Get best of the week

Can an epidemic be predicted?

On Habré there were already articles about forecasts of a coronavirus:


I will present a different forecast and, most importantly, an analysis of this forecast. How is the forecast different from others?
Two things:

  • made much earlier than the previous ones (in 2013);
  • made it ... an astrologer.

Perhaps someone already had a desire to laugh here. Well, do not deny yourself - laugh heartily. A good mood is always good. As the King and the Jester sang there: "I sincerely ask - laugh at me if this helps you ...". And how much you laugh, I ask under the cut: I do not urge anyone to change their worldview. I propose to evaluate the work of the author of the article based on facts, not emotions. If the author of the forecast is not right - it is necessary to say this correctly. And if right ... think about the reasons for his innocence.

We have the following questions at stake:

  1. How accurate is the forecast?
  2. Was the forecast really made in advance, and not in fact (and simply issued later as an “ancient prophecy”)?
  3. Is it possible to make a similar forecast by accident? What is the probability of making a similar forecast by accident?
  4. What other predictions did the author make, how accurate are they?

If there are errors in my story, feel free to report it. We are all people who tend to make mistakes (and learn from mistakes). I will be ready to correct my mistakes.
Gentlemen minus - please do not be shy and speak out in the comments. And it’s better to bring facts.

How it all began



UFO Care Minute


The pandemic COVID-19, a potentially severe acute respiratory infection caused by the SARS-CoV-2 coronavirus (2019-nCoV), has officially been announced in the world. There is a lot of information on Habré on this topic - always remember that it can be both reliable / useful, and vice versa.

We urge you to be critical of any published information.


Official sources

, .

Wash your hands, take care of your loved ones, stay at home whenever possible and work remotely.

Read publications about: coronavirus | remote work

It all started with the fact that this very forecast caught my eye completely by accident. There on page 60 I found the following text:
Winter from 2019 to 2020.

At the end of 2019, the arrival of influenza should be expected in late October and until the second of December.
The second outbreak of the epidemic - from January 4, 2020 to February 16, 2020 ...

... The main area of ​​infection is the eastern coast of the United States (especially to
Philadelphia), from China (from 115 g.v.d.) to Japan, the Russian Federation - Northeast Siberia and the Far
East.
Communicating with people about the reliability of this forecast, I realized that many people make a subjective conclusion purely intuitively, without specific facts. Or distorting some facts, taking them out of context. Thus, I was tired of arguing with each individually and I decided to write an article.

What my interlocutors did not like about this forecast


If we discard claims of an emotional nature (like, astrology is not a science and this is enough to reject a forecast without even reading it) - the main claims in fact were as follows:

  1. Incorrect list of regions of distribution;
  2. Incorrect burst dates
  3. there were no 2 bursts (there was one);
  4. Forecast from the series “Finger in the Sky”: there may be a coincidence. But, after all, a broken watch twice a day shows the correct time.

Let's deal with all this, armed with facts.

?


First of all - you need to figure out where to get the facts and how much you can trust them. On Habré there is an article describing a part of sources:
Coronavirus: information online services

I compare with open data from the Internet. But, you need to understand that they are also not the ultimate truth. For example, in my sources now there is no data at all on North Korea. But, this does not mean that there is no infection. Also a question of trust in official information: we do not know how reliably they diagnose “COVID-19” all over the world. Maybe someone from the infected is given a different diagnosis (flu, SARS, etc.) while the data sources in this article are precisely COVID-19. By the way, the forecast simply indicates the flu (apparently, as a general description of diseases of the upper respiratory tract, where COVID-19 refers).

A complete forecast of the 2013 Irina Senina epidemics for the northern hemisphere is available in PDF form here . The document is called " Research on the influence of the planets of the solar system on the emergence and spread of epidemics ." Up to about 57 pages, the information is more for astronomers, astrologers and all others who are interested in what planet configuration was revealed in different epochs of epidemics.
We immediately move to page 60 and find the very text:
Winter from 2019 to 2020.

At the end of 2019, the arrival of influenza should be expected in late October and until the second of December.
The second outbreak of the epidemic - from January 4, 2020 to February 16, 2020 ...

... The main area of ​​infection is the eastern coast of the United States (especially to
Philadelphia), from China (from 115 g.v.d.) to Japan, the Russian Federation - Northeast Siberia and the Far
East.
In addition to all this, I will focus on the question of interpreting words about date ranges. There are 2 interpretations:

  1. an event occurs the entire range of dates (a constant event, stretched over time);
  2. the event will occur once or several times during the specified period of time (event / events are not extended in time).

Having mastered the book S.A. Vronsky's "Astrology: science or superstition" I found the answer to this question and in the future I rely on the 2nd option (event / events are not extended in time).

And a little about the number of bursts: I perceive this information as the minimum number of bursts, and not the exact maximum number. From this point of view, I will continue to evaluate the quality of the forecast.

Now we are ready to try, without emotion, to evaluate the accuracy of the forecast. The data for the article was mainly taken on April 4 and 5, 2020 .

Let's start with the data from Hopkins University . According to these data, the east coast of the United States is now really much more red (more infections) than the west:



We’ll look separately at the area in the Philadelphia area (near New York):



In my opinion, the picture shows that if we talk about the area from Philadelphia to Boston - there is really more red than from Philadelphia to Miami or to Chicago. Unfortunately, the forecast did not indicate where to still from Philadelphia.

When I spoke with people about this forecast at the beginning of March, many pointed out the inaccuracy of the prevalence of infection by geography: then everyone on the lips besides China had Italy, which quickly broke into the leaders ( March 19 was in the first place ) and Iran ( second place in February 29, 2020, and according to Hopkins University on April 6, Iran fell to 7th place.) About the United States then did not speak at all. By the way, on April 6, according to the University of Hopkins, the USA takes the leading line in the number of confirmed cases of infection.


But it was not always so. The United States came to first place on March 29th,

and on March 22nd they were in 3rd.

Thus, over time, the forecast for the USA began to be justified. And now is not the end of the year - it’s too early to take stock of the world.

In the part "RF - North-East of Siberia and the Far East" - has not yet been confirmed. But, perhaps, everything is still ahead (as was the case with the USA).

In terms of the forecast "... from China (from 115 g.v.d.) to Japan ...". Let's see what is located on 115 gr.vd ... This is a whole line, where a lot of things fall. But, in particular, China. And the very epicenter of the spread of the epidemic.

If someone suddenly doubts that the coronavirus started in China - information from the WHO website
... in connection with the current outbreak of coronavirus infection COVID-19, which was first reported on December 31, 2019 in Wuhan, China
In the following picture, the red dot marks the position 30.483472 server latitude, 115.001202 east longitude. Wuhan is approximately 72 km away.



According to the website of Hopkins University (unfortunately, the data for China is mainly by province, not city):

  • China - 82 602 (the main place is Hubei Province - 67 803)
  • South Korea - 10,237
  • Japan - 3,139
  • Philippines - 3,246
  • Indonesia - 2,273 (although part of Indonesia falls into the southern hemisphere, and the server-based forecast).

All together - almost 100,000. Almost the same will be if from 115 g.v.d. go the other way (west) to Turkey.

What clearly does not fall into the forecast is the situation in Europe (countries from the top for infection):

  • Spain - 130,759
  • Italy - 124,632
  • Germany - 96 108
  • France - 89 953
  • Britain - 41,903
  • Switzerland - 21,100
  • Belgium - 19,691
  • Netherlands - 16 627
  • Austria - 11,861
  • Portugal - 11,278

According to rough approximate estimates, it turns out that from China (from 115 gr.v.d.) to Japan + the USA (entirely, without separating the eastern part from the western one) - about 412,000. The

top countries of Europe + countries to the east - about 625,000.

A brief summary of the geography of distribution for April 5 : the forecast is not 100% accurate. But also much more than 0%. If there are qualified commentators among them and will be able to calculate the accuracy of the geographic forecast more accurately, and not as crudely as mine, I will be glad to replenish the article with these data.

About the number of bursts and its dates


I recall from the forecast:
At the end of 2019, the arrival of influenza should be expected in late October and until the second of December.

The second outbreak of the epidemic - from January 4, 2020 to February 16, 2020


The site of the University of Hopkins has a schedule of new growths in patients (starting January 22). It is clearly seen that outbreaks of new infections were recorded on February 13-14 (15 100 and 6 500 respectively). And starting from February 19 - a strong decline (503) and almost until March (a new serious surge in March 13 - 16 900).



If we look at the logarithmic scale, the surge is even better visible: the interval from the very beginning (January 22) to February 13.



I was not too lazy to attach a protractor: from January 23 to 27, the steepness is approximately 45 degrees. From January 27 to January 28 - almost 60 degrees.

From February 12 to 13 - about 40 degrees. After these dates, the graph, although it has an increase, but the angle of inclination as a whole does not exceed 30 degrees.

Personally, I remember from the news from December 2019 - January 2020 that the surge was first in China. And in January (when the news about the downed Boeing in Iran ended) - they began to trumpet with might and main about cases of export of the virus to Europe, Iran and other countries. And in late December, mass media reported mass cases. Those. the disease itself was earlier than the end of December. As it turned out later from my communication with other people - not everyone remembers and considers the same way as I do. For example, there were people who believe that everything began in January. And in December there were no diseases, even in China. What do these people were sure of, and not speculated. In my opinion, this is a matter of psychology. I remember that in Alexey Berezhkov’s book “My Mute Afghan” this moment of psychology was also noticed: when colleagues, remembering the service in Afghanistan, after 20 years they assurethat they remember everything in detail - as if it were yesterday. Although, when comparing the stories, it turns out that the chronology of some is different. And the details contradict each other. Although in general, as a first approximation, the event by description is the same for everyone.

So you have to look for evidence. Unfortunately, the map of China on the website of Hopkins University does not answer the question of what was there in October-December 2019 (the schedule runs from January 2020). I have to delve into the news.

According to this source :
On December 31, 2019, the Chinese authorities informed the World Health Organization about an outbreak of unknown pneumonia in the city of Wuhan in the central part of the country (Hubei Province). Specialists found that the causative agent of the disease was a new coronavirus, later the disease was officially named COVID-19. WHO announces pandemic of new coronavirus in the world on March 11
Already from this brief information, it can be assumed that there were at least 2 outbreaks: in late December in China. From March 11 - around the world.

And here is the news that on December 30, a Chinese doctor spoke about this disease.

It reports the beginning of December.

According to this source :
On November 17, in Hubei Province, coronavirus was diagnosed in a 55-year-old man. In total, nine cases of the disease were detected in November.
This news has also been reprinted in various media. Of course, it should be borne in mind that not all news from the Internet (even from the media) can be fully trusted.

As a result, it is difficult to say exactly when this disease began. We can only say that at the end of December it already diverged greatly in parts of China. Given the incubation period (approximately 14 days), we get the middle of December. And taking into account the latency between the event itself and information about it ( Wenliang wrote on WeChat on December 30, that means the topic for this message appeared earlier than December 30) - the first half of December at least.

Unfortunately, in this forecast, everything is very brief and it is not written when the peak in certain areas will be passed. After searching the Internet a little more, I found updated information from the same author dated January 26 (post dated February 2, citing a source dated January 26):
it can be predicted that the main severity of the course of the disease and the spread of infections, when there can be an increase in mortality and complications from infections, will be from February 4 to February 16, 2020.

There is one more point, though I’m not very sure about it ... ... Theoretically, it is in these areas (from 106 grams of airborne forces to 123 grams) that increased mortality and complications can be expected.
This location, among other things, includes China. Again access the data Hopkins University and see:



Ie The sharp jump was from February 12 to 14. True, in addition to China, other countries (for example, Singapore, Thailand, Vietnam) fall in the northern hemisphere. But, due to the smaller number of people there, we can assume that China sets the basic statistics.

From here I will give a clipping of the forecast for the near future:
The greatest likelihood of the occurrence and spread of infections during these periods can be expected until June 21, 2020 in settlements located from 79 g. VD to 115 g. VD, and after June 21 - from 32 gr. VD to 64 gr. VD.
It is worth considering that this is not specifically about the coronavirus. And just about infections.

Brief Summary: Information on date ranges and bursts is confirmed.

When was the forecast made


For those who doubt that the forecast was made in advance, the Internet will help. Google on the phrase "Study of the influence of the planets of the solar system on the emergence and spread of epidemics." Through Yandex and Google found:

  1. May 2013 medical journal (unfortunately, the status of the journal is unclear to me: these are medical lovers or professionals with specialized education)
  2. May, 2013
  3. May, 2013
  4. June 2013
  5. October 2015
  6. May, 2013

Of course, it can be assumed that the author of the forecast was so powerful that he was able to access all of these resources (and I brought only a part) and modify them, "unwinding" the time on these resources. But, I strongly doubt something.

What other predictions did the author make, how accurate are they?


In the same article, forecasts are scheduled from 2013 end to 2050. I will not be able to give an answer about their quality (since this is a very painstaking work, which takes a lot of time and deserves separate articles). But I can highlight the problems that you may encounter when trying to answer this question.

Trying to figure out the forecast for 2016, I came across an interesting situation. Question: Was there an outbreak of fever in India in 2016? The answer depends on which data source is available and how to interpret it.

For example, on Wikipedia (the Russian version , the English version ) there is not a word about the 2016 outbreak in India.

If you refer to the data on page 67 from the book "Chikungunya and Zika: Global Status", then the 2016 epidemic is not visible at all.



To be careful, there is an answer why: against the backdrop of a very large epidemic in 2006, 2016 generally seems to be something insignificant.

If you turn to the data on the website of Rospotrebnadzor - there is information about the epidemic .

The outbreak of the epidemic in 2016 (relative to 2011-2015) can be seen in this article.



Summary: check information from different sources and be careful. Otherwise, you can easily make the wrong conclusion. But, in general, the analysis of epidemic stories in various areas of the globe is not an easy task and takes a lot of time.

Is it possible to make a similar forecast by accident? What is the probability of making a similar forecast by accident?


Perhaps this is the most interesting question. Unfortunately, in this article he will remain unanswered: of course, an accident could also be. But, I am not competent in such calculations. I hope that among the commentators to this article there are more competent people in this topic. And they will be able to accurately calculate this probability.

Total


At the moment:

  • forecast made in 2013 - confirmed;
  • information on ranges of dates and bursts was confirmed;
  • information on the geography of distribution is not 100% accurate, but also far from 0% (I would like to find a better assessment from competent people);
  • the probability of accidentally making such a forecast is a question requiring an answer from competent specialists.

Given that the forecast was made 7 years before the event, it is very brief and, as I assume, free of charge - a very good price / quality ratio. If suddenly someone did not impress with the accuracy of the forecast and he would like more ... well - any whim at your expense. Those who wish can easily find the author’s contact information on the Internet and try to agree on a more detailed forecast for this event for a fee.

More articles:


All Articles