Some quick (okay, long) updates on the current state of the world and covid-19 as I see it.
Subjects covered: medicines, vaccines, death rate, herd immunity, strain mutations, humidity/temperature, accuracy in statistics, long term effects of infection, course of action in LMIC vs developed countries, and a prediction on the stock market if anyone reads to the very end (anyone who reads this entire thing deserves a financial incentive/bribe).
Drugs
We shouldn’t be holding our breath for treatments like chloroquine or remdesivir anytime soon. Results are still preliminary, but so far the hardest data we have suggest little to no benefit at best, and active harm at worst.
Remdesivir is also inordinately difficult and expensive to make, and even in the best case scenario, Gilead is hoping to have a mere 1 million treatment courses available by end of the year — for a global population in the billions.
It will only be at best available for a fraction of the cases in rich countries, and right now it looks like its best hope is to be helpful for non-severe cases, of which there are far too many to possibly hope for it to have any meaningful impact.
At best, it’ll help the rich and famous escape serious harm when taken early on in a course of treatment before the disease progress becomes severe, but the average person can forget about it.
Chloroquine:
Researchers analyzed medical records of 368 male veterans hospitalized with confirmed coronavirus infection at Veterans Health Administration medical centers who died or were discharged by April 11.
About 28% who were given hydroxychloroquine plus usual care died, versus 11% of those getting routine care alone.
Remdesivir:
It showed that researchers studied 237 patients, administering the drug to 158 and comparing their progress with the remaining 79, who received a placebo.
After a month, 13.9% of the patients taking the drug had died compared to 12.8% of those receiving the placebo. The trial was stopped early because of side-effects.
“Remdesivir was not associated with clinical or virological benefits,” the summary states.
Those 1.5 million doses — roughly equivalent to 140,000 treatment courses — are either ready for distribution or in the final stages of production, and Gilead hopes to have another 360,000 courses ready to go by October, O’Day said. By year-end, if the company meets its goal, it will have turned out 1 million treatment courses.
Vaccines
A vaccine is an even more speculative bet at least 18 months out at best, and even that would be smashing any historical record for novel vaccine development.
Vaccines have never before been successfully developed for any coronavirus strain either, and the methods employed by the earliest vaccine trials are extremely speculative with many having never been successfully ever employed before for vaccine development.
There are reasonable odds we’ll eventually get a vaccine, and covid-19 does seem to be far more genetically preserved than common cold coronavirus strains, but it’s far from a sure bet and nothing people should be taking remotely as a given anytime soon.
“Like most vaccinologists, I don’t think this vaccine will be ready before 18 months,” says Annelies Wilder-Smith, professor of emerging infectious diseases at the London School of Hygiene and Tropical Medicine. That’s already extremely fast, and it assumes there will be no hitches.
The field of mRNA therapeutics is still relatively untested, with no mRNA vaccines or other therapeutics yet approved.
Death Rate
The overall population death rate probably falls into the magnitude of somewhere around ~0.5%, give or take a bunch. A number of various approaches from serology studies to closed data set population studies aboard cruise ships point to this broad order of magnitude.
Lest we think this a small number, note that at an estimated global herd immunity rate of 70% of the population overall being infected, this amounts to 27 million dead globally, and over 1 million dead in America — on the same order of magnitude as the 1918 Spanish Flu.
This is a good read on estimating the IFR and CFR rates from the Diamond Princess cruise ship, which is the one closed population we have the most information on — 3,711 total crew and passengers aboard, every single person tested, 712 confirmed infections, 13 deaths (1.8% death rate).
Normalizing as best as possible for the aged population and other factors, a reasonable IFR the researchers here found was 0.5% with a 95% confidence interval of 0.2–1.2%.
Herd Immunity
The whole herd immunity thing is the worst case scenario. We’re starting to get early serology results back, and while these should be taken with a huge grain of salt and massive caveats on specificity and sensitivity, if we take them directly at their word, the numbers broadly check out.
A small sampling of 3,000 shoppers in New York estimates broadly 13.9% of New Yorkers have had covid-19, which means we’re less than an order of magnitude away from herd immunity. This checks out with the 0.5% death rate figures above — we’re at 20,861 dead in New York out of a population of 19.5 million, or 0.1% (literally one out of every 1000 people in New York has died of covid-19 already, a truly unfathomable number).
At 13.9% of the population, that’s a 0.769% death rate (which may very well be substantially undercounted as excess deaths exceed the official covid-19 death rates in NY), and at a 70% herd immunity infection figure in NY, we would net out to just about exactly a 0.5% overall death rate for the population at large among everyone with or without the actual infection.
One should actually hope that *more* than 13.9% of New Yorkers have had covid-19, because if it’s really that low, the death rate is higher than feared.
Reasonable estimates of when we’ll hit herd immunity for covid-19 are around 70%:
For mumps, you need 92 percent of the population to be immune for the disease to stop spreading entirely. This is what’s known as the herd immunity threshold. COVID-19 is, fortunately, much less infectious than mumps, with an estimated R0 of roughly 3.
With this number, the proportion of people who need to be infected is lower but still high, sitting at around 70 percent of the entire population.
Which brings us to why herd immunity could never be considered a preventative measure.
If 70 percent of your population is infected with a disease, it is by definition not prevention. How can it be? Most of the people in your country are sick! And the hopeful nonsense that you can reach that 70 percent by just infecting young people is simply absurd. If only young people are immune, you’d have clusters of older people with no immunity at all, making it incredibly risky for anyone over a certain age to leave their house lest they get infected, forever.
People keep talking about Sweden doing relatively well despite not locking down and aiming to hit herd immunity. These claims are patently absurd. The only reason Sweden’s numbers look good is because their population is literally 3% that of the US.
Their per capita death rate for the entire country already exceeds that of the US (200 per 1mm vs 150 per 1mm), despite having a lower overall population density by far compared to the US.
Their death rate completely dwarfs that of their comparable neighbors, Norway and Finland (213 per 1mm vs 37 for Norway and 32 for Finland — literally 6X as many deaths and a far higher growth rate), and if you actually just look at Stockholm, the numbers are nothing short of horrific — over 1100 of the deaths have occurred in Stockholm County with a mere population of 2,377,081, coming out to a 462 per 1mm death rate.
Note also that the epidemiologists directing Sweden’s response are horrifically incompetent. They keep strutting around presenting themselves as a paragon model of success and righteousness and criticizing everyone else for their presumably incomprehensible lockdowns, and yet they are incapable of literally performing basic two second mathematical sanity checks.
The deputy state epidemiologist hilariously claimed that for every one person that tested positive for COVID-19, there were likely 999 other people who had the infection without knowing. This was immediately called out as ludicrous, as that would mean that more people have had covid-19 than literally exist in Sweden by far.
The error was later acknowledged as being the result of an idiotically cobbled together powerpoint presentation that apparently the deputy state epidemiologist had just absorbed without a mote of critical thought and had just parroted blindly at a literal press conference, despite the fact that even the journalists (who are not professional epidemiologists, but apparently more capable of performing basic mathematical operations) at the press conference were able to easily spot the absurdity and call it out:
During yesterday’s press conference, the deputy state epidemiologist Anders Wallenstein announced the report based on data models that indicated the spread of the coronavirus was much higher than previously thought. Among the claims were that for every one person tested positive for COVID-19, up to 999 others could also have the infection without knowing.
Swedish journalist Emanuel Karlsten queried this during a Q&A session at the press conference. Given there are more than 15,000 confirmed cases, the Agency’s estimate would mean that the total number of likely infections would be greater than the entire population of Sweden.
They also had to retract their initial model that was highly covered in the press in less than 24 hours after discovering a glaring error. Would think twice before taking anything these people put out at face value.
All their modeling is predicated on extraordinarily sparse sampling because they have horrific testing rates. They’ve only conducted 94,500 tests with 16,755 positive cases, which comes out to an absurd 17.7% positive test rate. If we take them at their confirmed testing numbers, they have a hilarious 12% death rate right now (2,021/16,755). Their entire modeling schemes are predicated not on a comprehensive accurate overall testing regime, but rather just extrapolative speculative sampling.
Mutations
Don’t worry too much about strain mutations. Here’s a good tweet storm about it:
Covid-19 is pretty strongly conserved — we note things like literally 5–10 base pairs of mutations for a genome of 30,000 base pairs — something like 99.97% conservation. To take the most obvious comparison, influenza, which mutates far more rapidly than covid-19 seems to, has never caused a pandemic through mere antigenic drift.
All the instances of pandemic flu have been caused by a single major shift driven by zoonotic transmission from an influenza A strain that has been slowly mutating in a parallel tract in some other animal species separate from humans before making a sudden leap and introducing a dramatically different influenza strain from anything any human has had immunity to before.
Mere drift has never produced pandemic results or caused an extraordinary uptick in fatality rate, even when it occurs at a more aggressive pace as is the case with influenza.
While influenza viruses change all the time due to antigenic drift, antigenic shift happens less frequently. Influenza pandemics occur very rarely; there have been four pandemics in the past 100 years. For more information, see pandemic flu. Type A viruses undergo both antigenic drift and shift and are the only influenza viruses known to cause pandemics, while influenza type B viruses change only by the more gradual process of antigenic drift.
Humidity and Temperature
Humidity and temperature are probably not going to save us. Ecuador, with its 80%+ humidity and 80 degree+ weather has experienced outbreaks that outpace NYC.
A New York Times analysis suggests that Ecuador’s death toll is 15 times higher than its official tally of coronavirus deaths, highlighting the damage the virus can do in developing countries.
In Guayaquil, fatalities during the first two weeks of April were eight times higher than usual, the data indicates — a far greater rise than that of New York City, where fatalities were four times higher in recent weeks.
Within weeks of the first case being identified in Guayaquil, hospitals were overflowing and burial systems collapsed under overwhelming demand, leaving dead bodies to pile in the streets and pushing families to bury loved ones in coffins made of cardboard.
The wave of deaths is all the more disturbing for being impossible to explain. There is no obvious reason for Ecuador to be devastated far more than other countries. Its population is relatively young, and most people live in rural areas, both factors that should reduce the risk, said Jenny Garcia, a demographer who studies Latin America at the Institut National d’Études Démographiques in France.
‘Outlier’ Outbreaks
This ‘impossibility’ of explanation leads me to only a few present reasonable suppositions (would love to hear any others if anyone has them) — barring something like Ecuadorians being more susceptible to covid-19 for some reason like substantively decreased health and higher disease load in general or something (in which case many developing countries are going to be doomed anyway) that is different from the current comorbidities most often studied in developed countries (like obesity, diabetes, heart disease, and all the other rich country diseases) but equally if not more devastating, the data suggests that most developing countries, if not all countries at large, are probably eventually going to get totally fucked.
Why Ecuador now and not other countries may only have to do with the exponential differential caused by small variations in early spread and initial cluster seeding in different countries — the same reason that Iran and Italy had massive outbreaks far earlier than other countries that are having them now. Ecuador is to Italy as Italy is to Wuhan, and all the other developing countries may only be ticking time bombs that will in months to come be Ecuadors themselves.
Undercounting
This leads to another side note — as much as developed countries like the US and Europe are undercounting deaths (see links below), developing countries and suppressive authoritarian regimes in general are likely undercounting deaths by untold, near unfathomable magnitudes. Iran comes to mind as a particular highlight — Iran had the first major outbreak outside of Wuhan, and by March 3rd, almost two months ago now, nearly 10% of their parliament (the hardest public figures to hide and deny) had been infected with covid-19, a number that far outpaces the rate of infection in other countries. They also have a higher comorbidity load of various diseases such as diabetes than even America.
With a population of 81 million and an economy and entire nation crippled by economic sanctions and unbridled spread months before almost anyone else that has both been unable and unwilling to truly lockdown, we really believe that Iran has only experienced 87,026 cases of covid-19 (0.1% of the population as of today, yet somehow 10% of parliament almost 2 months ago) and 5,481 deaths?
That’s fewer deaths than Belgium (6,490), which was hit far later and has a mere population of 11.46 million (14% of Iran) and has inordinately better and less overwhelmed medical care and far greater access to resources.
Extrapolating from the growth in New York and the near-herd immunity figures we’re seeing there even with a lockdown as soon as things started getting really bad, the most reasonable assumption I can see is to assume Iran got completely butt-fucked by this virus and at the very least 20%+ of the population (just a little more than New York) should be infected by now.
Assuming a similar death rate to New York (which is a very generous assumption since Iran was far more overwhelmed in their hospitals), over 80,000 people should be dead in actuality in Iran by now, making Iran actually the country with the largest number of deaths in the world.
This is just about the most conservative estimate and in all reality is likely far higher well into six figures already — and yet we’re all just here ambling along mostly taking Iran at their word that they have fewer deaths than Belgium.
The true death toll of covid-19 is very likely already far, far higher than it seems.
If not, everyone should board a plane for Iran right now, because they are by FAR doing infinitely better than anyone else who has been hit by a major outbreak of covid-19. Impossibly better.
Disease Trajectory
From what the early data seems to suggest, if you don’t beat this infection off when it’s still stuck in your nose and throat, before it gets into your lungs and then proceeds to spread aggressively to literally every part of your body from your brain stem to your GI tract, there are reasonably excellent odds that you may come down with long term/permanent damage, even if your course of infection is not severe enough to necessitate hospitalization and you recover just fine on your own in your own home.
Quite possibly 100% of hospitalized patients are going to have long term damage all over their bodies. Death is not the only metric to look at here. Outcomes are not binary — they are an extraordinary spectrum.
“[The disease] can attack almost anything in the body with devastating consequences,” says cardiologist Harlan Krumholz of Yale University and Yale-New Haven Hospital, who is leading multiple efforts to gather clinical data on COVID-19. “Its ferocity is breathtaking and humbling.”
Despite the more than 1000 papers now spilling into journals and onto preprint servers every week, a clear picture is elusive, as the virus acts like no pathogen humanity has ever seen.
In Brescia, Italy, a 53-year-old woman walked into the emergency room of her local hospital with all the classic symptoms of a heart attack, including telltale signs in her electrocardiogram and high levels of a blood marker suggesting damaged cardiac muscles. Further tests showed cardiac swelling and scarring, and a left ventricle — normally the powerhouse chamber of the heart — so weak that it could only pump one-third its normal amount of blood. But when doctors injected dye in the coronary arteries, looking for the blockage that signifies a heart attack, they found none. Another test revealed why: The woman had COVID-19.
How the virus attacks the heart and blood vessels is a mystery, but dozens of preprints and papers attest that such damage is common. A 25 March paper in JAMA Cardiology documented heart damage in nearly 20% of patients out of 416 hospitalized for COVID-19 in Wuhan, China. In another Wuhan study, 44% of 36 patients admitted to the ICU had arrhythmias.
Another striking set of symptoms in COVID-19 patients centers on the brain and central nervous system. Frontera says neurologists are needed to assess 5% to 10% of coronavirus patients at her hospital. But she says that “is probably a gross underestimate” of the number whose brains are struggling, especially because many are sedated and on ventilators.
Frontera has seen patients with the brain inflammation encephalitis, with seizures, and with a “sympathetic storm,” a hyperreaction of the sympathetic nervous system that causes seizurelike symptoms and is most common after a traumatic brain injury. Some people with COVID-19 briefly lose consciousness. Others have strokes.
On 3 April, a case study in the International Journal of Infectious Diseases, from a team in Japan, reported traces of new coronavirus in the cerebrospinal fluid of a COVID-19 patient who developed meningitis and encephalitis, suggesting it, too, can penetrate the central nervous system.
End up with covid-19, and you might end up unconscious barely holding on for dear life intubated with your leg amputated, even if you’re a totally normal 41 year old like this poor Broadway actor:
Autopsies have shown some people’s lungs fill with hundreds of microclots. Errant blood clots of a larger size can break off and travel to the brain or heart, causing a stroke or heart attack. On Saturday, Broadway actor Nick Cordero, 41, had his right leg amputated after being infected with the novel coronavirus and suffering from clots that blocked blood from getting to his toes.
Six scuba divers in their 40s who all recovered just fine without a severe course of illness in their own homes now have permanent lung damage and can never scuba dive again, despite feeling totally fine until exerting themselves:
He heads the emergency department in the hospital and is the responsible crisis coordinator for Covid 19 patients. In the clinic, doctors have treated dozens of coronavirus sufferers in recent weeks, from symptom-free spreaders to intensive care patients on the heart-lung machine. Among them were six active divers, all of whom did not have to be treated in hospital, but cured themselves in home quarantines. All of them were not severe cases, their illnesses were five to six weeks ago and they are considered to have recovered. But they can no longer dive. “The damage to the lungs is irreversible,” said Hartig in an interview with the APA.
This is shocking, we don’t understand what’s going on here. They are probably lifelong patients, so it doesn’t matter whether they dive again or not,” said the doctor. The bad news was made clear by lung CTs. “They didn’t get any better at all in imaging,” said Hartig. “As an emergency doctor with 20 years of experience, you swallow when you see something like this in a 40-year-old patient.”
In the control after several weeks, two patients showed significant oxygen deficiency when under stress as a typical sign of persistent lung shunt. In two, the bronchi were still very excitable when under stress, as in asthmatics. Four out of the six divers still showed impressive lung changes on the control CT. “I even called on the X-ray to see if they had swapped the pictures because a healthy patient was sitting in front of us,” said Hartig. “When they saw their own pictures, it was shocking for them,” said the doctor. “You have to check regularly with such lung damage.”
He therefore published an interim report in the diving magazine “Wetnote” to warn active divers. “After a Covid infection, even if you have only mild symptoms, you should definitely have a dive doctor examine you thoroughly, even if you still have an upright medical examination,” said Hartig.
These are truly unexpected outcomes — the flu definitely doesn’t usually go around fucking up every inch of your body. As Science Magazine noted, “the virus acts like no pathogen humanity has ever seen.”
To be clear, it still seems that the overwhelming mass majority of infected persons come out unscathed as they beat it off when it’s just still in their upper respiratory tract — this just highlights that if you get a severe course of illness, things can get really severe, almost unimaginably so. So if there’s ever been a time to stay as unimaginably healthy as you can possibly be, this is it.
Low and Middle Income Country Outlook
All of this said, my current read on the world is that low and middle income countries are completely fucked. They will likely be unable in full to pull off effective lockdowns long enough to prevent mass spread (at least 18 months for the earliest vaccine hopes, with dwindling chances of any treatment being found in the interim time that makes a meaningful difference and can be manufactured in time to be of use for these developing countries), and the cost of executing these draconian lockdowns does, in a tragic calculus, likely exceed the benefit of looser social distancing exhortations that still allow people to get enough food to not go hungry.
The world has never faced a hunger emergency like this, experts say. It could double the number of people facing acute hunger to 265 million by the end of this year.
Developed Country Outlook
For richer countries that can afford to print infinite money and meet its citizens’ basic needs indefinitely, the right approach is less clear. At the very least, it is almost certainly the right move to squash the curve as much as possible until it is certain the medical system will never be overloaded, that doctors and nurses and other medical workers will be fully adequately protected, and until we have enough insight into the course of the disease progression to be able to confidently recommend an authoritatively optimal approach to treatment and care, which in of themselves even without any medicine or vaccine being developed will dramatically lower the mortality and long term damage outcomes of covid-19.
We should likely go even further and ensure that the populace at large has access to basic PPE as well, and at the very least at least all the essential workers in places like grocery stores and transportation hubs. Massive testing on demand whenever there’s an iota of suspicion of infection should also likely be a pre-requisite for optimal outcomes.
And if we can successfully reduce the spread rate like China or Taiwan to a point that contact tracing can be successfully carried out, then this is an ideal outcome as much as one is manageable now. Taiwan has by leaps and bounds outperformed just about every single other country in the world here, and will now reap the immense beneficial fruit of their extraordinarily competent actions. Their citizens can go out without fear and keep the entire economy fully functional without a single death in ages.
These logistical efforts can yield immense benefits and successfully keep the virus in check without resorting to the worst case scenario of herd immunity and maxing out deaths long enough that we can at the very least provide the highest quality of care to every patient and at the very best provide a vaccine to eradicate this strain of covid-19.
Conclusion
All of that said, as the current state of the world goes, with the leadership (and lack thereof) and the unfathomable incompetency and missteps of the vast majority of governments and states here, I’m seeing the world as broadly pretty fucked.
Case in point, as of press time, President Trump was last seen suggesting that direct injections of disinfectant should be explored and sound like a promising treatment for covid-19.
Presumably, he feels as he did prior with an earlier exhortation to use chloroquine: “What do you have to lose”?
Really, you have to watch this:
Conclusion: Calls on SPY 400 5/1
