Alexander Evsin
Against the backdrop of the panic "We are all going to die!", Terrifying revelations of doctors and patients, quarantined states and apocalyptic news from Italy, I want to hear the voice of common sense.
Alexander Evsin - head of the situation center, deputy head of the data center (Center for Traffic Management of the Moscow Government). At the moment, the shift on duty is involved in large-scale anti-epidemiological measures in the city - in particular, it provides traffic management in the area of the construction of a new infectious diseases hospital.
Alexander Evsin himself is a specialist in assessing the degree of threats of various kinds, including large-scale epidemics. He is an analyst, and his posts on the Internet are interesting primarily because they are calm and logical.
You are criticized for writing about the epidemic without a hitch and knowledgeably, but you are not a medic.
I have worked in military intelligence for 17 years, where I have always been serious about global threats. By education - engineer-mathematician. I have been working professionally for 25 years in the field of information processing and its evaluation. I have vast practical experience in researching data on the widest range of applied areas. This always includes immersion in the subject area, study of the issue and consultation with specialists.
Of course, I studied the existing problem to the extent sufficient for risk assessment. So you can consider my personal opinion regarding COVID-19 as the look of a specialist in probabilistic-statistical methods of analysis.
And what does the analysis of statistics say?
In order to analyze the development of the process according to statistical characteristics, it is necessary to have data with a clear collection methodology, a methodology for registering primary data. What do we have in the case of the COVID-19 epidemic?
The most unreliable characteristic, as I see it, is the number of infected people. Because everywhere they use different counting systems. Some purposefully examine all patients with signs of acute respiratory viral infections, others look only at the most severe, still others look at the dead, fourth look at risk groups, and fifth look at small groups of random people. And nowhere do they examine all citizens in a row. Plus, in many countries or regions they simply did not test for COVID-19 due to the lack of tests.
- "But a much more accurate characteristic is the number of deaths of patients with coronavirus?"
- There are also possible options that significantly change the picture in local clusters, because in the group of deaths from coronavirus, the dead with coronavirus are introduced in some places.
Let me cite the example of the death of Francisco Garcia: a Spanish football coach died of coronavirus at 21. Such newspaper headlines came out all over the world. However, it soon became clear that the unhappy young man, having been in the hospital with symptoms of acute respiratory viral infections and pneumonia, was sick with leukemia, which he did not even know about. Leukemia, among other things, causes extreme vulnerability to any kind of infection. But he added to the statistics of the victims of the coronavirus.
Also, at first in Moscow, a patient suffering from many chronic diseases was recorded as a victim of coronavirus. And she died from a detached blood clot. Then our doctors sorted it out and, as far as I know, stopped considering coronavirus the cause of her death.
In other countries, it seems that any deceased replenishes the statistics of victims of the epidemic simply by the fact of the presence of a coronavirus in his body. Regardless of health effects. It is obvious that the Russian approach is more correct!
What else can be said about statistics? The data stream is of poor quality, dramatically clogged, so at the moment it cannot be the basis for any reliable modeling; it is impossible to trace how events will develop further.
The situation is aggravated by the fact that people do not understand the difference between mortality from infection in a population and mortality of the infection itself. Identify the speed of detection of virus infected and the real dynamics of the process.
Even WHO has already written that the real mortality from infection is lower than that detected at the moment. And pure mathematics says the same thing.
If the dead are detected reliably enough, the number of infected people is very unreliable. The latter, of course, much more!
- Like this?!
"Just don't faint." In the vast majority, the infection proceeds in a mild form, and sometimes asymptomatically. Maybe you and I already got sick and recovered, they just didn't do the tests. What does this mean? That the mortality rate is significantly overstated.
This is not virology, but mathematics. If in the expression M = (N dead / N infected) * 100 N infected are understated, the final M will be unambiguously exaggerated!
- Then how to understand the degree of danger?
- The degree of danger of coronavirus from the point of view of statistics can be determined only in comparison with the "normal" mortality rate of past years. Alas, we cannot trust this year's statistics due to their defiantly poor quality.
Let us try to compare two other comparable and quite reliable characteristics - the level of overall mortality from diseases over the past years and this year. If a surge this year is clearly visible, then we can say that this is a new threat of extreme degree.
While such a surge is not visible. Moreover, on a global scale throughout the world and for the whole year it will not be noticeable, since more than 57 million die every year, and since December 2019 only 16 thousand victims of coronavirus have been registered. This is 0.03% of the total number of deaths. Therefore, more local and detailed samples should be studied.
However, we have almost online alerts for coronavirus victims, but we don't have the same information about the total number of deaths in total. Internet mortality data is an approximation and not primary data. Approximation of the data, of course, will not be able to show a new cause of mortality, which occupies only 0.03% of the total.
It amazes me endlessly that the exact numbers are still not published on official, trustworthy resources. Moreover, the analysis of the available statistics does not give exactly any reason to expect an apocalypse.
I emphasize: only analysis of statistics. Because there is news from Italy, where, judging by what they write in the media and social networks, everything is very serious and tragic: more than 6 thousand dead with approximately 60 thousand infected. This gives a monstrously high mortality rate of 9-10%. This is not in any country. For example, in Germany, the death rate is 0.25%, which generally corresponds to the level of risk from seasonal flu. But viruses do not kill on a national basis ...
- Does information from Italy seem untrue to you?
- We discard the version of its unreliability. Let us draw attention to the key points of what the surgeon of the hospital in Bergamo, Daniel McKini, is at the epicenter of events.
"With amazement, I watched the reorganization of our hospital when the enemy was not yet so strong. The wards were released for new patients, intensive care was expanded. The emergency room has been converted to reduce the spread of infection. In the empty corridors, surreal silence was established. It's as if we were preparing for war. "
That is, we see that immediately before the influx of patients, the hospital was reorganized to receive a large number of people - and soon 20-30 people a day with serious pneumonia, SARS, etc. really started coming to them.
As a result, respirators and ventilators were no longer enough ... Most likely, the authorities did what the frightened society demanded of them: they isolated patients with coronavirus in a separate clinic, which was choked with such an influx.
- Could this lead to a collapse?
- Theoretically, yes. The hospital, allocated "under the coronavirus", received an order of magnitude more patients in serious and extremely serious condition. Let me remind you that a life-threatening condition from coronavirus occurs as a result of the development of pneumonia. Thus, the hospital is filled with severe patients with pneumonia of various etiologies. More than 17 million people annually suffer from pneumonia in the world and about 300 thousand die. The disease is contagious.
The danger of nosocomial infections should also not be underestimated!
In the history of medicine, there are cases when massive cross-infection occurred in hospitals. Even the widespread use of ventilators can become an independent threat in such conditions. After all, if the infection gets into such an apparatus, then it will be brought directly into the lungs of everyone else.
- So you think that the true cause of what is happening in Italy is not the particularly deadly properties of the virus, but the initial mistake in the logistics of patients?
- I can't talk about the true reason. But I believe that the same disease cannot have 50 times different mortality. Therefore, I am looking for the factor that determines the difference between 10% mortality in Italy and 0.25% in Germany. To verify this version, you need to know how the distribution of Italian patients to hospitals took place. You understand that you can arrange a DDos attack on any hospital simply by making the wrong transportation decisions.
It was reported that some clinics panicky get rid of patients with COVID-19, sending them to specialized places where resources are already lacking ...
Does an international commission work in Italy to figure out what is going on? Are they treated properly at all? It is reported that fungal infections are also found in the lungs of the deceased, and this is a consequence of severe immunosuppression, which can occur not from the virus itself, but from excessive use of antiviral drugs, including serious drugs used to treat HIV.
In conditions of excessively overestimated danger, COVID-19 is at risk for patients with other serious diseases: from diabetes to coronary heart disease.
When the clinic staff focuses only on the fight against coronavirus - of course, this can lead to new deaths.
- Watching what is happening on the Apennine Peninsula, Russians fear for their elderly relatives. Especially frightening is 8% of deaths at the age of 70-79 and 14.8% at the age of over 80.
- Unfortunately, people die without coronavirus. There are open sites where you can find out general information about mortality at a particular age. I will take the extreme values: for the ages of 10-14 and 75-79 years. Data taken from 1959-2009.
"And what do we see?"
- In 2009, for example, the overall mortality rate (for any reason) between the ages of 10-14 was 0.2%. Compare with the chart of mortality from coronavirus - the same 0.2%. At the age of 75-79 years, the total percentage of deaths (for any reason) was 30%. Compare with the schedule for coronavirus - there is only 8%! Three and a half times less !!!
- In the article "Coronavirus: we must act now", which was distributed in hundreds of thousands of reposts, blogger Thomas Pueyo draws graphs with direct extrapolation of data and a sharp geometric increase in the number of infected and dead in the very near future ...
- Who wrote this article? Mathematician? Virologist? Doctor? No. The blogger! What did the author publish before? Here is a translation of some names: "How to become the best in the world in something", "What Skywalker can teach", "8 reasons why your approach to explaining the reasons for other people's behavior is wrong", "How to manage the CEO's expectations" ... This just a dude who writes about everything in the world. He certainly cares about the virality, but in the media plan.
- And what is he wrong about?
- Honestly, in everything. Starting from the misuse of the source data and ending with the complete disregard for the biological, medical, and administrative components related to the spread of the virus and its consequences.
Any pathogenic virus, when detected and studied, first gives high statistical mortality rates. The reason is simple: when the first detection occurs, the proportion of severe outcomes is always high.
Because the sample comes from the most severe cases. You do not run to the center of virology with a cold? ..
The initial stage leads to a frightening statistic. It goes to newspapers with headlines like: "20% of people die from a new virus! Every 5th! " Society is tense. Meanwhile, specialists are beginning to look for where the threat came from, how many carriers ... Then another scary figure appears: the dynamics of the growth of patients. Every day, 5-10-20-100 people are found. But the reason is not in the outbreak as such, but in the targeted detection of new infections. What does the average person see? Next headings: "Growth dynamics - 200%!" ...
In my opinion, Thomas Pueyo draws inadmissible arithmetic conclusions from process indicators without understanding the essence of the processes themselves.
I can give a household explanation showing why direct extrapolation of data cannot be used. For example, in your family a girl was born in 2004, a second child appeared in 2008 - a girl again, a third - four years later, and this is a girl again. How reasonable is the conclusion that in 2024 you and your wife will have 6 girls? Not at all. This is a meaningless conclusion from the point of view of the real process, but absolutely true from the point of view of extrapolating data.
The blogger in his pseudoscientific work asks the question: "In South Korea, there was a frantic growth of COVID-19 - it is curious why this did not happen in Japan, Taiwan, Singapore, Thailand or Hong Kong?" I can answer why. Because Korea at the beginning of the epidemic organized a massive virus test. Anyone who sneezed was immediately diagnosed. And other countries surveyed only heavy ones. There are fewer infected, but a frightening mortality rate among those who have found the virus.
The dangers of COVID-19 need not be minimized. It is dangerous. And for some people at risk, it can become fatal. But this is not a reason for global panic. This is the reason to listen to the advice of doctors.
- So is strict isolation of citizens necessary then? Wuhan, where the epicenter of the epidemic was located, cites an example of perfectly organized quarantine - and suddenly everything began to decline ...
- I am against hard quarantines that stop the life of the city. Maybe China, with its powerful economy, can survive this, but for other countries it can end up much worse.
I consider it more correct to protect against the risk of infection at risk, that is, the elderly and the chronically ill. This is a more effective, in my opinion, model of behavior.
Excessively hard quarantine, if it gives a result, it simply transfers the peak from March to May, for example. But at the same time it will hit hard in all areas of life.
Undoubtedly, the death of every person is a tragedy. But we are talking about assessing the degree of danger. Only in Russia in 2018, 18 thousand people died in road accidents. An average of 45 people a day. About 2 people per hour! But no one trembles with horror, sitting in the car. And because of the coronavirus, air communications are interrupted, armies cordon off cities, billions are spent on protection from one of hundreds of ARVI pathogens ...
Either a really spontaneous media effect has occurred, or it is a prudent injection plan. Because the fall of the markets a week ago was estimated at more than 4 trillion dollars!
- The security measures that are now being taken in Moscow, right up to the most pessimistic option - with the closure of the city, also seem somewhat redundant?
- This is how modern mechanisms of social organization work. I must say that I perfectly understand the officials who are forced to make such decisions. They are logical in terms of mass communications. Although it is irrational in relation to objective reality. A very interesting phenomenon, which should certainly be carefully studied.
In my opinion, this is indeed an epidemic, but not in the medical sense, but in the communication - the panic virus.
Now even true liberals urge the state to tighten everything and everything. At the same time, let's be honest, on the part of objective data there is no reason for such a sharp restriction of rights and freedoms, a ban on businesses ...
Panic expectations, coupled with the disruption of many businesses, can be a much greater threat than infection. Disruption of food supplies to cities, the unemployed, left without a livelihood, are far from illusory categories.
- Are you more a skeptic or a pessimist?
"I'm used to being a pessimist." But at the same time I have an analytical mindset. A true analyst should doubt, trusting only reliable data and consistent logical conclusions.
My skepticism about the panic of recent days is not evidence that I and my colleagues are idle. Moreover, I am aware that I may not know something. Nevertheless, my concerns about the virus itself are moderate. No matter who calls it, this is a type of acute respiratory viral infection - an acute respiratory viral infection.
Despite all my personal views, I am obligated to comply with all orders of the authorities - and I will do so. At the moment, the Moscow government is constantly and with full dedication working on very large-scale anti-epidemiological measures that fully comply with the already formed panic public request.
The authorities are in a difficult situation: the panic request is so strong that the authorities will be forced to reinsure many times. The key to removing these restrictions is only in ourselves.
- The wave of the fight against coronavirus has also reached the Center for Organization of Traffic?
- Sure! Moreover, the data center went into such a regime earlier than citizens saw it. Since the end of February, official measures are being taken for disease prevention measures. Recently, about 34 km from MKAD, in the area of the village of Golokhvastovo, the construction of a new infectious diseases hospital has begun to provide potential treatment if there is a sharp increase in the incidence of COVID-19 coronavirus.
The road patrol of the data center in the construction area organizes traffic, the mobile situational center is an information processing center. The shift attendant constantly monitors information related to the epidemiological situation. Search and identification of taxi drivers who look unhealthy, catch a cold ...
Therefore, when asked why I am interested in the virus, that's why I am interested.
With the current informational background, refusal of activity would lead to an avalanche of reproaches, mistrust and indignation. Everything and even more is being done in Moscow to provide any option for further events.
- Imagine, the authorities are doing so much, but the end of the world will never come. And everyone will say: "Oooo!" So there was nothing serious! "
- It will be unprovable and unknowable. Six months have passed - the apocalypse never happened. Chthonic unknowability lies in the fact that, naturally, there will be no catastrophe. But you can never prove that it could not be.
I think that now society has divided in its relation to the epidemic. And that is a matter of faith. Someone later will believe that the world is saved, and some - that the world was deceived.
"And what does the analysis of statistics say?"
R.C.