a lot more attention and press (especially since avian flu hit Europe). It remains a potentially terrifying possibility, but at least some conversation is happening.

Odds are good that whatever happens will not be as bad as the worst we can imagine, but that it will be worst than the best we hope for.

Archimedes Death Ray

2.009 Product Engineering Processes: Archimedes Death Ray is a fantastic exploration of the "using mirrors to set ships on fire" myth/legend/experiment.

The MythBusters "busted" this myth, but this class at MIT seems to have made it at least "plausible".

Kung Fu Monkey: Bee-Fu!

Bee-Fu!:

The image of a dolphin rasing his t-shirt to reveal he's strapped, glaring at a barracuda and snarling "You want some of this you Sphyraenidae Sphyraenus fuck?!"

Too late for Superman

WorldNetDaily: Adult stem cells restore feeling in paraplegic

In an apparent major breakthrough, scientists in Korea report using umbilical cord blood stem cells to restore feeling and mobility to a spinal-cord injury patient.

The research, published in the peer-reviewed journal Cythotherapy, centered on a woman had been a paraplegic 19 years due to an accident.

A Person’s Got To Know Their Limitations

essays & effluvia: The Physics of 4 Wheel Drive and Snow