The Imperial College Study: Part 3B

[Links to the full series]

Part 1
Part 2
Part 3A
Part 3B

------------

 

1. How were the CFR and IFR calculated for the Imperial College study?

Since the Imperial College study is a microsimulation, it was simulating a whole population and hence needed to use an IFR, not a CFR. It got an IFR from this study: https://www.medrxiv.org/conte…/10.1101/2020.03.09.20033357v1

Let's look at how this study came up with their numbers:

The report estimated the CFR and IFR based on three different datasets and using multiple techniques in order to validate the results. They looked at data from mainland China (70,000+ cases as of February 11th, then cross checked with latest results as of March 3rd), data from the Diamond Princess cases, and data from cases being tracked outside China (about 2000 cases as of February 25th).

My description of their techniques is going to be very much an oversimplification because I find it hard to describe the statistical techniques employed in a short space. Here's my understanding of some of the key features of the technique used for the mainland China data:

1. They broke the population into 10 year age bands and assumed that covid-19 would attack each age band equally.
2. They took the actual age demographics of the infected areas and projected how many people should get sick in each age band.
3. The looked at how many people were diagnosed as sick in each age band. For the younger age bands, this was fewer than projected given an equal attack rate. This gave them an age-band-specific underreporting amount.
4. Most of the fatalities were in Wuhan, but Wuhan had a much higher fatality per reported case than mainland China. They assumed this was due to hospital overcrowding causing milder cases to get turned away, so they added in a further factor representing hospital overloading to scale down the Wuhan numbers to be in line with the rest of the Chinese numbers.
5. For each age-band, they then identified which CFR that--given the onset-to-death times which were observed--would have produced the observed total cumulative deaths as of the most recent data, given the underreporting factors that they identified.
6. They then aggregated these age-band specific CFRs into a population-wide CFR, which turned out to be 1.38%. Note, though, that this number is specific to the Chinese age demographics.

To estimate an IFR from this CFR, they used data from people repatriated out of China back to their homelands. All of these people were tested, and it was discovered that there were about as many asymptomatic people who tested positive as there were symptomatic people who tested positive. This led to the final IFR for the Chinese outbreak being estimated at 0.66%. I should note here, though, that the data sample size here was particularly small: a total of 6 asymptomatic people who tested positive from those flights.

To validate their IFR using the Diamond Princess data, they took a timeline of onset-of-symptoms for the 705 diagnosed passengers on the cruise. Then applying their age-specific onset-to-death results on the actual ages of the diagnosed passengers, they projected that by March 5th, between 3 and 14 people should have died if their IFR was correct. Since 7 passengers had died by that date, the Diamond Princess case was judged to be consistent with their results.

To separately estimate a CFR from all of cases outside China, they used two different methods, neither of which I have looked into enough to understand. At the time this study was done, this was a pretty small sample size (1334 cases out of 2000+ met their inclusion standards). Also, they didn't have individual-level onset-of-symptom or recovery data for a lot of those cases. For these two reasons, the CFR estimates cover a wide range, from 0.4% to 7.2%, with 1.2% being the best fit to the data. This basically validated the reasonableness of the 1.38% result from mainland China.

2. That's a lot of information. What's the bottom line for the Imperial College study again?

What the Imperial College study took from all of the above is that the Covid-19 IFR is about 0.66% for Chinese demographics. They also took the age-specific IFR from the study and applied it to the older Great Britain demographics to get an IFR that they used for their simulations of 0.9%. They also used the onset-to-death time periods and the percentage of hospitalizations from that study (which I didn't get into here but was another thing calculated from the same data).

The Imperial College Study: Part 3A

[Links to the full series]

Part 1
Part 2
Part 3A
Part 3B

------------

Now I'm going to look at what fatality rate the Imperial College used, and how it was derived. I'll be splitting this part up into two sections. First (3A), I'm going to look at what a CFR is in general and how it's often calculated. Second (3B), I'll look at specifically how it was determined in the Imperial College study.

1. What is a CFR?

CFR stands for "Case Fatality Rate". It is the proportion of those people who are diagnosed with a certain disease who die over some period of time. This distinguishes it from "Mortality", which is the proportion of a total population who die of a particular disease--so MERS, for example (another zoonotic coronavirus), has a massive CFR because it is almost always fatal if you catch it, but a tiny Mortality because very few people do.

CFR is sometimes a confusing number, however, because of an ambiguity in what "Case" stands for? Is a "case" only someone who has been officially diagnosed with the disease by a doctor? Or is it anybody who can be proved to have had the disease? CFR is actually used in both ways at different times. So, for example, you may have heard it said that the CFR for the seasonal flu is about 0.1%. This is using CFR in the second sense. What epidemiologists do is try to get a good estimate of the total number of people who have actually had the flu in a given season, whether or not they were diagnosed with the flu. Then they try to get a good estimate of the number of people who died because of the flu, again, whether or not they were diagnosed with it. Then they divide the second number by the first number.

This use of "CFR" can only be fully accurate after the fact, when an epidemic or an outbreak has passed. So people *also* use "CFR" to refer to the number of people who are *diagnosed* with a disease who then die. This is useful at the time of an outbreak because it gives a sense of the seriousness of the disease right away. But this usage is recognized to be always an overstatement of the eventual fatality of the disease, because it does not account for those people who come down with the disease but never seek medical treatment, and these people will always be heavily skewed towards the milder forms of the illness.

In order to be more clear about the difference in these two numbers (the percentage of diagnosed cases who die and the percentage of all infected people who die), I would like to use a term that I've seen in some papers, which is IFR, or "Infection Fatality Rate". This would be the percentage of all infected people who die, whereas CFR would be the percentage of those people who are diagnosed with the disease who die.

2. How is the CFR calculated in general?

*After* an outbreak has entirely resolved, CFR is very easy to calculate: just divide the number of diagnosed individuals who died by the total number of diagnoses. During an outbreak, however, these numbers are a moving target and it can be deceptive to try to calculate a CFR from those numbers. Here are some reasons why:

• Early on in an outbreak, a lot more people are going to be in an early phase of the disease, not a late phase. So a lot of *these* people might *eventually* die, but not be to that stage yet. So if you have a lot of testing and record a lot of sick people at an early stage, you might *understate* your CFR.
• On the other hand, if you *don't* do a lot of testing, the first wave of people in an outbreak you'll find out about are those people who show up at a hospital very sick. These are going to be disproportionately the percentage of the population who got the worst form of the disease or who got hit particularly hard by it. You can expect these people to die at a higher rate than the general population. So if you don't have good testing of your whole population, your initial CFR rates will be *overstated*.
• Also, the people who die earlier might be the people who are more likely to die anyway. For example, most of the people who die from covid-19 are probably older people *and* they are more likely to die earlier than the younger fatalities. If you try to calculate a CFR from initial death rates without correcting for an age factor, your initial CFR rates will be *overstated*.

In order to best calculate the true CFR from data as it comes in during an outbreak, you need to take both timeline and vulnerability data into account. For timeline correction, you need to find out what is the average time from onset-of-symptoms till death, and then plot both your death rate and infections chronologically, offset by that timelag. Given a sufficient length of time, the ratio between the two will converge on your true CFR. For vulnerability correction, you should identify the different groups that have different risk levels and split up your CFR calculation to do a separate one for each of those groups.

A key point I'd like to re-emphasize: the calculation of a CFR for a particular outbreak always gets more accurate as time goes along. This means you should pay attention to CFR as generated by those locations where outbreaks occurred *first*, when you are tracking a pandemic that is moving across the globe.

2. How is the IFR calculated in general?

In order to calculate the IFR, you need to be able to identify about how many people in your population get the disease without coming in to the doctor. You will then multiply your CFR by this fraction to get the lower IFR.
In practice, this is done in a couple of ways. The "gold standard" here is to do a serological study: you select random people from your population and look for the tell-tale antibodies in their blood that show that that person had a specific disease and developed an immunity to it. You can then get a good number for percentage of your total population that had that disease recently, whether or not they were caught in the official tallies.

Another way to establish this is by random surveys--you just ask people whether they had a particular disease or not and whether they went to the doctor for it. This is less accurate than the first method because the people you are trying to count are necessarily self-diagnosers, and the accuracy of their self-diagnosis is likely to be flawed.

The CDC has used both approaches in the past and come up with a roughly 50% proportion for the seasonal flu. That is, in order to calculate how many people in total have the flu each year, they multiply the cases determined from hospital records by about 2. Note, however, that this number is specific to the seasonal flu. In general, the more severe a disease is, the more people who have that disease will go to the doctor and the lower the proportion of that "undiagnosed" population will be.

The Imperial College Study: Part 2

[Links to the full series]

Part 1
Part 2
Part 3A
Part 3B

------------

Now I will start looking at the parameters that the Imperial College study used to characterize the coronavirus in its simulations. I'll start by looking at the basic reproduction number: the now-famous r(0) ("r naught").

1. What is r(0)?

R(0) is defined as the number of secondary infections a single instance of an infection will cause if exposed to a completely vulnerable population. It arises from a combination of multiple elements:

•   how naturally infective the particular virus is
•   how long an infected person is contagious *while* being still present in a population
•   how many people an infected person can be expected to contact during the infective period

Importantly, while the first two elements are virus dependent, the third is societally dependent. The more mobile and mixing a population is, the higher the r(0) will be. "Social distancing" is therefore a way of changing the r(0) of a particular disease by altering that third element.

From the perspective of a microsimulation, however, epidemiologists can just take the r(0) as a constant which determines the chance (on average) that any given non-infected individual coming into contact with an infected individual will become infected. This is because the microsimulation intrinsically accounts for degrees of social mixing by its rules that govern individual behaviors.

2. How is r(0) calculated in general?

This is an oversimplification, but I am aware of two methods of calculating an r(0).

First, an r(0) is said to be proportional to the early doubling rate of infections during an epidemic. During this early time period, there are so many more non-infected people compared to infected people that each instance of an infection will have essentially a "clean slate" of people to infect. Later on, an increasing percentage of infected individuals will come into contact with already-infected individuals, thus slowing the growth rate. Furthermore, in the earliest phases of an epidemic, the disease will be spreading without a lot of symptoms and without causing major alarm, so that social mixing will be normal as a whole. Later on, fear of the disease spreading through society will itself change mixing behavior and therefore the r(0) itself.

Second, you can examine a timeline of infection and death rates after the fact from a given epidemic event, and then run microsimulations for that population using different r(0) values until you reproduce the observed curve in the simulation.

3. How was the r(0) used in the Imperial College study calculated in particular?

The Imperial College simulation used the results from two different studies, each taking one of these approaches.

The first study was done by the China CDC (available here: https://www.nejm.org/doi/full/10.1056/NEJMoa2001316) looked at the first 425 cases reported in Wuhan and reconstructed an infection timeline by interviews which established when symptoms first occurred in each case. From this they saw an initial doubling time of the disease of about 7.5 days, which calculated out to an r(0) of 2.2, with a 95% confidence range of 1.4 to 3.9.

The second study was done by Julien Riou and Christian Althaus and funded by the Swiss National Science Foundation (available here: https://www.eurosurveillance.org/…/1560-7917.ES.2020.25.4.2…). They ran 2 million separate epidemic simulations with varying parameters and looked for those parameters that could reproduce the timeline of the infections as of January 29th, which at the time included 5,997 confirmed cases in China and 68 confirmed cases exported to other countries. Importantly, they allowed the number of cases *actually* in China in their simulations to vary substantially, in order to account for potential massive misreporting of the Chinese data. The results of this study was an r(0) of 2.2, with a 95% confidence range of 1.4 to 3.8.

4. How confident can we be in the results of these two studies?

Both studies that establish an r(0) of 2.2 for Covid-19 acknowledge the fact that they are operating on limited data, collected in a crisis situation. They both, therefore, should be treated as preliminary best estimates given the data that we had in mid February. Nevertheless, the fact that they agree very closely after taking very different approaches to the estimation does count for something.

Furthermore, I think there is good reason to believe that the true r(0) isn't lower than 2.2 given the newer European data that we have. If you look at infection growth rates for all of the European countries and for the United States since community spread started occurring in late February / early March, you can see a pretty consistent slope of the lines, yielding a doubling time of 3-4 days. This is significantly *faster* than the doubling time that was estimated for the Chinese outbreak. I believe that this makes an r(0) of significantly less than 2.2 very difficult to justify. I think the Imperial College agreed, because they used a base doubling time of 6.5 days (somewhere between current European numbers and the Chinese numbers) and a consequent r(0) value of 2.4. They did also explore values with a range from 2.0 to 2.6 to account for uncertainty.

So I think the calculations that the Imperial College did are a very reasonable interpretation of the best data that we have. I think newer studies taking the rapidly evolving new reports of infections given much greater worldwide testing would be good to do, but I very much doubt at the moment these studies would come to different conclusions right now. If anything, they would probably raise the estimated r(0), not lower it.

The Imperial College Study: Part 1

[Links to the full series]

Part 1
Part 2
Part 3A
Part 3B

------------

A bit ago I posted a link to this study by the Imperial College of London, which is a projection of how the U.S. and Great Britain can expect the Covid-19 disease to progress in the next two years under various scenarios of different levels of social isolation: https://www.imperial.ac.uk/…/Imperial-College-COVID19-NPI-m…
I want now to do a series of posts explaining this study and the current research behind it, because I think this might be the most influential study currently informing government decisions. I encourage you to read the study yourself, but I won't assume this.

1. What is the nature of this study?

This study is the results of a series of computer simulations run by epidemiologists. Specifically, it is an application of pre-existing "microsimulation" (https://en.wikipedia.org/wiki/Microsimulation) software with parameters set to simulate Covid-19 on a virtual population. A "microsimulation" is a computer model in which the behavior of a population is modeled by creating millions of individual virtual people and having them move around in a virtual world according to a set of behavioral rules. If you've ever played any of the Sim City games or any of the Tycoon games, you've played with a simple, small-scale microsimulation. A lot of people at this point have seen an *extremely* crude microsimulation recently published in an article by the Washington Post: https://www.washingtonpost.com/…/20…/world/corona-simulator/.

In epidemiology, these things are incredibly useful because you can simulate the spread and effects of a disease much more accurately than by just talking in broad percentages. You setup your population according to the actual characteristics of the real world: children will congregate in schools during school days but stay home on the weekends, old people will live in nursing home clusters, working-age people will congregate in business and will travel a range of distances to work using public transportation according to actual percentages from real-life that you enter, and so forth. Your virtual population will be created with a range of ages and existing health conditions that you also enter, based on real-world numbers that are correct for the population and time period you are modeling.

Then you characterize your disease: how long is the incubation period? What is the range of severity of symptoms and does this vary on an age basis or by existence of co-morbidities? What is the range of infectiousness of the disease, and how does it alter over the timeline of the progression of symptoms? How close does one person need to be to another person to spread the disease? How much does having the disease itself limit the motion of an individual and therefore the likelihood that one person will continue contacting other people?

Then you introduce the disease into the virtual population and play the simulation forward a certain number of times with different random seeding, and it gives you a composite average result of what happens.

2. What are some advantages to this approach?

There are a number of great advantages to this approach in epidemiology. A lot of statements about how diseases spread are actually just abstractions of just this sort of real-life behavior. We say that diseases spread exponentially in the early phases of an epidemic, but this is just a rough mathematical approximation of the behavior of a self-replicating virus as it spreads through networks of people. A good microsimulation will take more exact accounting of real-world social distances and movement. There will be a real-world mix of dense urban centers and more spread-out suburban neighborhoods, there will be an age and health spread in the population you are simulating that matches the real world, sick people will slow down their movements, etc.

There have been some people who have doubted that Covid-19 will behave as badly in the U.S. as it has in China and in Italy. They've raised potential differences between us and them that could make a difference: different baseline health of the population, different population density, different levels of cultural contact, different standards of hygiene, different access to health care. If the microsimulation is a good match to a given population, none of these problems would apply to its projections.

Furthermore, the type of human and virus behavior that needs to be modeled in order to get a realistic simulation is not actually that complex. Basically, the only human behavior that needs to be about right is movement and proximity. Viruses also are pretty simple organisms and it doesn't take too much data to be able to get a pretty accurate knowledge on the average behavior of an infection in humans on aggregate. So I think the results of these simulations tend to be fairly robust.

Another advantage is that you can use the same model to test multiple scenarios. Not sure the exact R(0) of your virus? You can run your simulation with a range and see how it affects the outcome.

3. What are some disadvantages to this approach?

Not a lot, but I can think of a few. First, as with any computer simulation, the results are only as good as the underlying assumptions and the underlying programming. This can cause it to miss some things. For example, we are now in an absolutely unprecedented level of public awareness and discussion about the Covid-19 pandemic. This itself is likely to change societal behavior--has the model taken this sort of public awareness and consequent fear of spreading infection into account? I don't know. Also, if the underlying characteristics of the virus are not correctly understood, then you will have an example of "garbage in, garbage out". Yes, you can run the simulation on a range of inputs, but if you understanding of the parameters is very out of line from reality, the results won't be very helpful.

Second, the results these types of simulation provide are very specific and realistic. But "realistic" is not the same thing as "real" and I think these simulations tend to produce a bit of over-confidence in their results because it looks like you're observing reality when you are just roughly simulating it.

4. How good a fit to reality is the model that was used in this study?

That, I am not sure of. However, I think there is good reason to trust it's a fairly good fit. With epidemiology, we have an opportunity to run such models and test their output every year because we use them things to predict the movements of seasonal flu. We've used them to model known outbreaks from the past to see if the results they produce match historical known results.

I'll stop my first post here. In a subsequent post, I'll get into the assumptions which this particular study used: how it characterized the virus and how it justified those characterizations.

Archived Facebook Coronavirus Posts: March 18th

[I'm transitioning a bunch of Facebook posts I made on the Coronavirus to my blog.  This was from March 18th]

A lot of people are now asking, what's the end game here? How long is extreme social distancing going to need to be applied?

I think there are two answers here. One would be a *final* answer, and I don't think anyone knows the this yet because we're not sure about how effective our chosen isolation tactics will be yet. But I don't think this means we have to decide, right away, on extreme *and* long term isolation. I believe there are some very concrete goals we can aim for during an *initial* extreme social distancing period. This would give us, not an end game per se, but an end game for the first period of this event, after which we can re-evaluate and adjust. So the following are my suggested goals for the first month of our semi-quarantined life in Europe and the United States.

1. More data. The most up-to-date data and the best analyses of this data has been painting a very grim picture of the upcoming year. However, many people aren't satisfied with this data and disagree with or outright disbelieve the interpretations of this data. But even with rather extreme social isolation for a month, we're going to continue to get a lot more data on the behavior of this virus around the world. I think a month of social isolation will provide the time for a lot more data to roll in. So we can believe the experts for a month and act as if the very dire predictions are correct, but then call for a re-examination of the data at the end of this period.

2. More testing. So far, the best model for reducing spread of covid-19 with the least amount of social and economic impact has been South Korea, and their strategy centers around extremely aggressive testing. The President should use whatever emergency powers are necessary to cut through whatever bureaucratic red tape still exists around getting massive amounts of tests available to the public. Further, we ought to do whatever we can to develop a good *antibody* test, as opposed to the current antigen test. When we are able to support easy testing for whomever wants it, up to a reasonable percentage of the entire population, then we will better equipped to deal with hunting and destroying this infection without needing as extreme an amount of social isolation.

3. A social media app for tracing infections developed by a trusted source. Part of South Korea's strategy is aggressive *tracking* of people, so that when one develops an infection, they can go ahead an warn a bunch of other people who have been in the same area to get tested. This sort of big-brother overseeing isn't likely to fly well in the States, so I think we need an op-in, anonymized version for the liberty lovers of the West. If not an app, then instead we need some other policies to enable us to track and contain the virus in the wild.

4. A lot more ICU resources. We should engage in a WWII, "build an airforce from scratch" style emergency manufacturing campaign to dramatically increase our ICU capacity: buildings (modular structures, temporarily commandeered hotels, etc.), PPE, ventillators, etc. Similarly, we should recruit heavily from the medical field and train up massive amounts of new people in the specific techniques required to care for ARDS while in an infectious environment.

The primary justification for extreme social distancing is that we don't have the ICU capacity to treat all the sick people we expect in a short period of time if this disease acts as expected. So, let's increase that capacity. We need to do that by a rather insane number (maybe 50 times existing capacity?)--but that sounds cheaper to me than shutting down the entire economy for 18 months while we wait for a vaccine. So let's get on this!

5. Research progress on non-vaccine medicines. It will take 12-18 months to develop a vaccine, we're told. However, a standard shortcut to finding working medicines without such a lengthy set of trial periods is to search for existing medications that have already been FDA approved and to see if they have some effectiveness against the disease you are fighting. We already have a number of interesting candidates in various stages of testing, the most interesting one to me being the combination of zinc and chloraquine. There are also some antivirals from the HIV world that might show up.

I don't know if it's reasonable to hope for results on this front in just a month, but whatever we can do to accelerate this process would be great. It's possible we could find a set of medications that have a real impact on the disease severity in just a month.

So those are my ideas. Anyone else have something they'd like to add?

Archived Facebook Coronavirus Posts: March 17th

[I'm transitioning a bunch of Facebook posts I made on the Coronavirus to my blog.  This was from March 17th]

As you may have heard, one potential vaccine for the coronavirus has already begun early human trials. Nevertheless, it is still thought that widespread availability of a vaccine is still 12-18 months away at minimum.

That may *seem* like a long time, but as someone who spent the last three years thinking my son had a very rare disease and following very intently the development of new potential treatments for that disease, I can assure you that this is *lightning* fast for a new treatment to get FDA approval.

So the question should be asked, why is it so slow to move from the scientific development of a potential cure or vaccine to the actual rollout to the public? Some of this has to do with proving efficacy of the treatment outside of the lab, and some of it has to do with proving the safety of the treatment. But a lot of it also has to do with medical research ethical rules that have been setup to avoid past abuses wherein humans were used as "guinea pigs" in order to drive scientific understanding.

Some of these rules are good and proper in ordinary situations, but are nevertheless not *intrinsically* necessary to medical ethics. That is, there are maybe other ways we could safeguard respect for human life that don't follow these rules strictly.

Given the *massive* economic and societal costs entailed in the prolonged quarantining of huge sectors of society, I think it is the right thing to do to review those ethical rules which are dictating a 12-18 month minimum path to the release of a vaccine. We *might* find some potential shortcuts.

Archived Facebook Coronavirus Posts: March 13th (post 2)

[I'm transitioning a bunch of Facebook posts I made on the Coronavirus to my blog.  This was from March 13th]

I thought it would be worth posting this video, which is of someone who has (mostly) recovered from the coronavirus describing what the early symptoms were like. https://www.youtube.com/watch?v=H2E1t3yMXgE&t=1800s

Of importance, I thought, were how mild the initial symptoms were: just a cough, really. The fever was tentative and barely registered at first.

The implication of this is that self-quarantining does need to happen quite aggressively, as soon as even mild symptoms arise. The *further* implication of this is that it would be really great if we had reliable and very easily accessible testing, because the stage at which you want to self-isolate with the coronavirus is the same stage at which it looks just like the cold or allergies or something. It would be really nice to be able to tell the difference very early on.

Archived Facebook Coronavirus Posts: March 13th

[I'm transitioning a bunch of Facebook posts I made on the Coronavirus to my blog.  This was from March 13th]

Some new information (to me) on COVID-19 coming out of the UK that has some specific implications for self-isolation: https://www.youtube.com/watch?v=BYTFk34nhoI

Key new information:

* Infectiousness of the disease starts before any symptoms show, but peaks around the time symptoms first show up. Then once symptoms begin, infectiousness goes way down after around 7 days.

* The most characteristic elements of COVID-19 are still a fever and/or a *dry* cough. The fever, however, is often not high--the UK criteria is only 100 degrees Fahrenheit.

* It is now confirmed that children do get the infection and spread it; it's just that they very frequently get an extremely mild form of the disease. While this is good news in the sense that most children are safer than everyone else, it does kind of make children our "plague rats" in this outbreak.

So, what are the self-isolation implications of these facts?

1. We need to have a high level of suspicion and self-quarantine *right away* as soon as even mild symptoms show up.

2. We especially need to pay attention to *dry* coughs and/or any level of fever.

3. We should probably quarantine our children if they are sick *at all*. Note that the UK is not recommending shutting down schools, though.

4. Seven days of self-quarantine should be enough unless you continue to get sicker.

Archived Facebook Coronavirus Posts: March 10th (post 2)

[I'm transitioning a bunch of Facebook posts I made on the Coronavirus to my blog.  This was from March 10th]

I'm annoyed by the persistent "coronavirus crisis is just being hyped by the liberal media!" theory people seem to be clinging to. I've been following this story very closely and I just do not see that as being the case.

If you still think the worry here is just caused by media overreaction, I would encourage you to watch and compare these two YouTube videos:

First, here is video about the new coronavirus from the early days of the outbreak in China by Dr. Eric Strong, a medical professor at Stanford and a practicing hospitalist: https://www.youtube.com/watch?v=AajszIH2ahs

Main takeaway (paraphrasing): "It's early yet, there's a lot we don't know, but don't worry, this could be no big deal at all and you're much more likely to get the flu than this new thing."

And this is the latest video on the situation he posted today:
https://www.youtube.com/watch?v=qsJ7LYZhUlQ

Main takeaway (exact quote): "For the country as a whole, including its health care system: I think things are going to get bad . . . like, really bad."

Archived Facebook Coronavirus Posts: March 10th

[I'm transitioning a bunch of Facebook posts I made on the Coronavirus to my blog.  This was from March 10th]

Some comments on the data on this article:

https://www.bloomberg.com/…/how-bad-is-the-coronavirus-let-…

For the record:

My own prediction--based on various news items, the behavior of the virus in general, and *some* preliminary studies--is that we will ultimately find out the CFR to be on the bottom end of the range presented in this article (I think 0.5% is likely to be close and I don't even think it's the true bottom of the possible range) *but* we will ultimately see it as more easily transmissible than is currently being indicated by top health organizations like the WHO and the CDC.

The WHO has been pretty dismissive of the theories that the virus can propagate asymptomatically, whereas I have seen several case studies where asymptomatic transmission seems pretty well proven. Also, I've seen peer reviewed studies that indicate pretty strongly that the virus can spread over the air farther and more easily than the WHO has been indicating *and* linger on exposes surfaces a lot longer than they've been saying.

Oh, and one important caveat on this article: a crucial set of numbers this article does not deal with is the percentage of deaths *by age*. I know it is true of the flu as well that it kills the elderly much more commonly than anyone else . . . but I feel like this may be even more true for the coronavirus. For example, a full 99% of all deaths in Italy (at least initially) were people over 60 years old. So I think the CFR is going to vary heavily by percentage of elderly in a population.

You can see this in action by looking at what happened at the Life Care Center in Washington. This case is actually worse than most people realize, because the initial news headlines had it as "13 residents have died of COVID-19". In reality, a full 26 residents died in the span of about a week there, compared to a typical 3-6 residents a month who die normally. Only 13 of those residents tested positive for COVID-19 (2 were inconclusive)--but they *haven't done* post-mortem testing yet on 11 remaining. So it's a safe assumption that almost *all* of those 26 deaths were due to COVID-19; that's a CFR of at least 21% *so far* in that population. *And* there is another 40 or so residents still hospitalized and in serious condition; fewer than half of the original 120 residents were unaffected.

So, we can expect COVID-19 to be really bad when it hits a nursing home, and I expect the worst headlines going forward to follow the Life Care Center pattern.

Archived Facebook Coronavirus Posts: March 2nd (second post)

[I'm transitioning a bunch of Facebook posts I made on the Coronavirus to my blog.  This was from March 2nd]

(More morbid coronavirus musings)

It just occurred to me that with Buttigieg and Klobuchar out, all potential candidates for the presidential election are now in the highest risk age group for Covid-19 fatalities, the 70+ range (I'm counting four Democrats and Trump here). So what's the chances that the coronavirus claims the life of one of the candidates before November?

Earlier, I estimated that 15% of the population was a reasonable estimate for how many American's were going to become infected by the coronavirus. With all the hand-shaking and baby-kissing that politicians have to do, I feel that it is reasonable to up the probabilities for our candidates to a full 25%.

Now, Covid-19 gets far more lethal with the age of the patient, all the way up to a full 15% fatality rate for the 80+ crowd (according to my current data). Our current crop of candidates are in the 70-80 range, though, so this gives them a much better 8% fatality rate. This gives an even 2% chance for each candidate that he or she will not make it to November, and a grand total of about 10% that *one* of the five will be picked off before the end (though not necessarily before dropping out for other reasons).

Archived Facebook Coronavirus Posts: March 2nd

[I'm transitioning a bunch of Facebook posts I made on the Coronavirus to my blog.  This was from March 2nd]

(Warning: musing on fatality statistics and the coronavirus follows. Not necessarily edifying material . . . though I suppose it is true that Catholics are urged to consider the Last Things during Lent.)

I've been reading on how Case Fatality Rates (CFR) are calculated in order to try to figure out whether Covid-19 is actually going to be worse than the seasonal flu or not, and roughly how much if so. The CFR is defined as how many people who are infected will end up dying. For the seasonal flu, it's surprisingly difficult to calculate, but I see numbers *in the US* generally from 0.02% to 0.08%. I've also seen a lot of people throw a "0.1%" number around, but I think that's a *worldwide* estimate. Death rates vary substantially according to quality of medical care in these cases.

The important factor that makes this difficult to judge in the real world is the difference between what you might define a "case" to be. In the early stages of a pandemic, usually the only way to define a "case" is either "someone admitted to a hospital with the right symptoms" or "someone who tested positive for the disease". In *either* case, this will inflate the actual fatality of the disease because it will not count those people who came down with the disease in a milder form and did not seek admission to the hospital, nor ever get tested.

In studies that have calculated the CFR from hospital admitted cases, we have seen estimates ranging from just under 1% up to 5%, but most people think these numbers will get smaller after all infected and not just hospital cases are considered. Can we attempt to do that now?

One approach would be to use the Diamond Princess cruise passengers as guinea pigs. In this case, with a captured audience that was under intense scrutiny, I think we can be confident that we actually know all the people who were infected. Everyone was tested and nobody who got sick in that situation would have shrugged it off as "no big deal". So what do the numbers that we know from the Diamond Princess cases tell us?

As of today, 6 out of the 705 people who have been confirmed to have caught the disease on the ship have died. Assuming all infected people were caught, this makes for a CFR of about .8. *But* we also need to correct for another factor, which is age. Cruise ships have a higher proportion of elderly passengers compared to the general population. My methods here are not very exacting (basically I found a graph of passenger age distribution on cruise ships and a graph of the general age population and squinted at them side-by-side), but I would estimate that cruise ships have 2-3 times the number of the crucial >70 age group compared to the general population. At least 5 out of the 6 people who have died so far from the Diamond Princess were older than 70. So in estimating the CFR from the Diamond Princess, I would divide by 3, and then adjust up a little bit more to account for the fact that there are still some passengers sick who may yet die. This gives me a real-world CFR of about 0.3%.

Two big problems with this estimation strategy: First, this is a small sample size. We really shouldn't put too much weight in what we can learn from a population of under 1000 patients. Second, as mentioned there are still people in this population who have not completed the course of the illness. So it is quite conceivable that another 20 people could still die from this group, which would dramatically change the final CFR.

Supposing, though, that 0.3% is a correct CFR for Covid-19. How many people would we expect to die from this illness if it becomes epidemic in the United States? This is also very difficult to determine, because we know even less about the total number of infected the coronavirus is likely to cause than we do about its fatality rates.

The seasonal flu is estimated to infect between 3% and 11% of the United States population each year. The new coronavirus is liable to infect more people: it has been estimated to have a base infectivity rate well over double that of the seasonal flu, it spreads asymptomatically or close to it, there is no "herd" immunity to the virus in the population, and it is unlikely to be slowed by a change of seasons (this normally happening to viruses *after* they become endemic in a population and not during their first introduction to the scene). So I think a guess of 15% of the population being infected is pretty reasonable.

This works out to a total death rate in the United States of about 150,000 people this year, overwhelmingly from the >70 year old population, but hitting a fair amount of younger people as well.

One important caveat to this number: the difference between a 1% hospitalized CFR and 5% CFR in China is largely between the Hubei province and everywhere else in China. To the extent you trust these numbers, the natural corollary is that as hospital systems get overwhelmed, as happened in Hubei, the fatality of the disease increases as the ability to care for the critical cases goes down. So *if*, in the United States, we start to see hospitals overwhelmed with critical cases, we could see the death rate go up to something closer to the 500-600 thousand mark.

Archived Facebook Coronavirus Posts: Feb 24th

[I'm transitioning a bunch of Facebook posts I made on the Coronavirus to my blog.  This was from February 24th]

OK, so--realistically--the coronavirus is a pandemic now. Community spread in Asia, the Middle East, Europe and (from what I'm hearing though this is not confirmed yet) Africa is now happening.

What does that mean for us?

1. At the governmental level, I don't think it changes a lot. Any government that wasn't already preparing as hard as possible for the pandemic was foolish. Furthermore, restricting travel and trying to track down sources of infection and quarantine them off is still worth doing. I don't think it's realistic to think we can *halt* the spread of the virus, but it should still be possible to *slow* the spread of the virus.

This is very important, because however bad this virus will be, it will be much, much worse in areas where it hits *suddenly*. The worst danger is the overwhelming of medical resources, where we are unable to care for all of the seriously ill because too many are infected at once.

So I think aggressive travel restrictions are still a good idea.

2. At the personal level, again I think the most useful thing is to do things that will slow the spread of the virus. That means, cutting down on unnecessary travel, cutting down on unnecessary social mingling. Stop shaking hands; try to keep more social space between you and other people. If you can, avoid visiting hospitals.

More importantly, I think it means responsible self-isolation. Do you have one child with a fever? Keep them all home. (Note: this is good advice even in normal flu season). Do you feel just a bit sick, but still well enough to do things? Unless necessary, stay at home! This particular virus spreads as badly as it does largely because it can spread from people with little to no symptoms.

For Catholics, if you are even a little bit sick, I'd recommend not receiving Communion on the tongue. I'd recommend either receiving on the hand or just abstaining.

I'm not suggesting here that we paralyze society and freeze all social activity--for now, I think we should all just add reasonable restrictions that we can practice and still function. All of these restrictions exist in a continuum, and given that there is no evidence of community spread here in the States yet, I think we can decide to practice just the "low hanging fruit" restrictions for now. We can decide to ramp up the restrictions as circumstances dictate.

Finally, we should all take care to practice impeccable hygiene. Wash your hands frequently and for 20 full seconds (the time it takes to sing "Happy Birthday" twice). Stop touching your face with your hands. Cover your mouth with the inside of your elbow when you sneeze or cough. Etc. There's no recommendation here that's not also a good idea in regular flu season, so I think now is a good opportunity to learn how to tighten up our regular hygiene practices.

Archived Facebook Coronavirus Posts: Feb 14th

[I'm transitioning a bunch of Facebook posts I made on the Coronavirus to my blog.  This was from February 14th]

With multiple children who are at relatively high risk to a pneumonia-causing disease, I have been following news of the novel coronavirus disease (aka Covid-19 now) pretty closely. I now think it is pretty likely--maybe more likely than not--that Covid-19 will become a global pandemic that will reach to the United States.

At any rate, it is now very reasonable to start talking about common sense preparations people can do *in case* we do end up with a medical crisis on our shores.

I think the first thing to do is to understand what such a crisis would look like, *if* it did happen. So here's my attempt to pre-construct a worst-case scenario that is reasonable to consider:

---------------

1. A certain threshold of number of infected people is crossed in your community. Most (85%) of these people are showing very mild or even no symptoms and spread the disease around before they know to contain themselves. The disease is now essentially anywhere and everywhere in your community. Let's say for the sake of argument that the disease infects 25% of the entire population within a relatively short time period.

2. In about 15% of cases, the Covid-19 causes serious problems, specifically pneumonia requiring hospital aid. This contingent consists primarily (but not entirely) of vulnerable people, specifically:
* the elderly
* immunocompromised
* diabetics
* people with Down Syndrome
* smokers or ex-smokers
* healthcare workers or anyone who has to spend a lot of time in hospitals.

Depending on the whose numbers are accurate here, maybe 1 or 2 percent of the total who fall sick will die, mostly from Acute Respiratory Distress Syndrome, or ARDS..

3. Suppose you live in a town of about 20,000 people. With the numbers above, about 5000 people will become infected and 750 will become seriously ill. Of these, about 50-100 will die even if they receive full care.

4. In a town of 20,000, the local hospital will not have anything near the resources to deal with 750 seriously ill people in a short period of time. The primary stressed resources will be personnel, hospital space, and ICU resources, particularly ventilators.

5. The likeliest consequence, I think, is that a distinction will have to be made between *seriously* ill people and *critically* ill people. People with "just" pneumonia will be told to stay at home, self-care and self-monitor, with occasional visits from over-worked health care workers for check-ups. Only the people at death's door are going to be admitted to the hospital for ICU treatment. This is what (we're pretty sure) has been happening in Wuhan.

6. There is some evidence that this state of affairs greatly increases the case fatality rate. That is, even with full hospital care the CFR might be 1%, *but* some estimates put the CFR in Wuhan at 18%. This difference *might* be because of the large number of people who are being left at home out of necessity.

7. Note that for this scenario I have chosen numbers that are very plausible and have put forward as the best guess actual proportions by people I consider pretty expert and knowledgeable. However, the numbers are not certain and it is quite possible that they could be worse in reality.

------------------

So this is the plausible scenario that we might try to plan for. What can we actually do in such a scenario? Well, first I think we can all educate ourselves on how to avoid picking up viruses of this sort as much as possible--learn all of the hygiene rules and be able to self-isolate as much as possible.

Second, I think we should learn the symptoms of ARDS, especially those of us with vulnerable family members. I have two links for this: an excellent YouTube video on ARDS (https://www.youtube.com/watch?v=okg7uq_HrhQ) which includes some information on recent research on how to improve survival rates from ARDS. And also here is a symptom run-down of ARDS from healthline: https://www.healthline.com/…/acute-respiratory-distress-syn….

Archived Facebook Coronavirus Posts: Jan 30th

[I'm transitioning a bunch of Facebook posts I made on the Coronavirus to my blog.  This was from January 30th]

Typical symptoms of the new coronavirus (2019-nCoV):

98% of the time: Fever
76% of the time: Cough
55% of the time: Shortness of Breath

*Note the absence of sneezing and runny nose.*

I've been following the news on the coronavirus pretty closely for the past few days. I believe that it's too early to say yet how severe this disease will become, worldwide. I think that those news outlets that are already saying, "hey, the flu is a bigger problem than the coronavirus" are being premature. Obviously it's not right to panic or stress out about this, but I think it's good to get some information out at this point.

Specifically, I think it's wise to be aware of the particular set of symptoms that are characteristic of this virus; if in the next few weeks you come down with specifically that combination of symptoms, then either get tested or stay home--especially if you have been travelling or have been around people who have been travelling. (I mean, "stay home if you have a fever" is good advice regardless.) And if you come down with just a normal cold or flu--with the sneezing and the runny nose--then *don't* run to the hospital panicking about the coronavirus.

Why am I concerned? There are a few things that concern me about the infection data that's been coming out on 2019-nCoV, which you can see here: https://gisanddata.maps.arcgis.com/…/opsdashboa…/index.html…

1. The graph of confirmed cases over time (the yellow line in the bottom left) has been following an exponential curve and is not leveling off.

2. The ratio of confirmed cases to confirmed fatalities is sitting at around 2.1 % for the whole dataset *but* if you look specifically at Hubei (the origin of the virus and where it has had the longest chance to play out), the ratio is actually 3.5%. I'm thinking maybe the 3.5% is the more accurate number? For reference, the ratio for the typical flu is 0.1% and the ratio for the 1918 Spanish Flu was 2.5%.

So we *may* be looking at something that spreads aggressively in a flu-like manner but kills 35 times as much.

3. This data is primarily from China, and I take it as a given that China is strongly disposed to downplay bad news as much as it can get away with.

4. There are only about 100 cases outside of China so far, and so far no deaths from these cases. *Hopefully* this means the mortality rate of the disease isn't as bad as the Chinese data makes it seem--*but* since the rest of the world is clearly multiple weeks behind China in the timeline, I don't think we can say that is the case *yet*. The non-Chinese mortality rate is something I'm going to be paying particular attention to going forward.

So, I'm currently in a cautious, "let's pay close attention in case this is a real serious problem", state right now, because I think that's what *could* happen.

Supply Chain Environmentalism and the "First 90%" Rule

The "First 90%" Rule

In 1970, Congress passed the Clean Air Act.  Without any real engineering plan behind its numbers, government decreed that in 5 years the emissions coming out of cars should be 90% cleaner than they were.  And as it turned out, this worked.  The automobile industry was able to implement improvements that drastically improved air quality across the world.  I don't think the original goal was quite reached in 5 years, but it didn't take all that much longer than that.

In the 40 years after this success, the EPA continued to make the requirements for automobile emissions stricter and stricter.  More and more engineering effort has been required to meet increasingly strict standards.  However much money was originally spent researching, designing and retooling in order to meet the original emissions standards, surely at least an order of magnitude more has been spent in meeting the further emission goals of the EPA.  And the results have been forthcoming, however whereas the original effort yielded a 90% decrease in emissions, the subsequent efforts have had a harder time eking out improvements.  In terms of total percentage of emissions reduced, all those subsequent efforts have yielded about 1/10th the reduction of the original smog reduction campaign, with an end result that cars today produce about 98% fewer emissions than did their '60s counterparts.

What I am trying to establish here is not that the EPA's increasingly strict standards were necessarily pointless--I put a very high premium on clean air myself, and I think I prefer to live in a world with 98% fewer emissions compared to the '60s as opposed to only 90% fewer.  However, it is clear that in this particular case, environmental regulations have been subject to a law of diminishing returns.  The first and easiest improvements that were made were the cheapest to do and also the most impactful.  Subsequent improvements were more costly and of less impact, though not useless.

This law of diminishing returns is not guaranteed to be the case for every type of environmental regulation, but I strongly suspect that it does hold for most types of environmental regulation.  And this has some very important implications for the manufacturing and energy production industries.  It should cause us to be circumspect about applying too-stringent environmental policies on industrial sectors that are apt to move to areas of least resistance.

Energy Production

One great example is fracking.  Energy is a fungible good, and therefore energy production is a very mobile industry.  If a country doesn't produce the energy it needs itself, it will import it from somewhere else.  The means of producing energy vary greatly in how bad they are for the environment.  Coal, for example, is much worse than oil or natural gas, all things taken into consideration.

Fracking stacks up very well against coal as a cleaner energy solution.  It's much better than coal--but it's not zero-impact.  It does have environmental costs.  From what I've read, the worst of the environmental costs can be mitigated with some fairly simple regulation, but even with this "low hanging fruit" regulation implemented, fracking is definitely going to have some environmental impact.  And so some people are against fracking entirely in the United States and want it completely banned.

But the impact of a complete ban against fracking would clearly and necessarily be an increase in the use of coal and oil fired power plants; this is inevitable.  It would also clearly increase the US total reliance on imported energy--and the places from which we import oil and gas do not give anything like the same consideration to the environment as we do.  For the sake of small improvements to the environment specifically in the States, we would be increasing the total harm to the environment world-wide.  Some people are just fine with this trade-off; I am not.  I think we should accept a lesser good in order to avoid a greater evil.

Manufacturing

Likewise, factory manufacturing has largely left the United States.  This is largely due to the average wage of Americans compared to other parts of the world, but it is also due at least in part to stricter environmental regulations.  Factory manufacturers have been acting in a very predictable way here; they compete with each other in terms of driving down cost of production.  The ones that survive, therefore, find areas with cheap labor and lax regulations on how rigorously they have to deal with the byproducts of manufacturing.  Our minimum wage laws and environmental regulations therefore do not have the effect of eliminating dirty factories staffed by impoverished workers; rather, they push that reality to other places of the world where we don't have to see it happening.

Manufacturing in China, in particular, has been a grotesque confluence of capitalism and communism.  Capitalism's blindness to consequences apart from the bottom-line and Communism's innate brutality and willingness to subject both man and nature to social engineering has produced a slow-motion humanitarian and ecological horror story.  Because China really doesn't give a c**p about the environment, as the clouds of pollution regularly infecting its skies attest.  They are not doing even the lowest cost mitigation work in order to achieve that first 90% of environmental benefit.

Solution?

What I believe we should be aiming for, then, is some way to impose minimal environmental and humanitarian standards upon our basic factory manufacturing supply chains, across the globe.  Increasing environmental standards on manufacturing in the US or in Europe is probably pointless and quite possibly even counter-productive.  We might even look to decrease our own standards slightly, if it could reduce exportation of manufacturing to third-world areas.  What is far more important for the environment as a whole is some way of getting that first 90% improvement on the bulk of manufacturing.

This is obviously very difficult to achieve.  These countries that have no regard for basic human rights or basic environmental cleanliness are also pretty much all willing to lie and cheat on everything they do, economically.  For the automobile industry, we were able to make global improvements by imposing standards on what is produced.  We can test cars that are manufactured in China and empirically determine how well they do on emissions.  We can't directly test a batch of chemicals shipped from China to see how clean was the process by which they were produced.

If we find a solution for this problem, it will probably include some requirement of openness and willingness to be inspected by impartial investigators . . . and this sort of thing has proven very difficult to make workable in the past, given unscrupulous sovereign nations who have a vested interest.  This is a strong motivation for us to favor nations as manufacturing partners who do not own their own manufacturing industries.  It will be much easier for us to demand accountability and good practices from factories, say, in Mexico where the owner of the factory is not a government with a vested interest in controlling a target percentage of world production capacity.  What this argues for is some type of tariff on goods produced from socialist states--a "freedom index" tariff of some sort.

Finally, I think it is very important for the sake of any effort to clean up the supply chain that we eschew environmental extremism.  It has been the nature of the environmentalist movement for some time that it lets the perfect be the enemy of the good; for the sake of a much cleaner manufacturing supply chain, we should very vigorously fight that tendency.  Let's work on making the supply chain "OK" rather than "pristine", because moving it to "OK" from "horrible" is going to be work enough for a lifetime.  Let's remember the importance of that first 90%.